البحث

Smithsonian Magazine شارك رابطًا

2025-05-19 22:18:40 ·

NASA Engineers Revive Long-Dormant Thrusters on Voyager 1, the Farthest Spacecraft From Earth, in a 'Miracle Save'

NASA Engineers Revive Long-Dormant Thrusters on Voyager 1, the Farthest Spacecraft From Earth, in a ‘Miracle ’
In the nick of time, NASA teams addressed clogging issues in the probe’s backup roll thrusters, before the only antenna capable of sending commands to it went offline

An artistic rendering of one of the Voyager probes.
NASA/JPL-Caltech

In a nail-biting mission to secure Voyager 1 before the only antenna that can send commands to the spacecraft goes offline for upgrades, NASA engineers revived thrusters aboard the probe that have been considered dead for more than two decades.
“It was such a glorious moment,” Todd Barber, the mission’s propulsion lead at NASA’s Jet Propulsion Laboratory, says in a statement. “Team morale was very high that day.”
Voyager 1 and Voyager 2 launched in 1977 to study Jupiter and Saturn as well as examine some of the planets’ moons. Since then, however, their mission has been repeatedly extended. The pair of probes also investigated Uranus and Neptune, 48 moons in the solar system and achieved what no human-made object had ever done before: They entered interstellar space. Voyager 1 first left our solar system in 2012, and Voyager 2 followed in 2018. Needless to say, the probes have transcended their mission, which was originally intended to last just five years, per Mashable’s Elisha Sauers.
The aging Voyagers have hit several hiccups in recent years, and the latest involved Voyager 1’s roll thrusters. Among various propulsion devices, the roll thrusters rotate the probes, keeping the antennas facing toward Earth for communication and the spacecraft pointed at a guide star, which is used for orientation. Each has a primary and backup set of roll thrusters, which NASA engineers toggle between using to avoid clogging their fuel tubes from overuse.
In 2004, however, Voyager 1’s primary roll thrusters stopped working after two internal heaters lost power, according to NASA’s statement. The team concluded that the heaters were as good as dead and continued the mission with only the backup roll thrusters.
“I think at that time, the team was OK with accepting that the primary roll thrusters didn’t work, because they had a perfectly good backup,” Kareem Badaruddin, Voyager mission manager at JPL, explains in the statement. “And frankly, they probably didn’t think the Voyagers were going to keep going for another 20 years.” After all, “it’s amazing that the two spacecraft are still working,” as former Voyager project scientist Ed Stone toldSpace.com’s Mike Wall in 2017, four decades after the mission’s launch.Now, the backup roll thrusters’ fuel tubes have accumulated enough buildup residue to likely cause serious problems later this year, so the team decided to re-evaluate the situation. Their work was even more time-sensitive given that Australia’s Deep Space Station 43—the only antenna in NASA’s Deep Space Network powerful enough to send commands to Voyager 1 and 2—was about to go offline. Planned upgrades between May 4, 2025, and February of next year are making it unusable, with only brief windows of operation in August and December.
Without the antenna, the team can’t send instructions to either probe. And if the backup roll thruster were to become totally clogged while the antenna is offline, it might be impossible for engineers to fix the issue.
The team suggested the thrusters’ heaters had stopped working because their circuits had unexpectedly flipped a switch back in 2004. If engineers could get the probe to flip the switch back, that might fix the problem.
Testing out this solution, however, was risky, because it first required turning the primary thrusters on, then flipping the switch and restarting the heaters. With the long-dormant thrusters operating, they were at risk of automatically firing if Voyager 1 moved too far from its guiding star. If this happened while the heaters were still off, it could result in an explosion. In other words, the engineers needed to keep the star tracker pointed as closely as possible to the guiding star while working.
As if that wasn’t dramatic enough, it currently takes around 23 hours for scientists’ command to reach Voyager 1, and then another 23 hours for the probe’s response to return to Earth. The team thus had to wait around two days to see if their approach had worked. If it hadn’t, they wouldn’t receive that message until 23 hours after their command might have already caused serious damage.
Fortunately, after the team sent their command in March, Voyager 1 reported a strong rise in the thruster heaters’ temperatures—a sign of success.
“These thrusters were considered dead. And that was a legitimate conclusion. It’s just that one of our engineers had this insight that maybe there was this other possible cause, and it was fixable,” Barber says in the statement. “It was yet another miracle save for Voyager.”
This isn’t the first time a team has had to scramble to keep a Voyager spacecraft alive. Given the mission’s unexpected longevity, NASA scientists and engineers must continuously mitigate age-related maintenance problems on the probes, per Space.com’s Samantha Mathewson.
Last fall, for instance, engineers briefly lost touch with Voyager 1 when it turned off its primary radio transmitter to save power. Instead, the probe reconnected with NASA using a backup transmitter that hadn’t been used in more than four decades. Voyager 1 also spent months sending incomprehensible data back to Earth, until NASA teams figured out a fix. Voyager 2 is not immune from these issues—it also dropped communication with NASA for some time in 2023.
Teams have begun shutting off certain instruments on the probes to preserve power. By doing so, NASA suggests the two spacecraft will be able to operate with at least one instrument into the next decade. Even once they lose power and the ability to communicate with scientists back on Earth, however, the probes will continue to silently travel away from our planet.
“In the absence of sun or wind or anything that’s going to wear them down, they could easily outlast us—our entire civilization, outlast our planet,” Jim Bell, an Arizona State University planetary scientist and author of the book, The Interstellar Age: Inside the Forty-Year Voyager Mission, told Scientific American’s Clara Moskowitz in 2015. “The Earth will eventually be swallowed by the sun and the Voyagers could still be out there.”

Get the latest stories in your inbox every weekday.
#nasa #engineers #revive #longdormant #thrusters

NASA Engineers Revive Long-Dormant Thrusters on Voyager 1, the Farthest Spacecraft From Earth, in a 'Miracle Save'
NASA Engineers Revive Long-Dormant Thrusters on Voyager 1, the Farthest Spacecraft From Earth, in a ‘Miracle ’ In the nick of time, NASA teams addressed clogging issues in the probe’s backup roll thrusters, before the only antenna capable of sending commands to it went offline An artistic rendering of one of the Voyager probes. NASA/JPL-Caltech In a nail-biting mission to secure Voyager 1 before the only antenna that can send commands to the spacecraft goes offline for upgrades, NASA engineers revived thrusters aboard the probe that have been considered dead for more than two decades. “It was such a glorious moment,” Todd Barber, the mission’s propulsion lead at NASA’s Jet Propulsion Laboratory, says in a statement. “Team morale was very high that day.” Voyager 1 and Voyager 2 launched in 1977 to study Jupiter and Saturn as well as examine some of the planets’ moons. Since then, however, their mission has been repeatedly extended. The pair of probes also investigated Uranus and Neptune, 48 moons in the solar system and achieved what no human-made object had ever done before: They entered interstellar space. Voyager 1 first left our solar system in 2012, and Voyager 2 followed in 2018. Needless to say, the probes have transcended their mission, which was originally intended to last just five years, per Mashable’s Elisha Sauers. The aging Voyagers have hit several hiccups in recent years, and the latest involved Voyager 1’s roll thrusters. Among various propulsion devices, the roll thrusters rotate the probes, keeping the antennas facing toward Earth for communication and the spacecraft pointed at a guide star, which is used for orientation. Each has a primary and backup set of roll thrusters, which NASA engineers toggle between using to avoid clogging their fuel tubes from overuse. In 2004, however, Voyager 1’s primary roll thrusters stopped working after two internal heaters lost power, according to NASA’s statement. The team concluded that the heaters were as good as dead and continued the mission with only the backup roll thrusters. “I think at that time, the team was OK with accepting that the primary roll thrusters didn’t work, because they had a perfectly good backup,” Kareem Badaruddin, Voyager mission manager at JPL, explains in the statement. “And frankly, they probably didn’t think the Voyagers were going to keep going for another 20 years.” After all, “it’s amazing that the two spacecraft are still working,” as former Voyager project scientist Ed Stone toldSpace.com’s Mike Wall in 2017, four decades after the mission’s launch.Now, the backup roll thrusters’ fuel tubes have accumulated enough buildup residue to likely cause serious problems later this year, so the team decided to re-evaluate the situation. Their work was even more time-sensitive given that Australia’s Deep Space Station 43—the only antenna in NASA’s Deep Space Network powerful enough to send commands to Voyager 1 and 2—was about to go offline. Planned upgrades between May 4, 2025, and February of next year are making it unusable, with only brief windows of operation in August and December. Without the antenna, the team can’t send instructions to either probe. And if the backup roll thruster were to become totally clogged while the antenna is offline, it might be impossible for engineers to fix the issue. The team suggested the thrusters’ heaters had stopped working because their circuits had unexpectedly flipped a switch back in 2004. If engineers could get the probe to flip the switch back, that might fix the problem. Testing out this solution, however, was risky, because it first required turning the primary thrusters on, then flipping the switch and restarting the heaters. With the long-dormant thrusters operating, they were at risk of automatically firing if Voyager 1 moved too far from its guiding star. If this happened while the heaters were still off, it could result in an explosion. In other words, the engineers needed to keep the star tracker pointed as closely as possible to the guiding star while working. As if that wasn’t dramatic enough, it currently takes around 23 hours for scientists’ command to reach Voyager 1, and then another 23 hours for the probe’s response to return to Earth. The team thus had to wait around two days to see if their approach had worked. If it hadn’t, they wouldn’t receive that message until 23 hours after their command might have already caused serious damage. Fortunately, after the team sent their command in March, Voyager 1 reported a strong rise in the thruster heaters’ temperatures—a sign of success. “These thrusters were considered dead. And that was a legitimate conclusion. It’s just that one of our engineers had this insight that maybe there was this other possible cause, and it was fixable,” Barber says in the statement. “It was yet another miracle save for Voyager.” This isn’t the first time a team has had to scramble to keep a Voyager spacecraft alive. Given the mission’s unexpected longevity, NASA scientists and engineers must continuously mitigate age-related maintenance problems on the probes, per Space.com’s Samantha Mathewson. Last fall, for instance, engineers briefly lost touch with Voyager 1 when it turned off its primary radio transmitter to save power. Instead, the probe reconnected with NASA using a backup transmitter that hadn’t been used in more than four decades. Voyager 1 also spent months sending incomprehensible data back to Earth, until NASA teams figured out a fix. Voyager 2 is not immune from these issues—it also dropped communication with NASA for some time in 2023. Teams have begun shutting off certain instruments on the probes to preserve power. By doing so, NASA suggests the two spacecraft will be able to operate with at least one instrument into the next decade. Even once they lose power and the ability to communicate with scientists back on Earth, however, the probes will continue to silently travel away from our planet. “In the absence of sun or wind or anything that’s going to wear them down, they could easily outlast us—our entire civilization, outlast our planet,” Jim Bell, an Arizona State University planetary scientist and author of the book, The Interstellar Age: Inside the Forty-Year Voyager Mission, told Scientific American’s Clara Moskowitz in 2015. “The Earth will eventually be swallowed by the sun and the Voyagers could still be out there.” Get the latest stories in your inbox every weekday. #nasa #engineers #revive #longdormant #thrusters

WWW.SMITHSONIANMAG.COM

NASA Engineers Revive Long-Dormant Thrusters on Voyager 1, the Farthest Spacecraft From Earth, in a 'Miracle Save'

NASA Engineers Revive Long-Dormant Thrusters on Voyager 1, the Farthest Spacecraft From Earth, in a ‘Miracle Save’ In the nick of time, NASA teams addressed clogging issues in the probe’s backup roll thrusters, before the only antenna capable of sending commands to it went offline An artistic rendering of one of the Voyager probes. NASA/JPL-Caltech In a nail-biting mission to secure Voyager 1 before the only antenna that can send commands to the spacecraft goes offline for upgrades, NASA engineers revived thrusters aboard the probe that have been considered dead for more than two decades. “It was such a glorious moment,” Todd Barber, the mission’s propulsion lead at NASA’s Jet Propulsion Laboratory (JPL), says in a statement. “Team morale was very high that day.” Voyager 1 and Voyager 2 launched in 1977 to study Jupiter and Saturn as well as examine some of the planets’ moons. Since then, however, their mission has been repeatedly extended. The pair of probes also investigated Uranus and Neptune, 48 moons in the solar system and achieved what no human-made object had ever done before: They entered interstellar space. Voyager 1 first left our solar system in 2012, and Voyager 2 followed in 2018. Needless to say, the probes have transcended their mission, which was originally intended to last just five years, per Mashable’s Elisha Sauers. The aging Voyagers have hit several hiccups in recent years, and the latest involved Voyager 1’s roll thrusters. Among various propulsion devices, the roll thrusters rotate the probes, keeping the antennas facing toward Earth for communication and the spacecraft pointed at a guide star, which is used for orientation. Each has a primary and backup set of roll thrusters, which NASA engineers toggle between using to avoid clogging their fuel tubes from overuse. In 2004, however, Voyager 1’s primary roll thrusters stopped working after two internal heaters lost power, according to NASA’s statement. The team concluded that the heaters were as good as dead and continued the mission with only the backup roll thrusters. “I think at that time, the team was OK with accepting that the primary roll thrusters didn’t work, because they had a perfectly good backup,” Kareem Badaruddin, Voyager mission manager at JPL, explains in the statement. “And frankly, they probably didn’t think the Voyagers were going to keep going for another 20 years.” After all, “it’s amazing that the two spacecraft are still working,” as former Voyager project scientist Ed Stone toldSpace.com’s Mike Wall in 2017, four decades after the mission’s launch.Now, the backup roll thrusters’ fuel tubes have accumulated enough buildup residue to likely cause serious problems later this year, so the team decided to re-evaluate the situation. Their work was even more time-sensitive given that Australia’s Deep Space Station 43—the only antenna in NASA’s Deep Space Network powerful enough to send commands to Voyager 1 and 2—was about to go offline. Planned upgrades between May 4, 2025, and February of next year are making it unusable, with only brief windows of operation in August and December. Without the antenna, the team can’t send instructions to either probe. And if the backup roll thruster were to become totally clogged while the antenna is offline, it might be impossible for engineers to fix the issue. The team suggested the thrusters’ heaters had stopped working because their circuits had unexpectedly flipped a switch back in 2004. If engineers could get the probe to flip the switch back, that might fix the problem. Testing out this solution, however, was risky, because it first required turning the primary thrusters on, then flipping the switch and restarting the heaters. With the long-dormant thrusters operating, they were at risk of automatically firing if Voyager 1 moved too far from its guiding star. If this happened while the heaters were still off, it could result in an explosion. In other words, the engineers needed to keep the star tracker pointed as closely as possible to the guiding star while working. As if that wasn’t dramatic enough, it currently takes around 23 hours for scientists’ command to reach Voyager 1, and then another 23 hours for the probe’s response to return to Earth. The team thus had to wait around two days to see if their approach had worked. If it hadn’t, they wouldn’t receive that message until 23 hours after their command might have already caused serious damage. Fortunately, after the team sent their command in March, Voyager 1 reported a strong rise in the thruster heaters’ temperatures—a sign of success. “These thrusters were considered dead. And that was a legitimate conclusion. It’s just that one of our engineers had this insight that maybe there was this other possible cause, and it was fixable,” Barber says in the statement. “It was yet another miracle save for Voyager.” This isn’t the first time a team has had to scramble to keep a Voyager spacecraft alive. Given the mission’s unexpected longevity, NASA scientists and engineers must continuously mitigate age-related maintenance problems on the probes, per Space.com’s Samantha Mathewson. Last fall, for instance, engineers briefly lost touch with Voyager 1 when it turned off its primary radio transmitter to save power. Instead, the probe reconnected with NASA using a backup transmitter that hadn’t been used in more than four decades. Voyager 1 also spent months sending incomprehensible data back to Earth, until NASA teams figured out a fix. Voyager 2 is not immune from these issues—it also dropped communication with NASA for some time in 2023. Teams have begun shutting off certain instruments on the probes to preserve power. By doing so, NASA suggests the two spacecraft will be able to operate with at least one instrument into the next decade. Even once they lose power and the ability to communicate with scientists back on Earth, however, the probes will continue to silently travel away from our planet. “In the absence of sun or wind or anything that’s going to wear them down, they could easily outlast us—our entire civilization, outlast our planet,” Jim Bell, an Arizona State University planetary scientist and author of the book, The Interstellar Age: Inside the Forty-Year Voyager Mission, told Scientific American’s Clara Moskowitz in 2015. “The Earth will eventually be swallowed by the sun and the Voyagers could still be out there.” Get the latest stories in your inbox every weekday.

·5 مشاهدة

الرجاء تسجيل الدخول , للأعجاب والمشاركة والتعليق على هذا!
NVIDIA شارك رابطًا

2025-05-19 21:45:19 ·

NVIDIA and Microsoft Accelerate Agentic AI Innovation, From Cloud to PC

Agentic AI is redefining scientific discovery and unlocking research breakthroughs and innovations across industries. Through deepened collaboration, NVIDIA and Microsoft are delivering advancements that accelerate agentic AI-powered applications from the cloud to the PC.
At Microsoft Build, Microsoft unveiled Microsoft Discovery, an extensible platform built to empower researchers to transform the entire discovery process with agentic AI. This will help research and development departments across various industries accelerate the time to market for new products, as well as speed and expand the end-to-end discovery process for all scientists.
Microsoft Discovery will integrate the NVIDIA ALCHEMI NIM microservice, which optimizes AI inference for chemical simulations, to accelerate materials science research with property prediction and candidate recommendation. The platform will also integrate NVIDIA BioNeMo NIM microservices, tapping into pretrained AI workflows to speed up AI model development for drug discovery. These integrations equip researchers with accelerated performance for faster scientific discoveries.
In testing, researchers at Microsoft used Microsoft Discovery to detect a novel coolant prototype with promising properties for immersion cooling in data centers in under 200 hours, rather than months or years with traditional methods.
Advancing Agentic AI With NVIDIA GB200 Deployments at Scale
Microsoft is rapidly deploying tens of thousands of NVIDIA GB200 NVL72 rack-scale systems across its Azure data centers, boosting both performance and efficiency.
Azure’s ND GB200 v6 virtual machines — built on a rack-scale architecture with up to 72 NVIDIA Blackwell GPUs per rack and advanced liquid cooling — deliver up to 35x more inference throughput compared with previous ND H100 v5 VMs accelerated by eight NVIDIA H100 GPUs, setting a new benchmark for AI workloads.
These innovations are underpinned by custom server designs, high-speed NVIDIA NVLink interconnects and NVIDIA Quantum InfiniBand networking — enabling seamless scaling to tens of thousands of Blackwell GPUs for demanding generative and agentic AI applications.
Microsoft chairman and CEO Satya Nadella and NVIDIA founder and CEO Jensen Huang also highlighted how Microsoft and NVIDIA’s collaboration is compounding performance gains through continuous software optimizations across NVIDIA architectures on Azure. This approach maximizes developer productivity, lowers total cost of ownership and accelerates all workloads, including AI and data processing — all while driving greater efficiency per dollar and per watt for customers.
NVIDIA AI Reasoning and Healthcare Microservices on Azure AI Foundry
Building on the NIM integration in Azure AI Foundry, announced at NVIDIA GTC, Microsoft and NVIDIA are expanding the platform with the NVIDIA Llama Nemotron family of open reasoning models and NVIDIA BioNeMo NIM microservices, which deliver enterprise-grade, containerized inferencing for complex decision-making and domain-specific AI workloads.
Developers can now access optimized NIM microservices for advanced reasoning in Azure AI Foundry. These include the NVIDIA Llama Nemotron Super and Nano models, which offer advanced multistep reasoning, coding and agentic capabilities, delivering up to 20% higher accuracy and 5x faster inference than previous models.
Healthcare-focused BioNeMo NIM microservices like ProteinMPNN, RFDiffusion and OpenFold2 address critical applications in digital biology, drug discovery and medical imaging, enabling researchers and clinicians to accelerate protein science, molecular modeling and genomic analysis for improved patient care and faster scientific innovation.
This expanded integration empowers organizations to rapidly deploy high-performance AI agents, connecting to these models and other specialized healthcare solutions with robust reliability and simplified scaling.
Accelerating Generative AI on Windows 11 With RTX AI PCs
Generative AI is reshaping PC software with entirely new experiences — from digital humans to writing assistants, intelligent agents and creative tools. NVIDIA RTX AI PCs make it easy to get it started with experimenting with generative AI and unlock greater performance on Windows 11.
At Microsoft Build, NVIDIA and Microsoft are unveiling an AI inferencing stack to simplify development and boost inference performance for Windows 11 PCs.
NVIDIA TensorRT has been reimagined for RTX AI PCs, combining industry-leading TensorRT performance with just-in-time, on-device engine building and an 8x smaller package size for seamless AI deployment to the more than 100 million RTX AI PCs.
Announced at Microsoft Build, TensorRT for RTX is natively supported by Windows ML — a new inference stack that provides app developers with both broad hardware compatibility and state-of-the-art performance. TensorRT for RTX is available in the Windows ML preview starting today, and will be available as a standalone software development kit from NVIDIA Developer in June.
Learn more about how TensorRT for RTX and Windows ML are streamlining software development. Explore new NIM microservices and AI Blueprints for RTX, and RTX-powered updates from Autodesk, Bilibili, Chaos, LM Studio and Topaz in the RTX AI PC blog, and join the community discussion on Discord.
Explore sessions, hands-on workshops and live demos at Microsoft Build to learn how Microsoft and NVIDIA are accelerating agentic AI.
#nvidia #microsoft #accelerate #agentic #innovation

NVIDIA and Microsoft Accelerate Agentic AI Innovation, From Cloud to PC
Agentic AI is redefining scientific discovery and unlocking research breakthroughs and innovations across industries. Through deepened collaboration, NVIDIA and Microsoft are delivering advancements that accelerate agentic AI-powered applications from the cloud to the PC. At Microsoft Build, Microsoft unveiled Microsoft Discovery, an extensible platform built to empower researchers to transform the entire discovery process with agentic AI. This will help research and development departments across various industries accelerate the time to market for new products, as well as speed and expand the end-to-end discovery process for all scientists. Microsoft Discovery will integrate the NVIDIA ALCHEMI NIM microservice, which optimizes AI inference for chemical simulations, to accelerate materials science research with property prediction and candidate recommendation. The platform will also integrate NVIDIA BioNeMo NIM microservices, tapping into pretrained AI workflows to speed up AI model development for drug discovery. These integrations equip researchers with accelerated performance for faster scientific discoveries. In testing, researchers at Microsoft used Microsoft Discovery to detect a novel coolant prototype with promising properties for immersion cooling in data centers in under 200 hours, rather than months or years with traditional methods. Advancing Agentic AI With NVIDIA GB200 Deployments at Scale Microsoft is rapidly deploying tens of thousands of NVIDIA GB200 NVL72 rack-scale systems across its Azure data centers, boosting both performance and efficiency. Azure’s ND GB200 v6 virtual machines — built on a rack-scale architecture with up to 72 NVIDIA Blackwell GPUs per rack and advanced liquid cooling — deliver up to 35x more inference throughput compared with previous ND H100 v5 VMs accelerated by eight NVIDIA H100 GPUs, setting a new benchmark for AI workloads. These innovations are underpinned by custom server designs, high-speed NVIDIA NVLink interconnects and NVIDIA Quantum InfiniBand networking — enabling seamless scaling to tens of thousands of Blackwell GPUs for demanding generative and agentic AI applications. Microsoft chairman and CEO Satya Nadella and NVIDIA founder and CEO Jensen Huang also highlighted how Microsoft and NVIDIA’s collaboration is compounding performance gains through continuous software optimizations across NVIDIA architectures on Azure. This approach maximizes developer productivity, lowers total cost of ownership and accelerates all workloads, including AI and data processing — all while driving greater efficiency per dollar and per watt for customers. NVIDIA AI Reasoning and Healthcare Microservices on Azure AI Foundry Building on the NIM integration in Azure AI Foundry, announced at NVIDIA GTC, Microsoft and NVIDIA are expanding the platform with the NVIDIA Llama Nemotron family of open reasoning models and NVIDIA BioNeMo NIM microservices, which deliver enterprise-grade, containerized inferencing for complex decision-making and domain-specific AI workloads. Developers can now access optimized NIM microservices for advanced reasoning in Azure AI Foundry. These include the NVIDIA Llama Nemotron Super and Nano models, which offer advanced multistep reasoning, coding and agentic capabilities, delivering up to 20% higher accuracy and 5x faster inference than previous models. Healthcare-focused BioNeMo NIM microservices like ProteinMPNN, RFDiffusion and OpenFold2 address critical applications in digital biology, drug discovery and medical imaging, enabling researchers and clinicians to accelerate protein science, molecular modeling and genomic analysis for improved patient care and faster scientific innovation. This expanded integration empowers organizations to rapidly deploy high-performance AI agents, connecting to these models and other specialized healthcare solutions with robust reliability and simplified scaling. Accelerating Generative AI on Windows 11 With RTX AI PCs Generative AI is reshaping PC software with entirely new experiences — from digital humans to writing assistants, intelligent agents and creative tools. NVIDIA RTX AI PCs make it easy to get it started with experimenting with generative AI and unlock greater performance on Windows 11. At Microsoft Build, NVIDIA and Microsoft are unveiling an AI inferencing stack to simplify development and boost inference performance for Windows 11 PCs. NVIDIA TensorRT has been reimagined for RTX AI PCs, combining industry-leading TensorRT performance with just-in-time, on-device engine building and an 8x smaller package size for seamless AI deployment to the more than 100 million RTX AI PCs. Announced at Microsoft Build, TensorRT for RTX is natively supported by Windows ML — a new inference stack that provides app developers with both broad hardware compatibility and state-of-the-art performance. TensorRT for RTX is available in the Windows ML preview starting today, and will be available as a standalone software development kit from NVIDIA Developer in June. Learn more about how TensorRT for RTX and Windows ML are streamlining software development. Explore new NIM microservices and AI Blueprints for RTX, and RTX-powered updates from Autodesk, Bilibili, Chaos, LM Studio and Topaz in the RTX AI PC blog, and join the community discussion on Discord. Explore sessions, hands-on workshops and live demos at Microsoft Build to learn how Microsoft and NVIDIA are accelerating agentic AI. #nvidia #microsoft #accelerate #agentic #innovation

BLOGS.NVIDIA.COM

NVIDIA and Microsoft Accelerate Agentic AI Innovation, From Cloud to PC

Agentic AI is redefining scientific discovery and unlocking research breakthroughs and innovations across industries. Through deepened collaboration, NVIDIA and Microsoft are delivering advancements that accelerate agentic AI-powered applications from the cloud to the PC. At Microsoft Build, Microsoft unveiled Microsoft Discovery, an extensible platform built to empower researchers to transform the entire discovery process with agentic AI. This will help research and development departments across various industries accelerate the time to market for new products, as well as speed and expand the end-to-end discovery process for all scientists. Microsoft Discovery will integrate the NVIDIA ALCHEMI NIM microservice, which optimizes AI inference for chemical simulations, to accelerate materials science research with property prediction and candidate recommendation. The platform will also integrate NVIDIA BioNeMo NIM microservices, tapping into pretrained AI workflows to speed up AI model development for drug discovery. These integrations equip researchers with accelerated performance for faster scientific discoveries. In testing, researchers at Microsoft used Microsoft Discovery to detect a novel coolant prototype with promising properties for immersion cooling in data centers in under 200 hours, rather than months or years with traditional methods. Advancing Agentic AI With NVIDIA GB200 Deployments at Scale Microsoft is rapidly deploying tens of thousands of NVIDIA GB200 NVL72 rack-scale systems across its Azure data centers, boosting both performance and efficiency. Azure’s ND GB200 v6 virtual machines — built on a rack-scale architecture with up to 72 NVIDIA Blackwell GPUs per rack and advanced liquid cooling — deliver up to 35x more inference throughput compared with previous ND H100 v5 VMs accelerated by eight NVIDIA H100 GPUs, setting a new benchmark for AI workloads. These innovations are underpinned by custom server designs, high-speed NVIDIA NVLink interconnects and NVIDIA Quantum InfiniBand networking — enabling seamless scaling to tens of thousands of Blackwell GPUs for demanding generative and agentic AI applications. Microsoft chairman and CEO Satya Nadella and NVIDIA founder and CEO Jensen Huang also highlighted how Microsoft and NVIDIA’s collaboration is compounding performance gains through continuous software optimizations across NVIDIA architectures on Azure. This approach maximizes developer productivity, lowers total cost of ownership and accelerates all workloads, including AI and data processing — all while driving greater efficiency per dollar and per watt for customers. NVIDIA AI Reasoning and Healthcare Microservices on Azure AI Foundry Building on the NIM integration in Azure AI Foundry, announced at NVIDIA GTC, Microsoft and NVIDIA are expanding the platform with the NVIDIA Llama Nemotron family of open reasoning models and NVIDIA BioNeMo NIM microservices, which deliver enterprise-grade, containerized inferencing for complex decision-making and domain-specific AI workloads. Developers can now access optimized NIM microservices for advanced reasoning in Azure AI Foundry. These include the NVIDIA Llama Nemotron Super and Nano models, which offer advanced multistep reasoning, coding and agentic capabilities, delivering up to 20% higher accuracy and 5x faster inference than previous models. Healthcare-focused BioNeMo NIM microservices like ProteinMPNN, RFDiffusion and OpenFold2 address critical applications in digital biology, drug discovery and medical imaging, enabling researchers and clinicians to accelerate protein science, molecular modeling and genomic analysis for improved patient care and faster scientific innovation. This expanded integration empowers organizations to rapidly deploy high-performance AI agents, connecting to these models and other specialized healthcare solutions with robust reliability and simplified scaling. Accelerating Generative AI on Windows 11 With RTX AI PCs Generative AI is reshaping PC software with entirely new experiences — from digital humans to writing assistants, intelligent agents and creative tools. NVIDIA RTX AI PCs make it easy to get it started with experimenting with generative AI and unlock greater performance on Windows 11. At Microsoft Build, NVIDIA and Microsoft are unveiling an AI inferencing stack to simplify development and boost inference performance for Windows 11 PCs. NVIDIA TensorRT has been reimagined for RTX AI PCs, combining industry-leading TensorRT performance with just-in-time, on-device engine building and an 8x smaller package size for seamless AI deployment to the more than 100 million RTX AI PCs. Announced at Microsoft Build, TensorRT for RTX is natively supported by Windows ML — a new inference stack that provides app developers with both broad hardware compatibility and state-of-the-art performance. TensorRT for RTX is available in the Windows ML preview starting today, and will be available as a standalone software development kit from NVIDIA Developer in June. Learn more about how TensorRT for RTX and Windows ML are streamlining software development. Explore new NIM microservices and AI Blueprints for RTX, and RTX-powered updates from Autodesk, Bilibili, Chaos, LM Studio and Topaz in the RTX AI PC blog, and join the community discussion on Discord. Explore sessions, hands-on workshops and live demos at Microsoft Build to learn how Microsoft and NVIDIA are accelerating agentic AI.

·71 مشاهدة

الرجاء تسجيل الدخول , للأعجاب والمشاركة والتعليق على هذا!
NeowinFeed شارك رابطًا

2025-05-19 20:47:04 ·

Windows 10 emergency update KB5061768 fixes BitLocker boot loops

When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works.

Windows 10 emergency update KB5061768 fixes BitLocker boot loops

Taras Buria

Neowin
@TarasBuria ·

May 19, 2025 15:10 EDT

As promised, Microsoft is rolling out an out-of-band emergency update for Windows 10 to address problems with Intel-based systems booting into BitLocker recovery due to LSASS issues caused by KB5058379. The company recently acknowledged a known issue on devices with Intel Trusted Execution Technologyon 10th-gen and later systems and promised to release a fix as soon as possible.
The company did not take too long to address the issue, and the necessary fix is now available on the Microsoft Update Catalog as KB5061768, with build numbers 19044.5853and 19045.5853.
Here is the changelog with a single fix:Fixed: A known issue on devices with Intel Trusted Execution Technologyenabled on 10th generation or later Intel vPro processors. On these systems, installing the May 13, 2025, Windows security updatemight cause the Local Security Authority Subsystem Serviceprocess to terminate unexpectedly, triggering an Automatic Repair prompting for the BitLocker recovery key to continue.

In the updated post on the Windows Health Dashboard website, Microsoft says that KB5061768 is recommended for those who use affected systems and have not yet deployed KB5058379. It will prevent BitLocker problems from happening. As for systems that already cannot boot due to BitLocker loops, Microsoft suggests the following:

To start Windows, temporarily disable Intel VT for Direct I/Oand Intel Trusted Execution Technologyin your BIOS / UEFI settings. This step will require entering your BitLocker recovery key.
Install update KB5061768 via the Microsoft Update Catalog.
After restarting Windows, return to your BIOS / UEFI settings and re-enable Intel VT for Direct I/Oand TXT.As for known bugs, there is one issue with the Noto font: text can appear blurry or unclear in Chrome, Edge, and other Chromium-based browsers in certain languages like Chinese, Japanese, and Korean when the display scaling is set to 100%. Microsoft is investigating this problem, and users can mitigate it by setting the display scaling to 125% or higher.
KB5061768, which fixes BitLocker boot loop issues, is only available in the Microsoft Update Catalog. Microsoft is not rolling it out via Windows Update, so grab it using this link.

Tags

Report a problem with article

Follow @NeowinFeed
#windows #emergency #update #kb5061768 #fixes

Windows 10 emergency update KB5061768 fixes BitLocker boot loops
When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works. Windows 10 emergency update KB5061768 fixes BitLocker boot loops Taras Buria Neowin @TarasBuria · May 19, 2025 15:10 EDT As promised, Microsoft is rolling out an out-of-band emergency update for Windows 10 to address problems with Intel-based systems booting into BitLocker recovery due to LSASS issues caused by KB5058379. The company recently acknowledged a known issue on devices with Intel Trusted Execution Technologyon 10th-gen and later systems and promised to release a fix as soon as possible. The company did not take too long to address the issue, and the necessary fix is now available on the Microsoft Update Catalog as KB5061768, with build numbers 19044.5853and 19045.5853. Here is the changelog with a single fix:Fixed: A known issue on devices with Intel Trusted Execution Technologyenabled on 10th generation or later Intel vPro processors. On these systems, installing the May 13, 2025, Windows security updatemight cause the Local Security Authority Subsystem Serviceprocess to terminate unexpectedly, triggering an Automatic Repair prompting for the BitLocker recovery key to continue. In the updated post on the Windows Health Dashboard website, Microsoft says that KB5061768 is recommended for those who use affected systems and have not yet deployed KB5058379. It will prevent BitLocker problems from happening. As for systems that already cannot boot due to BitLocker loops, Microsoft suggests the following: To start Windows, temporarily disable Intel VT for Direct I/Oand Intel Trusted Execution Technologyin your BIOS / UEFI settings. This step will require entering your BitLocker recovery key. Install update KB5061768 via the Microsoft Update Catalog. After restarting Windows, return to your BIOS / UEFI settings and re-enable Intel VT for Direct I/Oand TXT.As for known bugs, there is one issue with the Noto font: text can appear blurry or unclear in Chrome, Edge, and other Chromium-based browsers in certain languages like Chinese, Japanese, and Korean when the display scaling is set to 100%. Microsoft is investigating this problem, and users can mitigate it by setting the display scaling to 125% or higher. KB5061768, which fixes BitLocker boot loop issues, is only available in the Microsoft Update Catalog. Microsoft is not rolling it out via Windows Update, so grab it using this link. Tags Report a problem with article Follow @NeowinFeed #windows #emergency #update #kb5061768 #fixes

WWW.NEOWIN.NET

Windows 10 emergency update KB5061768 fixes BitLocker boot loops

When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works. Windows 10 emergency update KB5061768 fixes BitLocker boot loops Taras Buria Neowin @TarasBuria · May 19, 2025 15:10 EDT As promised, Microsoft is rolling out an out-of-band emergency update for Windows 10 to address problems with Intel-based systems booting into BitLocker recovery due to LSASS issues caused by KB5058379. The company recently acknowledged a known issue on devices with Intel Trusted Execution Technology (TXT) on 10th-gen and later systems and promised to release a fix as soon as possible. The company did not take too long to address the issue, and the necessary fix is now available on the Microsoft Update Catalog as KB5061768, with build numbers 19044.5853 (Windows 10 version 21H2) and 19045.5853 (Windows 10 version 22H2). Here is the changelog with a single fix: [OS Security (Known Issue)] Fixed: A known issue on devices with Intel Trusted Execution Technology (TXT) enabled on 10th generation or later Intel vPro processors. On these systems, installing the May 13, 2025, Windows security update (KB5058379) might cause the Local Security Authority Subsystem Service (LSASS) process to terminate unexpectedly, triggering an Automatic Repair prompting for the BitLocker recovery key to continue. In the updated post on the Windows Health Dashboard website, Microsoft says that KB5061768 is recommended for those who use affected systems and have not yet deployed KB5058379 (May 2025 security update). It will prevent BitLocker problems from happening. As for systems that already cannot boot due to BitLocker loops, Microsoft suggests the following: To start Windows, temporarily disable Intel VT for Direct I/O (also known as VTD or VTX) and Intel Trusted Execution Technology (TXT) in your BIOS / UEFI settings. This step will require entering your BitLocker recovery key. Install update KB5061768 via the Microsoft Update Catalog. After restarting Windows, return to your BIOS / UEFI settings and re-enable Intel VT for Direct I/O (also known as VTD or VTX) and TXT. (This step will require entering your BitLocker recovery key.) As for known bugs, there is one issue with the Noto font (introduced in the recent preview update as a fallback font): text can appear blurry or unclear in Chrome, Edge, and other Chromium-based browsers in certain languages like Chinese, Japanese, and Korean when the display scaling is set to 100% (96 DPI). Microsoft is investigating this problem, and users can mitigate it by setting the display scaling to 125% or higher. KB5061768, which fixes BitLocker boot loop issues, is only available in the Microsoft Update Catalog. Microsoft is not rolling it out via Windows Update, so grab it using this link. Tags Report a problem with article Follow @NeowinFeed

·76 مشاهدة

الرجاء تسجيل الدخول , للأعجاب والمشاركة والتعليق على هذا!
GamersNexus شارك رابطًا

2025-05-19 20:46:56 ·

$8000* Disaster Prebuilt PC - Corsair & Origin Fail Again

PC Builds * Disaster Prebuilt PC - Corsair & Origin Fail AgainMay 19, 2025Last Updated: 2025-05-19We test Origin's expensive PC’s thermals, acoustics, power, frequency, and perform a tear-downThe HighlightsOur Origin Genesis PC comes with an RTX 5090, 9800X3D, and 32GB of system memoryDue to poor system thermals, the memory on the GPU fails our testingThe fans in the system don’t ramp up until the liquid-cooled CPU gets warm, which means the air-cooled GPU temperature suffersOriginal MSRP: +Release Date: January 2025Table of ContentsAutoTOC Our fully custom 3D Emblem Glasses celebrate our 15th Anniversary! We hand-assemble these on the East Coast in the US with a metal badge, strong adhesive, and high-quality pint glass. They pair excellently with our 3D 'Debug' Drink Coasters. Purchases keep us ad-free and directly support our consumer-focused reviews!IntroWe paid for Origin PC’s 5090-powered Genesis when it launched, or after taxes. Today, a similar build has a list price of Markup is to over DIY. This computer costs as much as an RTX Pro 6000, or a used car, or a brand new Kia Rio with a lifetime warranty in 2008 with passenger doors that fall off…The point is, this is expensive, and it also sucks.Editor's note: This was originally published on May 16, 2025 as a video. This content has been adapted to written format for this article and is unchanged from the original publication.CreditsTest Lead, Host, WritingSteve BurkeVideo Editing, CameraMike GaglioneTesting, WritingJeremy ClaytonCameraTim PhetdaraWriting, Web EditingJimmy ThangThe RTX 5090 is the most valuable thing in this for its 32GB of VRAM, and to show you how much they care about the only reason you’d buy this prebuilt, Origin incinerates the memory at 100 degrees Celsius by choosing to not spin the fans for 8 minutes while under load. The so-called “premium” water cooling includes tubes made out of discolored McDonald’s toy plastic that was left in the sun too long, making it look old, degraded, and dirty.But there are some upsides for this expensive computer. For example, it’s quiet, to its credit, mostly because the fans don’t spin…for 8 minutes.OverviewOriginally, this Origin Genesis pre-built cost – and that’s after taxes and a discount off the initial sticker price of We ordered it immediately after the RTX 5090 launch, which turned out to be one of the only reliable ways to actually get a 5090 with supply as bad as it was. It took a while to come in, but it did arrive in the usual Origin crate.We reviewed one of these a couple years ago that was a total disaster of a combo. The system had a severely underclocked CPU, ridiculously aggressive fan behavior, chipped paint, and a nearly unserviceable hardline custom liquid cooling loop. Hopefully this one has improved. And hopefully isn’t 1GHz below spec.Parts and PriceOrigin PC RTX 5090 + 9800X3D "Genesis" Part Prices | GamersNexusPart NameRetail Price 4/25MotherboardMSI PRO B650-P WIFICPURyzen 7 9800X3DGraphics CardNVIDIA RTX 5090 Founders EditionRAMCorsair Vengeance DDR5-6000SSD 1Corsair MP600 CORE XT 1TB PCIe 4 M.2 SSDCustom Loop"Hydro X iCUE LINK Cooling" / Pump, Rad, Block, FittingsFans12x Corsair iCUE LINK RX120 120mm FanCaseCorsair 7000D AirflowPSUCorsair RM1200x SHIFT 80+ Gold PSURGB/Fan Controller2x Corsair iCUE Link System HubOperating SystemWindows 11N/AT-ShirtORIGIN PC T-ShirtN/AMousepadORIGIN PC Mouse PadN/AShipping"ORIGIN Maximum Protection Shipping Process: ORIGIN Wooden Crate Armor"N/A???"The ORIGIN Difference: Unrivaled Quality & Performance"PricelessTotal retail cost of all parts as of April 2025We’ll price it out based on the original, pre-tariff build before taxes and with a 10% off promo. Keep in mind that the new price is to depending on when you buy.The good news is that nothing is proprietary – all of its parts are standard. The bad news is that this means we can directly compare it to retail parts which, at the time we wrote this piece, would cost making for a markup compared to the pre-tax subtotal. That’s a huge amount to pay for someone to screw the parts together. Given the price of the system, the MSI PRO B650-P WIFI motherboard and 1TB SSD are stingy and the 7000D Airflow case is old at this point. The parts don’t match the price.Just two months after we ordered and around when it finally arrived, Origin now offers a totally different case and board with the Gigabyte X870E Aorus Elite. The base SSD is still just 1TB though – only good enough for roughly two or three full Call of Duty installs. The detailed packing sheet lists 22 various water cooling fittings, but, curiously, the build itself only has 15, plus one more in the accessory kit, making it 16 by our count. We don’t know how Origin got 22 here, but it isn’t 22. Hopefully we weren’t charged for 22. Oh, and it apparently comes with “1 Integrated High-Definition.” Good. That’s good. We wouldn’t want 0 integrated high definitions.Similar to last time, you also get “The ORIGIN Difference: Unrivaled Quality & Performance” as a line item. Putting intangible, unachievable promises on the literal receipt is the Origin way: Origin’s quality is certainly rivaled.Against DIY, pricing is extreme and insane as an absolute dollar amount when the other SIs are around -markup at the high end. In order for this system to be “worth” more than DIY, it would need to be immaculate and it’s not. The only real value the PC offers is the 5090. Finding a 5090 Founders Edition now for is an increasingly unlikely scenario. Lately, price increases with scarcity and tariffs have resulted in 5090s closer to or more, so the markup with that instead would be if we assume a 5090 costs That’s still a big markup, and the motherboard is still disappointing, the tubes are still discolored, the SSD is too small, and it still has problems with the fans not properly spinning, but it’s less insane.Build QualityGetting into the parts choices:This new Genesis has a loop that’s technically set up better than the last one, but it only cools the CPU. That means we have a computer with water cooling, but only on the coolest of the two silicon parts -- the one that pulls under 150W. That leaves the 575W RTX 5090 FE to fend for itself, and that doesn’t always go well.Originally, Origin didn’t have the option to water cool the 5090. It’s just a shame that Origin isn’t owned by a gigantic PC hardware company that manufactures its own water cooling components and even has its own factories and is publicly traded and transacts billions of dollars a year to the point that it might have had enough access to make a block... A damn shame. Maybe we’ll buy from a bigger company next time.At least now, with the new sticker price of you can spend another and add a water block to the GPU. Problem solved -- turns out, we just needed to spend even more money. Here’s a closer look at Origin’s “premium” cooling solution, complete with saggy routing that looks deflated and discolored tubing that has that well-hydrated catheter tube coloring to it.The fluid is clean and the contents of the block are fine, but the tubing is the problem. In fact, the included drain tube is the correct coloring, making it even more obvious how discolored the loop is.Corsair says its XT Softline tubing is “UV-resistant tubing made to withstand the test of time without any discoloration or deforming.”So clearly something is wrong. Or not “clearly,” actually, seeing as it’s not clear. The tubing looks gross. It shouldn’t look gross. The spare piece in the accessory kit doesn’t look gross. The coolant is even Corsair’s own XL8 clear fluid, making it even more inexcusable.We’re not the only ones to have this problem, though – we found several posts online with the same issue and very little in the way of an official response from Corsair or Origin. We only saw one reply asking the user to contact support.Even without the discoloration, it comes off as looking amateurish from the way it just hangs around the inside of the case. There’s not a lot you can do about long runs of flexible tubing, unless maybe you’re the one building it and have complete control of everything in the pipeline... There is one thing we can compliment about the loop: Origin actually added a ball valve at the bottom underneath the pump for draining and maintenance, which is something that we directly complained about on the previous Origin pre-built. We’re glad to see that get addressed.The fans in the build are part of Corsair’s relatively new LINK family, so they’re all daisy chained together with a single USB-C-esque cable and controlled together in tandem by two of Corsair’s hubs. It’s an interestingsystem that extends to include the pump and CPU block – both of which have liquid temperature sensors.Tear-down Grab a GN15 Large Anti-Static Modmat to celebrate our 15th Anniversary and for a high-quality PC building work surface. The Modmat features useful PC building diagrams and is anti-static conductive. Purchases directly fund our work!We’re starting the tear-down by looking at the cable management side. Opening up the swinging side panel, we noticed masking tape on the dust filter, which we’re actually okay with as it’s to keep it in place during shipping and is removable.  Internally, they’ve included all of the unused PSU cables in the system’s accessories box, which we’ll talk more about down below. The cable routing makes sense and is generally well managed. While they tied the cables together, not all of the ties were tied down to the chassis. The system uses the cable management channel for the 24-pin connector. Overall, it’s clean and they’ve done well here. Looking at the other side of the system, we can see that the power cable leading into the 5090 is mostly seated, and isn’t a concern to us. Removing the water block’s cable, it had a little piece of plastic which acted as a pull tab. That’s actually kind of nice.Removing the screws on the water block reveal that they are captive, which is nice. Looking at the pattern, we can see that they used pre-applied paste via a silk screen. That allowed contact for all 8 legs of the IHS, which looked good with overall even pressure. The block application was also good. Looking at how well all of the cables were seated, everything was fine from the CPU fan header down to the front panel connectors. Removing the heat sync off the NVMe SSD, we didn’t see any plastic on the thermal pad, which is good. Look at the 16GB DDR 6000 RAM modules, they are in the correct slots and Origin outfitted them with Corsair 36-44-44-96 sticks, which are not the greatest timings. Examining the tightness of all the screws on the motherboard, we didn’t encounter any loose screws. Removing the motherboard from the case, everything looked fine. Looking at the motherboard out of the case, it’s a lower-end board than we’d like to see out of a premium system. Looking at the fans, they are immaculately installed, which is partially due to how they’re connected together. This results in a very clean setup.  The back side of the PC has a massive radiator. And overall, the system has very clean cable management and the assembly was mostly good. This relegates the system’s biggest issues being the value and its water-cooling setup. We didn’t drain the loop so we’re going to keep running it and see what it looks like down the road. Thermal BenchmarksSystem Thermals at Steady StateGetting into the benchmarking, we’ll start with thermals.Right away, the 96-degree result on the memory junction is a problem -- especially because this is an average, which means we have spikes periodically to 100 degrees. The technical rating on this memory is 105 degrees for maximum safety spec. This is getting way too close and is hotter than what we saw in our 5090 FE review. This is also when all of the thermal pads are brand new. The Origin pre-built uses a large case with 12 fans, so it should be impossible for the GPU to be this hot. The Ryzen 9800X3D hit 87C at steady-state – which is also not great for how much cooling is in this box. All of the various motherboard and general system temperature sensors fell well within acceptable ranges.Finally, the watercooling parts provide a couple of liquid temperatures. The pump is on the “cool” side of the loop and read 36.7C at steady state, while the coolant in the block on the “hot” side of the loop got up to 41.3C. You typically want liquid temperature to stay under 55Cto not violate spec on the pump and tubing, so this is fine.We need to plot these over time to uncover some very strange behavior.CPU Temperature vs. Fan Speeds Over TimeCPU temperature during the test starts out on a slow ramp upwards during the idle period. When the CPU load first starts, we see an immediate jump to about 72C, a brief drop, then a long and steady rise from roughly 250 seconds to 750 seconds into the test where it levels off at the 87C mark. The VRM temperature follows the same general curve, but takes longer to reach steady-state. Adding the liquid temperatures to the chart shows the same breakpoints.Finally, adding pump and fan speeds gives us the big reveal for why the curves look like this. The pump stair steps up in speed while the temperatures rise, but the fans don’t even turn on for over 8 minutes into the load’s runtime. Once they’re actually running, they average out to just 530RPM, which is so slow that they might as well be off.This is an awful configuration. Response to liquid temperature isn’t new, but this is done without any thought whatsoever. If you tie all fans to liquid temperature, and if you have parts not cooled by liquid like VRAM on the video card, then you’re going to have a bad time. And that’s the next chart. But before that one, this is an overcorrection from how Origin handled the last custom loop PC we reviewed from the company, which immediately ramped the fans up high as it could as soon as the CPU started doing anything. Maybe now they can find a middle ground since we’ve found the two extremes of thoughtless cooling.GPU Temperature vs. Fan Speeds Over TimeThis chart shows GPU temperatures versus GPU fan speed.The GPU temperature under load rises to around 83C before coming back down when the case fans finally kick on. As a reminder, 83-84 degrees is when NVIDIA starts hard throttling the clocks more than just from GPU Boost, so they’re dropping clocks as a result of this configuration.The 5090’s VRAM already runs hot on an open bench – 89 to 90 degrees Celsius – and that gets pushed up to peak at 100C in the Origin pre-built. This is unacceptable. Adding the GPU fan speed to the chart shows us how the Founders Edition cooler attempts to compensate by temporarily boosting fan speed to 56% during this time, which also means that Origin isn’t even benefiting as much from the noise levels as it should from the slower fans. Balancing them better would benefit noise more.As neat of a party trick as it is to have the case fans stay off unless they’re needed in the loop, Origin should have kept at least one or two running at all times, like rear exhaust, to give the GPU some help. Besides, letting the hot air linger could potentially encourage local hot spots to form on subcomponents that aren’t directly monitored, which can lead to problems.Power At The WallNow we’ll look at full system load power consumption by logging it at the wall – so everything, even efficiency losses from the PSU, is taken into account.Idle, it pulled a relatively high 125W. At the 180 second mark, the CPU load kicks in. There’s a jump at 235 seconds when the GPU load kicks in.We see a slight ramp upwards in power consumption after that, which tracks with increasing leakage as the parts heat up, before settling in at an average of 884W at steady state. AcousticsNext we’ll cover dBA over time as measured in our hemi-anechoic chamber.At idle, the fans are off, which makes for a functionally silent system at the noise floor. The first fans to come on in the system are on the GPU, bringing noise levels up to a still-quiet range of 25-28dBA at 1 meter. The loudest point is 30.5 dBA when the GPU fans briefly ramp and before system fans kick in. CPU Frequency vs. Original ReviewFor CPU frequency, fortunately for Origin, it didn’t randomly throttle it by 1GHz this time. The 9800X3D managed to stay at 5225MHz during the CPU-only load portion of torture test – the same frequency that we recorded in our original review for the CPU so that’ good. At steady state with the GPU dumping over 500W of heat into the case, the average core frequency dropped by 50MHz. If Origin made better use of its dozen or so fans, it should hold onto more of that frequency. BIOS ConfigurationBIOS for the Origin pre-built is set up sensibly, at least. The build date is January 23, which was the latest available in the time between when we ordered the system at the 50 series launch and when the system was actually assembled.Scrutinizing the chosen settings revealed nothing out of line. The DDR5-6000 memory profile was enabled and the rest of the core settings were properly set to Auto. This was all fine.Setup and SoftwareThe Windows install was normal with no bloatware. That’s also good.The desktop had a few things on it. A “Link Windows 10 Key to Microsoft Account” PDF is helpful for people who don’t know what to do if their system shows the Activate Windows watermark. Confusingly, it hasn’t been updated to say “11” instead of “10.” It also shepherds the user towards using a Microsoft account. That’s not necessarily a bad thing, but we don’t like how it makes it seem necessary because it’s not and you shouldn’t. There’s also an “Origin PC ReadMe” PDF that doesn’t offer much except coverage for Origin’s ass with disclaimers and points of contact for support. One useful thing is that it points the user to “C:\\ORIGIN PC” to find “important items.”That folder has Origin branded gifs, logos, and wallpapers, as well as CPU-Z, Teamviewer, and a Results folder. Teamviewer is almost certainly for Origin’s support teams to be able to remotely inspect the PC during support calls. It makes sense to have that stuff on there. The results folder contains an OCCT test report that shows a total of 1 hour and 52 minutes of testing. A CPU test for 12 minutes, CPU + RAM, memory, and 3D adaptive tests for 30 minutes each, then finishing with 10 minutes of OCCT’s “power” test, which is a combined full system load. It’s great that Origin actually does testing and provides this log as a baseline for future issues, and just for base expectations. This is good and gives you something to work from. Not having OCCT pre-installed to actually run again for comparison is a support oversight. It’s free for personal use at least, so the user could go download it easily.There weren’t any missing drivers in Device Manager and NVIDIA’s 572.47 driver from February 20 was the latest at the time of the build – both good things. There wasn’t any bundled bloatware installed, so points to Origin for that.iCUE itself isn’t as bad as it used to be, but it’s still clunky, like the preloaded fan profiles not showing their set points. PackagingOn to packaging.The Origin Genesis pre-built came in a massive wooden crate that was big enough for two people to move around. Considering this PC was after taxes, we’re definitely OK with the wooden crate and its QR code opening instructions.Origin uses foam, a fabric cover, a cardboard box within a crate, and the crate for the PC. The case had two packs of expanding foam inside it, allowing the GPU to arrive undamaged and installed. The sticker on the side panel also had clear instructions. These are good things. Unfortunately, there’s a small chip in the paint on top of the case, but not as bad as the last Origin paint issues we had and we think it’s unrelated to the packaging itself.AccessoriesThe accessory kit is basic, and came inside of a box with the overused cringey adage “EAT SLEEP GAME REPEAT” printed on it. Inside are the spare PSU cables, an AC power cable, stock 5090 FE power adapter, standard motherboard and case accessories, a G1/4 plug tool and extra plugs, and a piece of soft tubing with a fitting on one end that can be used to help drain the cooling loop. All of this is good.Conclusion Visit our Patreon page to contribute a few dollars toward this website's operationAdditionally, when you purchase through links to retailers on our site, we may earn a small affiliate commission.During this review process, the price went even higher. You already shouldn’t buy this, but just to drive it home:Now, for the same configuration, the Genesis now costs after the discount, off the new sticker price of That’s an increase of over making the premium over current DIY pricing roughly -Now, there are good reasons for the price to go up. Tariffs have a real impact on pricing and we’re going to see it everywhere, and tariffs are also outside of Corsair’s control. We don’t fault them for that. But that doesn’t change the fact that the cost over DIY is so insanely elevated. Even Corsair’s own competitors offer better value than this, like Maingear.At sticker price, you’d have to be drunk on whatever is discoloring Origin’s loop to buy it. Nobody should buy this, especially not for gaming. If you’re doing productivity or creative work that would seriously benefit from the 5090’s 32GB of VRAM, then look elsewhere for a better deal. This costs nearly as much as an RTX Pro 6000, which has 96GB of VRAM and is better.It would actually be cheaper to get scalped for a 5090 on Ebay and then buy the whole rest of the computer than to buy this Origin system. That’s how crazy this is.The upcharge, even assuming a 5090 price of is just way too high versus other system integrators. Seriously, Alienware is cheaper at this point – by thousands of dollars. Alienware.We can’t recommend this PC. Ignoring the price, the memory on the video card is hitting 100 degrees C in workloads when the fans aren’t turning on because the fans are set to turn on based on the liquid temperature and the liquid doesn’t touch the GPU. For that reason alone, it gets a failing grade. For our thermal testing, pre-builts have to pass the torture test. If they don’t, they instantly fail. That’s how it always works for our pre-built reviews. This system has, unfortunately, instantly failed.
#disaster #prebuilt #corsair #ampamp #origin

$8000* Disaster Prebuilt PC - Corsair & Origin Fail Again
PC Builds * Disaster Prebuilt PC - Corsair & Origin Fail AgainMay 19, 2025Last Updated: 2025-05-19We test Origin's expensive PC’s thermals, acoustics, power, frequency, and perform a tear-downThe HighlightsOur Origin Genesis PC comes with an RTX 5090, 9800X3D, and 32GB of system memoryDue to poor system thermals, the memory on the GPU fails our testingThe fans in the system don’t ramp up until the liquid-cooled CPU gets warm, which means the air-cooled GPU temperature suffersOriginal MSRP: +Release Date: January 2025Table of ContentsAutoTOC Our fully custom 3D Emblem Glasses celebrate our 15th Anniversary! We hand-assemble these on the East Coast in the US with a metal badge, strong adhesive, and high-quality pint glass. They pair excellently with our 3D 'Debug' Drink Coasters. Purchases keep us ad-free and directly support our consumer-focused reviews!IntroWe paid for Origin PC’s 5090-powered Genesis when it launched, or after taxes. Today, a similar build has a list price of Markup is to over DIY. This computer costs as much as an RTX Pro 6000, or a used car, or a brand new Kia Rio with a lifetime warranty in 2008 with passenger doors that fall off…The point is, this is expensive, and it also sucks.Editor's note: This was originally published on May 16, 2025 as a video. This content has been adapted to written format for this article and is unchanged from the original publication.CreditsTest Lead, Host, WritingSteve BurkeVideo Editing, CameraMike GaglioneTesting, WritingJeremy ClaytonCameraTim PhetdaraWriting, Web EditingJimmy ThangThe RTX 5090 is the most valuable thing in this for its 32GB of VRAM, and to show you how much they care about the only reason you’d buy this prebuilt, Origin incinerates the memory at 100 degrees Celsius by choosing to not spin the fans for 8 minutes while under load. The so-called “premium” water cooling includes tubes made out of discolored McDonald’s toy plastic that was left in the sun too long, making it look old, degraded, and dirty.But there are some upsides for this expensive computer. For example, it’s quiet, to its credit, mostly because the fans don’t spin…for 8 minutes.OverviewOriginally, this Origin Genesis pre-built cost – and that’s after taxes and a discount off the initial sticker price of We ordered it immediately after the RTX 5090 launch, which turned out to be one of the only reliable ways to actually get a 5090 with supply as bad as it was. It took a while to come in, but it did arrive in the usual Origin crate.We reviewed one of these a couple years ago that was a total disaster of a combo. The system had a severely underclocked CPU, ridiculously aggressive fan behavior, chipped paint, and a nearly unserviceable hardline custom liquid cooling loop. Hopefully this one has improved. And hopefully isn’t 1GHz below spec.Parts and PriceOrigin PC RTX 5090 + 9800X3D "Genesis" Part Prices | GamersNexusPart NameRetail Price 4/25MotherboardMSI PRO B650-P WIFICPURyzen 7 9800X3DGraphics CardNVIDIA RTX 5090 Founders EditionRAMCorsair Vengeance DDR5-6000SSD 1Corsair MP600 CORE XT 1TB PCIe 4 M.2 SSDCustom Loop"Hydro X iCUE LINK Cooling" / Pump, Rad, Block, FittingsFans12x Corsair iCUE LINK RX120 120mm FanCaseCorsair 7000D AirflowPSUCorsair RM1200x SHIFT 80+ Gold PSURGB/Fan Controller2x Corsair iCUE Link System HubOperating SystemWindows 11N/AT-ShirtORIGIN PC T-ShirtN/AMousepadORIGIN PC Mouse PadN/AShipping"ORIGIN Maximum Protection Shipping Process: ORIGIN Wooden Crate Armor"N/A???"The ORIGIN Difference: Unrivaled Quality & Performance"PricelessTotal retail cost of all parts as of April 2025We’ll price it out based on the original, pre-tariff build before taxes and with a 10% off promo. Keep in mind that the new price is to depending on when you buy.The good news is that nothing is proprietary – all of its parts are standard. The bad news is that this means we can directly compare it to retail parts which, at the time we wrote this piece, would cost making for a markup compared to the pre-tax subtotal. That’s a huge amount to pay for someone to screw the parts together. Given the price of the system, the MSI PRO B650-P WIFI motherboard and 1TB SSD are stingy and the 7000D Airflow case is old at this point. The parts don’t match the price.Just two months after we ordered and around when it finally arrived, Origin now offers a totally different case and board with the Gigabyte X870E Aorus Elite. The base SSD is still just 1TB though – only good enough for roughly two or three full Call of Duty installs. The detailed packing sheet lists 22 various water cooling fittings, but, curiously, the build itself only has 15, plus one more in the accessory kit, making it 16 by our count. We don’t know how Origin got 22 here, but it isn’t 22. Hopefully we weren’t charged for 22. Oh, and it apparently comes with “1 Integrated High-Definition.” Good. That’s good. We wouldn’t want 0 integrated high definitions.Similar to last time, you also get “The ORIGIN Difference: Unrivaled Quality & Performance” as a line item. Putting intangible, unachievable promises on the literal receipt is the Origin way: Origin’s quality is certainly rivaled.Against DIY, pricing is extreme and insane as an absolute dollar amount when the other SIs are around -markup at the high end. In order for this system to be “worth” more than DIY, it would need to be immaculate and it’s not. The only real value the PC offers is the 5090. Finding a 5090 Founders Edition now for is an increasingly unlikely scenario. Lately, price increases with scarcity and tariffs have resulted in 5090s closer to or more, so the markup with that instead would be if we assume a 5090 costs That’s still a big markup, and the motherboard is still disappointing, the tubes are still discolored, the SSD is too small, and it still has problems with the fans not properly spinning, but it’s less insane.Build QualityGetting into the parts choices:This new Genesis has a loop that’s technically set up better than the last one, but it only cools the CPU. That means we have a computer with water cooling, but only on the coolest of the two silicon parts -- the one that pulls under 150W. That leaves the 575W RTX 5090 FE to fend for itself, and that doesn’t always go well.Originally, Origin didn’t have the option to water cool the 5090. It’s just a shame that Origin isn’t owned by a gigantic PC hardware company that manufactures its own water cooling components and even has its own factories and is publicly traded and transacts billions of dollars a year to the point that it might have had enough access to make a block... A damn shame. Maybe we’ll buy from a bigger company next time.At least now, with the new sticker price of you can spend another and add a water block to the GPU. Problem solved -- turns out, we just needed to spend even more money. Here’s a closer look at Origin’s “premium” cooling solution, complete with saggy routing that looks deflated and discolored tubing that has that well-hydrated catheter tube coloring to it.The fluid is clean and the contents of the block are fine, but the tubing is the problem. In fact, the included drain tube is the correct coloring, making it even more obvious how discolored the loop is.Corsair says its XT Softline tubing is “UV-resistant tubing made to withstand the test of time without any discoloration or deforming.”So clearly something is wrong. Or not “clearly,” actually, seeing as it’s not clear. The tubing looks gross. It shouldn’t look gross. The spare piece in the accessory kit doesn’t look gross. The coolant is even Corsair’s own XL8 clear fluid, making it even more inexcusable.We’re not the only ones to have this problem, though – we found several posts online with the same issue and very little in the way of an official response from Corsair or Origin. We only saw one reply asking the user to contact support.Even without the discoloration, it comes off as looking amateurish from the way it just hangs around the inside of the case. There’s not a lot you can do about long runs of flexible tubing, unless maybe you’re the one building it and have complete control of everything in the pipeline... There is one thing we can compliment about the loop: Origin actually added a ball valve at the bottom underneath the pump for draining and maintenance, which is something that we directly complained about on the previous Origin pre-built. We’re glad to see that get addressed.The fans in the build are part of Corsair’s relatively new LINK family, so they’re all daisy chained together with a single USB-C-esque cable and controlled together in tandem by two of Corsair’s hubs. It’s an interestingsystem that extends to include the pump and CPU block – both of which have liquid temperature sensors.Tear-down Grab a GN15 Large Anti-Static Modmat to celebrate our 15th Anniversary and for a high-quality PC building work surface. The Modmat features useful PC building diagrams and is anti-static conductive. Purchases directly fund our work!We’re starting the tear-down by looking at the cable management side. Opening up the swinging side panel, we noticed masking tape on the dust filter, which we’re actually okay with as it’s to keep it in place during shipping and is removable.  Internally, they’ve included all of the unused PSU cables in the system’s accessories box, which we’ll talk more about down below. The cable routing makes sense and is generally well managed. While they tied the cables together, not all of the ties were tied down to the chassis. The system uses the cable management channel for the 24-pin connector. Overall, it’s clean and they’ve done well here. Looking at the other side of the system, we can see that the power cable leading into the 5090 is mostly seated, and isn’t a concern to us. Removing the water block’s cable, it had a little piece of plastic which acted as a pull tab. That’s actually kind of nice.Removing the screws on the water block reveal that they are captive, which is nice. Looking at the pattern, we can see that they used pre-applied paste via a silk screen. That allowed contact for all 8 legs of the IHS, which looked good with overall even pressure. The block application was also good. Looking at how well all of the cables were seated, everything was fine from the CPU fan header down to the front panel connectors. Removing the heat sync off the NVMe SSD, we didn’t see any plastic on the thermal pad, which is good. Look at the 16GB DDR 6000 RAM modules, they are in the correct slots and Origin outfitted them with Corsair 36-44-44-96 sticks, which are not the greatest timings. Examining the tightness of all the screws on the motherboard, we didn’t encounter any loose screws. Removing the motherboard from the case, everything looked fine. Looking at the motherboard out of the case, it’s a lower-end board than we’d like to see out of a premium system. Looking at the fans, they are immaculately installed, which is partially due to how they’re connected together. This results in a very clean setup.  The back side of the PC has a massive radiator. And overall, the system has very clean cable management and the assembly was mostly good. This relegates the system’s biggest issues being the value and its water-cooling setup. We didn’t drain the loop so we’re going to keep running it and see what it looks like down the road. Thermal BenchmarksSystem Thermals at Steady StateGetting into the benchmarking, we’ll start with thermals.Right away, the 96-degree result on the memory junction is a problem -- especially because this is an average, which means we have spikes periodically to 100 degrees. The technical rating on this memory is 105 degrees for maximum safety spec. This is getting way too close and is hotter than what we saw in our 5090 FE review. This is also when all of the thermal pads are brand new. The Origin pre-built uses a large case with 12 fans, so it should be impossible for the GPU to be this hot. The Ryzen 9800X3D hit 87C at steady-state – which is also not great for how much cooling is in this box. All of the various motherboard and general system temperature sensors fell well within acceptable ranges.Finally, the watercooling parts provide a couple of liquid temperatures. The pump is on the “cool” side of the loop and read 36.7C at steady state, while the coolant in the block on the “hot” side of the loop got up to 41.3C. You typically want liquid temperature to stay under 55Cto not violate spec on the pump and tubing, so this is fine.We need to plot these over time to uncover some very strange behavior.CPU Temperature vs. Fan Speeds Over TimeCPU temperature during the test starts out on a slow ramp upwards during the idle period. When the CPU load first starts, we see an immediate jump to about 72C, a brief drop, then a long and steady rise from roughly 250 seconds to 750 seconds into the test where it levels off at the 87C mark. The VRM temperature follows the same general curve, but takes longer to reach steady-state. Adding the liquid temperatures to the chart shows the same breakpoints.Finally, adding pump and fan speeds gives us the big reveal for why the curves look like this. The pump stair steps up in speed while the temperatures rise, but the fans don’t even turn on for over 8 minutes into the load’s runtime. Once they’re actually running, they average out to just 530RPM, which is so slow that they might as well be off.This is an awful configuration. Response to liquid temperature isn’t new, but this is done without any thought whatsoever. If you tie all fans to liquid temperature, and if you have parts not cooled by liquid like VRAM on the video card, then you’re going to have a bad time. And that’s the next chart. But before that one, this is an overcorrection from how Origin handled the last custom loop PC we reviewed from the company, which immediately ramped the fans up high as it could as soon as the CPU started doing anything. Maybe now they can find a middle ground since we’ve found the two extremes of thoughtless cooling.GPU Temperature vs. Fan Speeds Over TimeThis chart shows GPU temperatures versus GPU fan speed.The GPU temperature under load rises to around 83C before coming back down when the case fans finally kick on. As a reminder, 83-84 degrees is when NVIDIA starts hard throttling the clocks more than just from GPU Boost, so they’re dropping clocks as a result of this configuration.The 5090’s VRAM already runs hot on an open bench – 89 to 90 degrees Celsius – and that gets pushed up to peak at 100C in the Origin pre-built. This is unacceptable. Adding the GPU fan speed to the chart shows us how the Founders Edition cooler attempts to compensate by temporarily boosting fan speed to 56% during this time, which also means that Origin isn’t even benefiting as much from the noise levels as it should from the slower fans. Balancing them better would benefit noise more.As neat of a party trick as it is to have the case fans stay off unless they’re needed in the loop, Origin should have kept at least one or two running at all times, like rear exhaust, to give the GPU some help. Besides, letting the hot air linger could potentially encourage local hot spots to form on subcomponents that aren’t directly monitored, which can lead to problems.Power At The WallNow we’ll look at full system load power consumption by logging it at the wall – so everything, even efficiency losses from the PSU, is taken into account.Idle, it pulled a relatively high 125W. At the 180 second mark, the CPU load kicks in. There’s a jump at 235 seconds when the GPU load kicks in.We see a slight ramp upwards in power consumption after that, which tracks with increasing leakage as the parts heat up, before settling in at an average of 884W at steady state. AcousticsNext we’ll cover dBA over time as measured in our hemi-anechoic chamber.At idle, the fans are off, which makes for a functionally silent system at the noise floor. The first fans to come on in the system are on the GPU, bringing noise levels up to a still-quiet range of 25-28dBA at 1 meter. The loudest point is 30.5 dBA when the GPU fans briefly ramp and before system fans kick in. CPU Frequency vs. Original ReviewFor CPU frequency, fortunately for Origin, it didn’t randomly throttle it by 1GHz this time. The 9800X3D managed to stay at 5225MHz during the CPU-only load portion of torture test – the same frequency that we recorded in our original review for the CPU so that’ good. At steady state with the GPU dumping over 500W of heat into the case, the average core frequency dropped by 50MHz. If Origin made better use of its dozen or so fans, it should hold onto more of that frequency. BIOS ConfigurationBIOS for the Origin pre-built is set up sensibly, at least. The build date is January 23, which was the latest available in the time between when we ordered the system at the 50 series launch and when the system was actually assembled.Scrutinizing the chosen settings revealed nothing out of line. The DDR5-6000 memory profile was enabled and the rest of the core settings were properly set to Auto. This was all fine.Setup and SoftwareThe Windows install was normal with no bloatware. That’s also good.The desktop had a few things on it. A “Link Windows 10 Key to Microsoft Account” PDF is helpful for people who don’t know what to do if their system shows the Activate Windows watermark. Confusingly, it hasn’t been updated to say “11” instead of “10.” It also shepherds the user towards using a Microsoft account. That’s not necessarily a bad thing, but we don’t like how it makes it seem necessary because it’s not and you shouldn’t. There’s also an “Origin PC ReadMe” PDF that doesn’t offer much except coverage for Origin’s ass with disclaimers and points of contact for support. One useful thing is that it points the user to “C:\\ORIGIN PC” to find “important items.”That folder has Origin branded gifs, logos, and wallpapers, as well as CPU-Z, Teamviewer, and a Results folder. Teamviewer is almost certainly for Origin’s support teams to be able to remotely inspect the PC during support calls. It makes sense to have that stuff on there. The results folder contains an OCCT test report that shows a total of 1 hour and 52 minutes of testing. A CPU test for 12 minutes, CPU + RAM, memory, and 3D adaptive tests for 30 minutes each, then finishing with 10 minutes of OCCT’s “power” test, which is a combined full system load. It’s great that Origin actually does testing and provides this log as a baseline for future issues, and just for base expectations. This is good and gives you something to work from. Not having OCCT pre-installed to actually run again for comparison is a support oversight. It’s free for personal use at least, so the user could go download it easily.There weren’t any missing drivers in Device Manager and NVIDIA’s 572.47 driver from February 20 was the latest at the time of the build – both good things. There wasn’t any bundled bloatware installed, so points to Origin for that.iCUE itself isn’t as bad as it used to be, but it’s still clunky, like the preloaded fan profiles not showing their set points. PackagingOn to packaging.The Origin Genesis pre-built came in a massive wooden crate that was big enough for two people to move around. Considering this PC was after taxes, we’re definitely OK with the wooden crate and its QR code opening instructions.Origin uses foam, a fabric cover, a cardboard box within a crate, and the crate for the PC. The case had two packs of expanding foam inside it, allowing the GPU to arrive undamaged and installed. The sticker on the side panel also had clear instructions. These are good things. Unfortunately, there’s a small chip in the paint on top of the case, but not as bad as the last Origin paint issues we had and we think it’s unrelated to the packaging itself.AccessoriesThe accessory kit is basic, and came inside of a box with the overused cringey adage “EAT SLEEP GAME REPEAT” printed on it. Inside are the spare PSU cables, an AC power cable, stock 5090 FE power adapter, standard motherboard and case accessories, a G1/4 plug tool and extra plugs, and a piece of soft tubing with a fitting on one end that can be used to help drain the cooling loop. All of this is good.Conclusion Visit our Patreon page to contribute a few dollars toward this website's operationAdditionally, when you purchase through links to retailers on our site, we may earn a small affiliate commission.During this review process, the price went even higher. You already shouldn’t buy this, but just to drive it home:Now, for the same configuration, the Genesis now costs after the discount, off the new sticker price of That’s an increase of over making the premium over current DIY pricing roughly -Now, there are good reasons for the price to go up. Tariffs have a real impact on pricing and we’re going to see it everywhere, and tariffs are also outside of Corsair’s control. We don’t fault them for that. But that doesn’t change the fact that the cost over DIY is so insanely elevated. Even Corsair’s own competitors offer better value than this, like Maingear.At sticker price, you’d have to be drunk on whatever is discoloring Origin’s loop to buy it. Nobody should buy this, especially not for gaming. If you’re doing productivity or creative work that would seriously benefit from the 5090’s 32GB of VRAM, then look elsewhere for a better deal. This costs nearly as much as an RTX Pro 6000, which has 96GB of VRAM and is better.It would actually be cheaper to get scalped for a 5090 on Ebay and then buy the whole rest of the computer than to buy this Origin system. That’s how crazy this is.The upcharge, even assuming a 5090 price of is just way too high versus other system integrators. Seriously, Alienware is cheaper at this point – by thousands of dollars. Alienware.We can’t recommend this PC. Ignoring the price, the memory on the video card is hitting 100 degrees C in workloads when the fans aren’t turning on because the fans are set to turn on based on the liquid temperature and the liquid doesn’t touch the GPU. For that reason alone, it gets a failing grade. For our thermal testing, pre-builts have to pass the torture test. If they don’t, they instantly fail. That’s how it always works for our pre-built reviews. This system has, unfortunately, instantly failed. #disaster #prebuilt #corsair #ampamp #origin

GAMERSNEXUS.NET

$8000* Disaster Prebuilt PC - Corsair & Origin Fail Again

PC Builds $8000* Disaster Prebuilt PC - Corsair & Origin Fail AgainMay 19, 2025Last Updated: 2025-05-19We test Origin's expensive PC’s thermals, acoustics, power, frequency, and perform a tear-downThe HighlightsOur Origin Genesis PC comes with an RTX 5090, 9800X3D, and 32GB of system memoryDue to poor system thermals, the memory on the GPU fails our testingThe fans in the system don’t ramp up until the liquid-cooled CPU gets warm, which means the air-cooled GPU temperature suffersOriginal MSRP: $6,050+Release Date: January 2025Table of ContentsAutoTOC Our fully custom 3D Emblem Glasses celebrate our 15th Anniversary! We hand-assemble these on the East Coast in the US with a metal badge, strong adhesive, and high-quality pint glass. They pair excellently with our 3D 'Debug' Drink Coasters. Purchases keep us ad-free and directly support our consumer-focused reviews!IntroWe paid $6,050 for Origin PC’s 5090-powered Genesis when it launched, or $6,500 after taxes. Today, a similar build has a list price of $8,396. Markup is $1,700 to $2,500 over DIY. This computer costs as much as an RTX Pro 6000, or a used car, or a brand new Kia Rio with a lifetime warranty in 2008 with passenger doors that fall off…The point is, this is expensive, and it also sucks.Editor's note: This was originally published on May 16, 2025 as a video. This content has been adapted to written format for this article and is unchanged from the original publication.CreditsTest Lead, Host, WritingSteve BurkeVideo Editing, CameraMike GaglioneTesting, WritingJeremy ClaytonCameraTim PhetdaraWriting, Web EditingJimmy ThangThe RTX 5090 is the most valuable thing in this for its 32GB of VRAM, and to show you how much they care about the only reason you’d buy this prebuilt, Origin incinerates the memory at 100 degrees Celsius by choosing to not spin the fans for 8 minutes while under load. The so-called “premium” water cooling includes tubes made out of discolored McDonald’s toy plastic that was left in the sun too long, making it look old, degraded, and dirty.But there are some upsides for this expensive computer. For example, it’s quiet, to its credit, mostly because the fans don’t spin…for 8 minutes.OverviewOriginally, this Origin Genesis pre-built cost $6,488 – and that’s after taxes and a $672 discount off the initial sticker price of $6,722. We ordered it immediately after the RTX 5090 launch, which turned out to be one of the only reliable ways to actually get a 5090 with supply as bad as it was (and continues to be). It took a while to come in, but it did arrive in the usual Origin crate.We reviewed one of these a couple years ago that was a total disaster of a combo. The system had a severely underclocked CPU, ridiculously aggressive fan behavior (which is the opposite of the system we’re reviewing today), chipped paint, and a nearly unserviceable hardline custom liquid cooling loop. Hopefully this one has improved. And hopefully isn’t 1GHz below spec.Parts and PriceOrigin PC RTX 5090 + 9800X3D "Genesis" Part Prices | GamersNexusPart NameRetail Price 4/25MotherboardMSI PRO B650-P WIFI$190CPURyzen 7 9800X3D$480Graphics CardNVIDIA RTX 5090 Founders Edition$2,000RAMCorsair Vengeance DDR5-6000 (2x16GB)$93SSD 1Corsair MP600 CORE XT 1TB PCIe 4 M.2 SSD$70Custom Loop"Hydro X iCUE LINK Cooling" / Pump, Rad, Block, Fittings$712Fans12x Corsair iCUE LINK RX120 120mm Fan$360CaseCorsair 7000D Airflow$240PSUCorsair RM1200x SHIFT 80+ Gold PSU$230RGB/Fan Controller2x Corsair iCUE Link System Hub$118Operating SystemWindows 11N/AT-ShirtORIGIN PC T-ShirtN/AMousepadORIGIN PC Mouse PadN/AShipping"ORIGIN Maximum Protection Shipping Process: ORIGIN Wooden Crate Armor"N/A???"The ORIGIN Difference: Unrivaled Quality & Performance"PricelessTotal retail cost of all parts as of April 2025$4,493We’ll price it out based on the original, pre-tariff $6,050 build before taxes and with a 10% off promo. Keep in mind that the new price is $7,500 to $8,400, depending on when you buy.The good news is that nothing is proprietary – all of its parts are standard. The bad news is that this means we can directly compare it to retail parts which, at the time we wrote this piece, would cost $4,493, making for a $1,557 markup compared to the pre-tax subtotal. That’s a huge amount to pay for someone to screw the parts together. Given the price of the system, the MSI PRO B650-P WIFI motherboard and 1TB SSD are stingy and the 7000D Airflow case is old at this point. The parts don’t match the price.Just two months after we ordered and around when it finally arrived, Origin now offers a totally different case and board with the Gigabyte X870E Aorus Elite. The base SSD is still just 1TB though – only good enough for roughly two or three full Call of Duty installs. The detailed packing sheet lists 22 various water cooling fittings, but, curiously, the build itself only has 15, plus one more in the accessory kit, making it 16 by our count. We don’t know how Origin got 22 here, but it isn’t 22. Hopefully we weren’t charged for 22. Oh, and it apparently comes with “1 Integrated High-Definition.” Good. That’s good. We wouldn’t want 0 integrated high definitions.Similar to last time, you also get “The ORIGIN Difference: Unrivaled Quality & Performance” as a line item. Putting intangible, unachievable promises on the literal receipt is the Origin way: Origin’s quality is certainly rivaled.Against DIY, pricing is extreme and insane as an absolute dollar amount when the other SIs are around $500-$800 markup at the high end. In order for this system to be “worth” $1,500 more than DIY, it would need to be immaculate and it’s not. The only real value the PC offers is the 5090. Finding a 5090 Founders Edition now for $2,000 is an increasingly unlikely scenario. Lately, price increases with scarcity and tariffs have resulted in 5090s closer to $2,800 or more, so the markup with that instead would be $777 if we assume a 5090 costs $2,800. That’s still a big markup, and the motherboard is still disappointing, the tubes are still discolored, the SSD is too small, and it still has problems with the fans not properly spinning, but it’s less insane.Build QualityGetting into the parts choices:This new Genesis has a loop that’s technically set up better than the last one, but it only cools the CPU. That means we have a $6,500 computer with water cooling, but only on the coolest of the two silicon parts -- the one that pulls under 150W. That leaves the 575W RTX 5090 FE to fend for itself, and that doesn’t always go well.Originally, Origin didn’t have the option to water cool the 5090. It’s just a shame that Origin isn’t owned by a gigantic PC hardware company that manufactures its own water cooling components and even has its own factories and is publicly traded and transacts billions of dollars a year to the point that it might have had enough access to make a block... A damn shame. Maybe we’ll buy from a bigger company next time.At least now, with the new sticker price of $8,400, you can spend another $200 and add a water block to the GPU. Problem solved -- turns out, we just needed to spend even more money. Here’s a closer look at Origin’s “premium” cooling solution, complete with saggy routing that looks deflated and discolored tubing that has that well-hydrated catheter tube coloring to it.The fluid is clean and the contents of the block are fine, but the tubing is the problem. In fact, the included drain tube is the correct coloring, making it even more obvious how discolored the loop is.Corsair says its XT Softline tubing is “UV-resistant tubing made to withstand the test of time without any discoloration or deforming.”So clearly something is wrong. Or not “clearly,” actually, seeing as it’s not clear. The tubing looks gross. It shouldn’t look gross. The spare piece in the accessory kit doesn’t look gross. The coolant is even Corsair’s own XL8 clear fluid, making it even more inexcusable.We’re not the only ones to have this problem, though – we found several posts online with the same issue and very little in the way of an official response from Corsair or Origin. We only saw one reply asking the user to contact support.Even without the discoloration, it comes off as looking amateurish from the way it just hangs around the inside of the case. There’s not a lot you can do about long runs of flexible tubing, unless maybe you’re the one building it and have complete control of everything in the pipeline... There is one thing we can compliment about the loop: Origin actually added a ball valve at the bottom underneath the pump for draining and maintenance, which is something that we directly complained about on the previous Origin pre-built. We’re glad to see that get addressed.The fans in the build are part of Corsair’s relatively new LINK family, so they’re all daisy chained together with a single USB-C-esque cable and controlled together in tandem by two of Corsair’s hubs. It’s an interesting (if expensive) system that extends to include the pump and CPU block – both of which have liquid temperature sensors.Tear-down Grab a GN15 Large Anti-Static Modmat to celebrate our 15th Anniversary and for a high-quality PC building work surface. The Modmat features useful PC building diagrams and is anti-static conductive. Purchases directly fund our work! (or consider a direct donation or a Patreon contribution!)We’re starting the tear-down by looking at the cable management side. Opening up the swinging side panel, we noticed masking tape on the dust filter, which we’re actually okay with as it’s to keep it in place during shipping and is removable.  Internally, they’ve included all of the unused PSU cables in the system’s accessories box, which we’ll talk more about down below. The cable routing makes sense and is generally well managed. While they tied the cables together, not all of the ties were tied down to the chassis. The system uses the cable management channel for the 24-pin connector. Overall, it’s clean and they’ve done well here. Looking at the other side of the system, we can see that the power cable leading into the 5090 is mostly seated, and isn’t a concern to us. Removing the water block’s cable, it had a little piece of plastic which acted as a pull tab. That’s actually kind of nice.Removing the screws on the water block reveal that they are captive, which is nice. Looking at the pattern, we can see that they used pre-applied paste via a silk screen. That allowed contact for all 8 legs of the IHS, which looked good with overall even pressure. The block application was also good. Looking at how well all of the cables were seated, everything was fine from the CPU fan header down to the front panel connectors. Removing the heat sync off the NVMe SSD, we didn’t see any plastic on the thermal pad, which is good. Look at the 16GB DDR 6000 RAM modules, they are in the correct slots and Origin outfitted them with Corsair 36-44-44-96 sticks, which are not the greatest timings. Examining the tightness of all the screws on the motherboard, we didn’t encounter any loose screws. Removing the motherboard from the case, everything looked fine. Looking at the motherboard out of the case, it’s a lower-end board than we’d like to see out of a premium system. Looking at the fans, they are immaculately installed, which is partially due to how they’re connected together. This results in a very clean setup.  The back side of the PC has a massive radiator. And overall, the system has very clean cable management and the assembly was mostly good. This relegates the system’s biggest issues being the value and its water-cooling setup. We didn’t drain the loop so we’re going to keep running it and see what it looks like down the road. Thermal BenchmarksSystem Thermals at Steady StateGetting into the benchmarking, we’ll start with thermals.Right away, the 96-degree result on the memory junction is a problem -- especially because this is an average, which means we have spikes periodically to 100 degrees. The technical rating on this memory is 105 degrees for maximum safety spec. This is getting way too close and is hotter than what we saw in our 5090 FE review. This is also when all of the thermal pads are brand new. The Origin pre-built uses a large case with 12 fans, so it should be impossible for the GPU to be this hot. The Ryzen 9800X3D hit 87C at steady-state – which is also not great for how much cooling is in this box. All of the various motherboard and general system temperature sensors fell well within acceptable ranges.Finally, the watercooling parts provide a couple of liquid temperatures. The pump is on the “cool” side of the loop and read 36.7C at steady state, while the coolant in the block on the “hot” side of the loop got up to 41.3C. You typically want liquid temperature to stay under 55C (at the most) to not violate spec on the pump and tubing, so this is fine.We need to plot these over time to uncover some very strange behavior.CPU Temperature vs. Fan Speeds Over TimeCPU temperature during the test starts out on a slow ramp upwards during the idle period. When the CPU load first starts, we see an immediate jump to about 72C, a brief drop, then a long and steady rise from roughly 250 seconds to 750 seconds into the test where it levels off at the 87C mark. The VRM temperature follows the same general curve, but takes longer to reach steady-state. Adding the liquid temperatures to the chart shows the same breakpoints.Finally, adding pump and fan speeds gives us the big reveal for why the curves look like this. The pump stair steps up in speed while the temperatures rise, but the fans don’t even turn on for over 8 minutes into the load’s runtime. Once they’re actually running, they average out to just 530RPM, which is so slow that they might as well be off.This is an awful configuration. Response to liquid temperature isn’t new, but this is done without any thought whatsoever. If you tie all fans to liquid temperature, and if you have parts not cooled by liquid like VRAM on the video card, then you’re going to have a bad time. And that’s the next chart. But before that one, this is an overcorrection from how Origin handled the last custom loop PC we reviewed from the company, which immediately ramped the fans up high as it could as soon as the CPU started doing anything. Maybe now they can find a middle ground since we’ve found the two extremes of thoughtless cooling.GPU Temperature vs. Fan Speeds Over TimeThis chart shows GPU temperatures versus GPU fan speed.The GPU temperature under load rises to around 83C before coming back down when the case fans finally kick on. As a reminder, 83-84 degrees is when NVIDIA starts hard throttling the clocks more than just from GPU Boost, so they’re dropping clocks as a result of this configuration.The 5090’s VRAM already runs hot on an open bench – 89 to 90 degrees Celsius – and that gets pushed up to peak at 100C in the Origin pre-built. This is unacceptable. Adding the GPU fan speed to the chart shows us how the Founders Edition cooler attempts to compensate by temporarily boosting fan speed to 56% during this time, which also means that Origin isn’t even benefiting as much from the noise levels as it should from the slower fans. Balancing them better would benefit noise more.As neat of a party trick as it is to have the case fans stay off unless they’re needed in the loop, Origin should have kept at least one or two running at all times, like rear exhaust, to give the GPU some help. Besides, letting the hot air linger could potentially encourage local hot spots to form on subcomponents that aren’t directly monitored, which can lead to problems.Power At The WallNow we’ll look at full system load power consumption by logging it at the wall – so everything, even efficiency losses from the PSU, is taken into account.Idle, it pulled a relatively high 125W. At the 180 second mark, the CPU load kicks in. There’s a jump at 235 seconds when the GPU load kicks in.We see a slight ramp upwards in power consumption after that, which tracks with increasing leakage as the parts heat up, before settling in at an average of 884W at steady state. AcousticsNext we’ll cover dBA over time as measured in our hemi-anechoic chamber.At idle, the fans are off, which makes for a functionally silent system at the noise floor. The first fans to come on in the system are on the GPU, bringing noise levels up to a still-quiet range of 25-28dBA at 1 meter. The loudest point is 30.5 dBA when the GPU fans briefly ramp and before system fans kick in. CPU Frequency vs. Original ReviewFor CPU frequency, fortunately for Origin, it didn’t randomly throttle it by 1GHz this time. The 9800X3D managed to stay at 5225MHz during the CPU-only load portion of torture test – the same frequency that we recorded in our original review for the CPU so that’ good. At steady state with the GPU dumping over 500W of heat into the case, the average core frequency dropped by 50MHz. If Origin made better use of its dozen or so fans, it should hold onto more of that frequency. BIOS ConfigurationBIOS for the Origin pre-built is set up sensibly, at least. The build date is January 23, which was the latest available in the time between when we ordered the system at the 50 series launch and when the system was actually assembled.Scrutinizing the chosen settings revealed nothing out of line. The DDR5-6000 memory profile was enabled and the rest of the core settings were properly set to Auto. This was all fine.Setup and SoftwareThe Windows install was normal with no bloatware. That’s also good.The desktop had a few things on it. A “Link Windows 10 Key to Microsoft Account” PDF is helpful for people who don’t know what to do if their system shows the Activate Windows watermark. Confusingly, it hasn’t been updated to say “11” instead of “10.” It also shepherds the user towards using a Microsoft account. That’s not necessarily a bad thing, but we don’t like how it makes it seem necessary because it’s not and you shouldn’t. There’s also an “Origin PC ReadMe” PDF that doesn’t offer much except coverage for Origin’s ass with disclaimers and points of contact for support. One useful thing is that it points the user to “C:\\ORIGIN PC” to find “important items.”That folder has Origin branded gifs, logos, and wallpapers, as well as CPU-Z, Teamviewer, and a Results folder. Teamviewer is almost certainly for Origin’s support teams to be able to remotely inspect the PC during support calls. It makes sense to have that stuff on there. The results folder contains an OCCT test report that shows a total of 1 hour and 52 minutes of testing. A CPU test for 12 minutes, CPU + RAM, memory, and 3D adaptive tests for 30 minutes each, then finishing with 10 minutes of OCCT’s “power” test, which is a combined full system load. It’s great that Origin actually does testing and provides this log as a baseline for future issues, and just for base expectations. This is good and gives you something to work from. Not having OCCT pre-installed to actually run again for comparison is a support oversight. It’s free for personal use at least, so the user could go download it easily.There weren’t any missing drivers in Device Manager and NVIDIA’s 572.47 driver from February 20 was the latest at the time of the build – both good things. There wasn’t any bundled bloatware installed, so points to Origin for that.iCUE itself isn’t as bad as it used to be, but it’s still clunky, like the preloaded fan profiles not showing their set points. PackagingOn to packaging.The Origin Genesis pre-built came in a massive wooden crate that was big enough for two people to move around. Considering this PC was $6,500 after taxes (at the time), we’re definitely OK with the wooden crate and its QR code opening instructions.Origin uses foam, a fabric cover, a cardboard box within a crate, and the crate for the PC. The case had two packs of expanding foam inside it, allowing the GPU to arrive undamaged and installed. The sticker on the side panel also had clear instructions. These are good things. Unfortunately, there’s a small chip in the paint on top of the case, but not as bad as the last Origin paint issues we had and we think it’s unrelated to the packaging itself.AccessoriesThe accessory kit is basic, and came inside of a box with the overused cringey adage “EAT SLEEP GAME REPEAT” printed on it. Inside are the spare PSU cables (that we’re happy to see included), an AC power cable, stock 5090 FE power adapter, standard motherboard and case accessories, a G1/4 plug tool and extra plugs, and a piece of soft tubing with a fitting on one end that can be used to help drain the cooling loop. All of this is good.Conclusion Visit our Patreon page to contribute a few dollars toward this website's operation (or consider a direct donation or buying something from our GN Store!) Additionally, when you purchase through links to retailers on our site, we may earn a small affiliate commission.During this review process, the price went even higher. You already shouldn’t buy this, but just to drive it home:Now, for the same configuration, the Genesis now costs $7,557 after the discount, off the new sticker price of $8,396. That’s an increase of over $1,000, making the premium over current DIY pricing roughly $1,700-$2,500.Now, there are good reasons for the price to go up. Tariffs have a real impact on pricing and we’re going to see it everywhere, and tariffs are also outside of Corsair’s control. We don’t fault them for that. But that doesn’t change the fact that the cost over DIY is so insanely elevated. Even Corsair’s own competitors offer better value than this, like Maingear.At $8,400 sticker price, you’d have to be drunk on whatever is discoloring Origin’s loop to buy it. Nobody should buy this, especially not for gaming. If you’re doing productivity or creative work that would seriously benefit from the 5090’s 32GB of VRAM, then look elsewhere for a better deal. This costs nearly as much as an RTX Pro 6000, which has 96GB of VRAM and is better.It would actually be cheaper to get scalped for a 5090 on Ebay and then buy the whole rest of the computer than to buy this Origin system. That’s how crazy this is.The upcharge, even assuming a 5090 price of $2,800, is just way too high versus other system integrators. Seriously, Alienware is cheaper at this point – by thousands of dollars. Alienware.We can’t recommend this PC. Ignoring the price, the memory on the video card is hitting 100 degrees C in workloads when the fans aren’t turning on because the fans are set to turn on based on the liquid temperature and the liquid doesn’t touch the GPU. For that reason alone, it gets a failing grade. For our thermal testing, pre-builts have to pass the torture test. If they don’t, they instantly fail. That’s how it always works for our pre-built reviews. This system has, unfortunately, instantly failed.

·128 مشاهدة

الرجاء تسجيل الدخول , للأعجاب والمشاركة والتعليق على هذا!
PCMag شارك رابطًا

2025-05-19 20:43:42 ·

Apple Magic Mouse vs. Logitech MX Master 3S: Should You Finally Ditch Your Dinky Magic Mouse?

Logitech MX Master 3S Wireless Mouse5.0 OutstandingBottom LineSilent buttons and an 8,000dpi sensor bring Logitech's flagship MX Master 3S Wireless Mouse just one or two clicks from perfection.Street Price365.0VSApple Magic Mouse3.0 GoodBottom LineThis 2024 update to Apple's Magic Mouse swaps the Lightning port for a USB-C one, but the rest of the design—including its most puzzling flaw—remains frozen in time.Street Price329.0

Design and Build QualityThe Apple Magic Mouse is one of the sleekest input devices you can buy. Its glossy, featureless shell evokes a futuristic feeling, a bit like magic, if you will. The aesthetic has more or less defined the mouse for the better part of a decade. While there have been minor changes over the years, like moving away from disposable batteries or—most recently—the switch to a USB-C port from the old Lightning connector, the mouse has looked largely the same since the Magic Mouse 2 was introduced in 2015. You even still charge its battery the same awkward way, by flipping it upside-down to access the USB-C port.It’s a fairly small mouse, too. In my hands, it’s a little too small, and the slight bump at the center barely offers my hand any support. The distinctive body is well-suited for both right- and left-handed users, but the hyper-glossy exterior is prone to smudging. If you're buying the Magic Mouse by itself, you can choose from two colors, though additional options are available with the purchase of an iMac, including green, orange, pink, purple, and yellow. While the Logitech MX Master 3S's curvaceous shape is strictly for right-handers, the larger size offers more real estate for larger hands. While it's not expressly intended to be an ergonomic mouse, the MX Master 3S does offer supreme comfort. A slight wing on the mouse's left flank holds your thumb perfectly, while the soft ridges and smooth, rubber-like exterior feel good in your hands. The Logitech MX Master 3S comes in three color schemes: black, pale gray, and graphite.Both mice look good paired with any Mac. But while the Apple Magic Mouse is a gorgeous mouse, as far as comfort is concerned, there are few that can compete with the Logitech MX Master 3S.Winner: Logitech MX Master 3SFeatures and CustomizationThe Logitech MX Master 3S and Apple Magic Mouse each offer unique features that power users will enjoy. The Magic Mouse supports Apple's Multi-Touch gestures. In addition to traditional right and left clicks, you can perform certain other gestures by dragging your fingers across the mouse. For example, you can drag your finger down the center of the mouse to scroll, or swipe two fingers across the mouse to switch between apps. One thing to note: Apple says the latest Magic Mouse requires a Mac with macOS 15.1 or later, or an iPad with iPadOS 18.1 or later. You can use the mouse with Bluetooth-enabled Macs running earlier OS versions, but functionality will be limited.The Logitech MX Master 3S, on the other hand, comes with additional buttons and a horizontal scroll wheel. The physical MagSpeed Electromagnetic scroll wheel is fast, precise, and nearly silent. A small button sits just under the scroll wheel and cycles between sensor resolution presets up to 8,000 dots per inch. The Magic Mouse does not have an adjustable sensor resolution, and Apple does not list its maximum DPI. Recommended by Our EditorsAs far as software is concerned, the MX Master 3S supports Logi Options+, Logitech’s free, proprietary software that offers tons of additional functionality. Logitech Flow allows you to seamlessly work between two Windows or macOS computers, and even transfer text, images, and files between them. Logitech Smart Actions is a macro builder designed to automate repetitive tasks, and the Logi AI Prompt Builder allows users to build prompts and queries for ChatGPT. The Apple Magic Mouse can’t do any of that, and is obviously only compatible with Apple devices, though it does have a few basic customizable options accessed via the macOS System Preferences app. Winner: Logitech MX Master 3SBattery Life and Wireless ConnectivityBattery life is another important factor in choosing a wireless mouse. The Logitech MX Master 3S offers 70 days of power on a full charge and gets three hours of use from a one-minute quick charge. The MX Master 3S also saves up to three Bluetooth connections, which can be swapped with a press of a button located underneath the mouse. Or, you can use the Logi Bolt USB dongle, which promises a drop-off-free connection up to 33 feet away.The Apple Magic Mouse, on the other hand, offers “about a month or more” of power. It doesn't have a USB dongle, with Bluetooth as its sole connection option. You also can't charge it while you're using it, while the MX Master 3S has a USB-C port for charging right at the front of the mouse.Winner: Logitech MX Master 3SPriceChances are, you want to get the most out of whatever money you've got leftover to spend on peripherals after buying a Mac, and if you're buying an iMac, you'll be tempted to not shell out any additional money at all, since your purchase already includes a mouse. In fact, neither the Apple nor the Logitech mouse is budget-friendly. The Logitech MX Master 3S originally retailed for when we first reviewed it back in 2022. However, it has crept up in price since then, now retailing for For that sum, you’ll get a USB-C charging cordand the Logitech Bolt USB receiver in addition to the mouse itself.The Apple Magic Mouse, on the other hand, retails for and includes a USB-Ccharging cable—if you get the white mouse. For some reason, Apple charges an additional for the black option. Apple products are typically expensive endeavors, but in this case, the Magic Mouse is actually the most economical option.Winner: Apple Magic Mouse
#apple #magic #mouse #logitech #master

Apple Magic Mouse vs. Logitech MX Master 3S: Should You Finally Ditch Your Dinky Magic Mouse?
Logitech MX Master 3S Wireless Mouse5.0 OutstandingBottom LineSilent buttons and an 8,000dpi sensor bring Logitech's flagship MX Master 3S Wireless Mouse just one or two clicks from perfection.Street Price365.0VSApple Magic Mouse3.0 GoodBottom LineThis 2024 update to Apple's Magic Mouse swaps the Lightning port for a USB-C one, but the rest of the design—including its most puzzling flaw—remains frozen in time.Street Price329.0 Design and Build QualityThe Apple Magic Mouse is one of the sleekest input devices you can buy. Its glossy, featureless shell evokes a futuristic feeling, a bit like magic, if you will. The aesthetic has more or less defined the mouse for the better part of a decade. While there have been minor changes over the years, like moving away from disposable batteries or—most recently—the switch to a USB-C port from the old Lightning connector, the mouse has looked largely the same since the Magic Mouse 2 was introduced in 2015. You even still charge its battery the same awkward way, by flipping it upside-down to access the USB-C port.It’s a fairly small mouse, too. In my hands, it’s a little too small, and the slight bump at the center barely offers my hand any support. The distinctive body is well-suited for both right- and left-handed users, but the hyper-glossy exterior is prone to smudging. If you're buying the Magic Mouse by itself, you can choose from two colors, though additional options are available with the purchase of an iMac, including green, orange, pink, purple, and yellow. While the Logitech MX Master 3S's curvaceous shape is strictly for right-handers, the larger size offers more real estate for larger hands. While it's not expressly intended to be an ergonomic mouse, the MX Master 3S does offer supreme comfort. A slight wing on the mouse's left flank holds your thumb perfectly, while the soft ridges and smooth, rubber-like exterior feel good in your hands. The Logitech MX Master 3S comes in three color schemes: black, pale gray, and graphite.Both mice look good paired with any Mac. But while the Apple Magic Mouse is a gorgeous mouse, as far as comfort is concerned, there are few that can compete with the Logitech MX Master 3S.Winner: Logitech MX Master 3SFeatures and CustomizationThe Logitech MX Master 3S and Apple Magic Mouse each offer unique features that power users will enjoy. The Magic Mouse supports Apple's Multi-Touch gestures. In addition to traditional right and left clicks, you can perform certain other gestures by dragging your fingers across the mouse. For example, you can drag your finger down the center of the mouse to scroll, or swipe two fingers across the mouse to switch between apps. One thing to note: Apple says the latest Magic Mouse requires a Mac with macOS 15.1 or later, or an iPad with iPadOS 18.1 or later. You can use the mouse with Bluetooth-enabled Macs running earlier OS versions, but functionality will be limited.The Logitech MX Master 3S, on the other hand, comes with additional buttons and a horizontal scroll wheel. The physical MagSpeed Electromagnetic scroll wheel is fast, precise, and nearly silent. A small button sits just under the scroll wheel and cycles between sensor resolution presets up to 8,000 dots per inch. The Magic Mouse does not have an adjustable sensor resolution, and Apple does not list its maximum DPI. Recommended by Our EditorsAs far as software is concerned, the MX Master 3S supports Logi Options+, Logitech’s free, proprietary software that offers tons of additional functionality. Logitech Flow allows you to seamlessly work between two Windows or macOS computers, and even transfer text, images, and files between them. Logitech Smart Actions is a macro builder designed to automate repetitive tasks, and the Logi AI Prompt Builder allows users to build prompts and queries for ChatGPT. The Apple Magic Mouse can’t do any of that, and is obviously only compatible with Apple devices, though it does have a few basic customizable options accessed via the macOS System Preferences app. Winner: Logitech MX Master 3SBattery Life and Wireless ConnectivityBattery life is another important factor in choosing a wireless mouse. The Logitech MX Master 3S offers 70 days of power on a full charge and gets three hours of use from a one-minute quick charge. The MX Master 3S also saves up to three Bluetooth connections, which can be swapped with a press of a button located underneath the mouse. Or, you can use the Logi Bolt USB dongle, which promises a drop-off-free connection up to 33 feet away.The Apple Magic Mouse, on the other hand, offers “about a month or more” of power. It doesn't have a USB dongle, with Bluetooth as its sole connection option. You also can't charge it while you're using it, while the MX Master 3S has a USB-C port for charging right at the front of the mouse.Winner: Logitech MX Master 3SPriceChances are, you want to get the most out of whatever money you've got leftover to spend on peripherals after buying a Mac, and if you're buying an iMac, you'll be tempted to not shell out any additional money at all, since your purchase already includes a mouse. In fact, neither the Apple nor the Logitech mouse is budget-friendly. The Logitech MX Master 3S originally retailed for when we first reviewed it back in 2022. However, it has crept up in price since then, now retailing for For that sum, you’ll get a USB-C charging cordand the Logitech Bolt USB receiver in addition to the mouse itself.The Apple Magic Mouse, on the other hand, retails for and includes a USB-Ccharging cable—if you get the white mouse. For some reason, Apple charges an additional for the black option. Apple products are typically expensive endeavors, but in this case, the Magic Mouse is actually the most economical option.Winner: Apple Magic Mouse #apple #magic #mouse #logitech #master

ME.PCMAG.COM

Apple Magic Mouse vs. Logitech MX Master 3S: Should You Finally Ditch Your Dinky Magic Mouse?

Logitech MX Master 3S Wireless Mouse5.0 OutstandingBottom LineSilent buttons and an 8,000dpi sensor bring Logitech's flagship MX Master 3S Wireless Mouse just one or two clicks from perfection.Street Price365.0VSApple Magic Mouse (USB-C)3.0 GoodBottom LineThis 2024 update to Apple's Magic Mouse swaps the Lightning port for a USB-C one, but the rest of the design—including its most puzzling flaw—remains frozen in time.Street Price329.0 Design and Build QualityThe Apple Magic Mouse is one of the sleekest input devices you can buy. Its glossy, featureless shell evokes a futuristic feeling, a bit like magic, if you will. The aesthetic has more or less defined the mouse for the better part of a decade. While there have been minor changes over the years, like moving away from disposable batteries or—most recently—the switch to a USB-C port from the old Lightning connector, the mouse has looked largely the same since the Magic Mouse 2 was introduced in 2015. You even still charge its battery the same awkward way, by flipping it upside-down to access the USB-C port.(Credit: Zackery Cuevas)It’s a fairly small mouse, too. In my hands, it’s a little too small, and the slight bump at the center barely offers my hand any support. The distinctive body is well-suited for both right- and left-handed users, but the hyper-glossy exterior is prone to smudging (though it can be easily wiped away with a microfiber cloth). If you're buying the Magic Mouse by itself, you can choose from two colors (black and white), though additional options are available with the purchase of an iMac, including green, orange, pink, purple, and yellow. While the Logitech MX Master 3S's curvaceous shape is strictly for right-handers, the larger size offers more real estate for larger hands. While it's not expressly intended to be an ergonomic mouse, the MX Master 3S does offer supreme comfort. A slight wing on the mouse's left flank holds your thumb perfectly, while the soft ridges and smooth, rubber-like exterior feel good in your hands. The Logitech MX Master 3S comes in three color schemes: black, pale gray, and graphite.Both mice look good paired with any Mac. But while the Apple Magic Mouse is a gorgeous mouse, as far as comfort is concerned, there are few that can compete with the Logitech MX Master 3S.Winner: Logitech MX Master 3SFeatures and CustomizationThe Logitech MX Master 3S and Apple Magic Mouse each offer unique features that power users will enjoy. The Magic Mouse supports Apple's Multi-Touch gestures. In addition to traditional right and left clicks, you can perform certain other gestures by dragging your fingers across the mouse. For example, you can drag your finger down the center of the mouse to scroll, or swipe two fingers across the mouse to switch between apps. One thing to note: Apple says the latest Magic Mouse requires a Mac with macOS 15.1 or later, or an iPad with iPadOS 18.1 or later. You can use the mouse with Bluetooth-enabled Macs running earlier OS versions, but functionality will be limited.(Credit: Molly Flores)The Logitech MX Master 3S, on the other hand, comes with additional buttons and a horizontal scroll wheel. The physical MagSpeed Electromagnetic scroll wheel is fast, precise, and nearly silent. A small button sits just under the scroll wheel and cycles between sensor resolution presets up to 8,000 dots per inch (DPI). The Magic Mouse does not have an adjustable sensor resolution, and Apple does not list its maximum DPI. Recommended by Our EditorsAs far as software is concerned, the MX Master 3S supports Logi Options+, Logitech’s free, proprietary software that offers tons of additional functionality. Logitech Flow allows you to seamlessly work between two Windows or macOS computers, and even transfer text, images, and files between them. Logitech Smart Actions is a macro builder designed to automate repetitive tasks, and the Logi AI Prompt Builder allows users to build prompts and queries for ChatGPT. The Apple Magic Mouse can’t do any of that, and is obviously only compatible with Apple devices, though it does have a few basic customizable options accessed via the macOS System Preferences app. Winner: Logitech MX Master 3SBattery Life and Wireless ConnectivityBattery life is another important factor in choosing a wireless mouse. The Logitech MX Master 3S offers 70 days of power on a full charge and gets three hours of use from a one-minute quick charge. The MX Master 3S also saves up to three Bluetooth connections, which can be swapped with a press of a button located underneath the mouse. Or, you can use the Logi Bolt USB dongle, which promises a drop-off-free connection up to 33 feet away. (Credit: Zackery Cuevas)The Apple Magic Mouse, on the other hand, offers “about a month or more” of power. It doesn't have a USB dongle, with Bluetooth as its sole connection option. You also can't charge it while you're using it (you know, because it needs to be upside-down to charge), while the MX Master 3S has a USB-C port for charging right at the front of the mouse.Winner: Logitech MX Master 3SPriceChances are, you want to get the most out of whatever money you've got leftover to spend on peripherals after buying a Mac, and if you're buying an iMac, you'll be tempted to not shell out any additional money at all, since your purchase already includes a mouse. In fact, neither the Apple nor the Logitech mouse is budget-friendly. The Logitech MX Master 3S originally retailed for $99 when we first reviewed it back in 2022. However, it has crept up in price since then, now retailing for $119. For that sum, you’ll get a USB-C charging cord (USB-A-to-C) and the Logitech Bolt USB receiver in addition to the mouse itself. (Credit: Molly Flores)The Apple Magic Mouse, on the other hand, retails for $79.99 and includes a USB-C (USB-C-to-C) charging cable—if you get the white mouse. For some reason, Apple charges an additional $20 for the black option. Apple products are typically expensive endeavors, but in this case, the Magic Mouse is actually the most economical option.Winner: Apple Magic Mouse (USB-C)

·100 مشاهدة

الرجاء تسجيل الدخول , للأعجاب والمشاركة والتعليق على هذا!
Unity شارك رابطًا

2025-05-19 20:38:59 ·

Everything you need to know about Memory Profiler 1.0.0

The new Memory Profiler – version 1.0.0 – delivers improvements to the profiling suite that offer a more complete view of memory usage and make it easier and faster to optimize performance or detect memory issues.We in the profiler team are very happy to announce that, as of September 2022, the Memory Profiler package moved from Experimental to Verified and is now accessible for everyone using the Unity 2022.2 Tech Stream.Optimizing memory is crucial for applications that run on devices with limited capabilities in order to prevent them from crashing out of memory. Moreover, if you’re planning to ship on multiple devices, your memory footprint might be in need of fine tuning to make sure you get the best out of each platform.The Memory Profiler is a tool that helps you with these challenges by allowing you to take a snapshot of your game and review its memory usage at a specific moment in time.By taking a “snapshot” of your game, you can see what is occupying most of its memory or identify problems that might cause applications to crash.Note: While Memory Profiler’s complete features and memory tracking are only available for 2022.2 onwards, you can take advantage of the new user interfaceto inspect older captures by opening Memory Profiler in an empty 2022+ project and importing older snapshots.Memory Profiler was originally released as an experimental package a couple of years ago, but this release includes a major overhaul of the UI and user experienceto make it faster and easier to use.The original experimental package was used by some early adopters, and helped in many cases to identify memory issues. However, it was unwieldy, slow, and required expert knowledge in its first iteration.When we set up to work on a new version of the package, we had the following intentions in mind:Simplify the most common and important workflows, so that you would be able to identify and fix issues fasterProvide more complete and reliable information to make it easier to compare memory usage across devicesShowcase information in a more accessible way, so that users without in-depth knowledge of memory can still effectively optimize applicationsTo achieve this, we’ve been working closely with the early adopters of the experimental package in order to understand their needs and pain points, and we have iterated several times on the design to make sure it provides the best UX possible.Thanks to their input, our team was able to greatly streamline and simplify the tool, reducing jargon and making it easier to inspect objects, details, and dependencies directly in the tool.Memory Profiler consists of three main sections that showcase its new and improved UI, which we’ll break down here:Snapshot listA Main section, which is itself divided into three workflow viewsSelection DetailsThe Snapshot list is the area where you can capture and see snapshots that you have taken in your game, as well as high-level information about memory usage and date of capture. Here, you can capture snapshots either with the “Capture” button in the toolbar or – if no snapshot is available – through the snapshot button in the Main section.You can also see whether memory usage is changing across snapshots and can get a sense of whether project performance is improving or not.Finally, you can also use Snapshot list to select whether you want to look at single snapshots or compared snapshots.This is the core area of Memory Profiler, divided into three workflow views: Summary, Unity Objects, and All of Memory. Each of these views is accessible from the tabs in the top of the Main section.The Summary view is the default view that opens as you load or capture a snapshot. It provides information on how much memory you are using, how much is “resident” on the device, and how much is committed but not currently on device. It also provides information about how memory is distributed across categories, to simplify choosing where to start your investigation.Finally, Summary view provides entry points for more detailed analysis, such as the “Top Unity Objects in Memory” or the “Managed memory breakdown.”The Unity Objects view is where you will likely spend the most time. It lists all the main types of objects loaded in memory, such as textures, shaders, fonts, and so on. Usually, this is where you can identify objects that are too big, don’t need to be there because they have been loaded mistakenly, or were kept in memory by some hanging reference.The All of Memory view enables you to see all memory, divided by category – Native, Managed, Graphics, or Executables. You can see the data captured by the Memory Profiler and investigate more specific aspects to see what is composing your project’s memory.Selection details offers more information about items selected in the Main section, including descriptions of different categories of memory and potential references or details of selected objects.This section offers more information to help you understand what you are looking at, and makes it easier to determine why it is in memory. Moreover, if you have the project open in the Editor, you will have the possibility to “ping” objects and inspect assets directly in the Scene or Project window.Each of the views described above is available for inspecting single snapshots or for comparing them.If you want to compare a snapshot, you can select “Compare” in the snapshot list, then select a second snapshot from your list. Being in the compare tab of the snapshot list will automatically turn all views into Compare mode.In Compare mode, the Summary view provides the two captures’ memory breakdown, side by side, so you can see the main differences in memory usage between the snapshots.In the Unity Objects and All of Memory views, there is a new, dedicated UI that allows you to see how different memory categories have changed in size or the amount of objects contained in them.If you select a category in the top table, you will be able to inspect individual differences for snapshot A and B in the tables below.If you want to go back to inspecting a single snapshot, you can select “Single” again in the snapshot list. The second snapshot will stay latently loaded so you can quickly switch between the two, as needed.In the coming months, we hope to provide you with further resources and information on how you can effectively optimize memory. Stay tuned for more blog posts covering how to identify common problems in memory, as well as a deep dive into how memory works in Unity.For more general resources on how to optimize application performance, we recommend the following:Memory in Unity manualThe Ultimate guide to profiling Unity gamese-bookOptimize your console and PC game performancee-bookOptimize your mobile game performance e-bookThis redesign of the Memory Profiler integrates input from many early adopters of the experimental version of this package. While this version brings major improvements in both the reliability and usability of this tool, our work is far from done.We are also currently working on better understanding workflows related to reducing fragmentation, so that we might be able to help you with this aspect in the future. At the same time, we are gathering feedback on usability and UX issues that are still open and need improvement.If you have any feedback or suggestions, we invite you to share your thoughts through our roadmap page, where you can vote for existing features in progress or suggest new ones. You can also reach out to ask questions in the forums.We’re constantly monitoring both channels and look forward to working together to further improve Unity’s performance and tooling.Install or learn more about the Memory Profiler 1.0.0 package.
#everything #you #need #know #about

Everything you need to know about Memory Profiler 1.0.0
The new Memory Profiler – version 1.0.0 – delivers improvements to the profiling suite that offer a more complete view of memory usage and make it easier and faster to optimize performance or detect memory issues.We in the profiler team are very happy to announce that, as of September 2022, the Memory Profiler package moved from Experimental to Verified and is now accessible for everyone using the Unity 2022.2 Tech Stream.Optimizing memory is crucial for applications that run on devices with limited capabilities in order to prevent them from crashing out of memory. Moreover, if you’re planning to ship on multiple devices, your memory footprint might be in need of fine tuning to make sure you get the best out of each platform.The Memory Profiler is a tool that helps you with these challenges by allowing you to take a snapshot of your game and review its memory usage at a specific moment in time.By taking a “snapshot” of your game, you can see what is occupying most of its memory or identify problems that might cause applications to crash.Note: While Memory Profiler’s complete features and memory tracking are only available for 2022.2 onwards, you can take advantage of the new user interfaceto inspect older captures by opening Memory Profiler in an empty 2022+ project and importing older snapshots.Memory Profiler was originally released as an experimental package a couple of years ago, but this release includes a major overhaul of the UI and user experienceto make it faster and easier to use.The original experimental package was used by some early adopters, and helped in many cases to identify memory issues. However, it was unwieldy, slow, and required expert knowledge in its first iteration.When we set up to work on a new version of the package, we had the following intentions in mind:Simplify the most common and important workflows, so that you would be able to identify and fix issues fasterProvide more complete and reliable information to make it easier to compare memory usage across devicesShowcase information in a more accessible way, so that users without in-depth knowledge of memory can still effectively optimize applicationsTo achieve this, we’ve been working closely with the early adopters of the experimental package in order to understand their needs and pain points, and we have iterated several times on the design to make sure it provides the best UX possible.Thanks to their input, our team was able to greatly streamline and simplify the tool, reducing jargon and making it easier to inspect objects, details, and dependencies directly in the tool.Memory Profiler consists of three main sections that showcase its new and improved UI, which we’ll break down here:Snapshot listA Main section, which is itself divided into three workflow viewsSelection DetailsThe Snapshot list is the area where you can capture and see snapshots that you have taken in your game, as well as high-level information about memory usage and date of capture. Here, you can capture snapshots either with the “Capture” button in the toolbar or – if no snapshot is available – through the snapshot button in the Main section.You can also see whether memory usage is changing across snapshots and can get a sense of whether project performance is improving or not.Finally, you can also use Snapshot list to select whether you want to look at single snapshots or compared snapshots.This is the core area of Memory Profiler, divided into three workflow views: Summary, Unity Objects, and All of Memory. Each of these views is accessible from the tabs in the top of the Main section.The Summary view is the default view that opens as you load or capture a snapshot. It provides information on how much memory you are using, how much is “resident” on the device, and how much is committed but not currently on device. It also provides information about how memory is distributed across categories, to simplify choosing where to start your investigation.Finally, Summary view provides entry points for more detailed analysis, such as the “Top Unity Objects in Memory” or the “Managed memory breakdown.”The Unity Objects view is where you will likely spend the most time. It lists all the main types of objects loaded in memory, such as textures, shaders, fonts, and so on. Usually, this is where you can identify objects that are too big, don’t need to be there because they have been loaded mistakenly, or were kept in memory by some hanging reference.The All of Memory view enables you to see all memory, divided by category – Native, Managed, Graphics, or Executables. You can see the data captured by the Memory Profiler and investigate more specific aspects to see what is composing your project’s memory.Selection details offers more information about items selected in the Main section, including descriptions of different categories of memory and potential references or details of selected objects.This section offers more information to help you understand what you are looking at, and makes it easier to determine why it is in memory. Moreover, if you have the project open in the Editor, you will have the possibility to “ping” objects and inspect assets directly in the Scene or Project window.Each of the views described above is available for inspecting single snapshots or for comparing them.If you want to compare a snapshot, you can select “Compare” in the snapshot list, then select a second snapshot from your list. Being in the compare tab of the snapshot list will automatically turn all views into Compare mode.In Compare mode, the Summary view provides the two captures’ memory breakdown, side by side, so you can see the main differences in memory usage between the snapshots.In the Unity Objects and All of Memory views, there is a new, dedicated UI that allows you to see how different memory categories have changed in size or the amount of objects contained in them.If you select a category in the top table, you will be able to inspect individual differences for snapshot A and B in the tables below.If you want to go back to inspecting a single snapshot, you can select “Single” again in the snapshot list. The second snapshot will stay latently loaded so you can quickly switch between the two, as needed.In the coming months, we hope to provide you with further resources and information on how you can effectively optimize memory. Stay tuned for more blog posts covering how to identify common problems in memory, as well as a deep dive into how memory works in Unity.For more general resources on how to optimize application performance, we recommend the following:Memory in Unity manualThe Ultimate guide to profiling Unity gamese-bookOptimize your console and PC game performancee-bookOptimize your mobile game performance e-bookThis redesign of the Memory Profiler integrates input from many early adopters of the experimental version of this package. While this version brings major improvements in both the reliability and usability of this tool, our work is far from done.We are also currently working on better understanding workflows related to reducing fragmentation, so that we might be able to help you with this aspect in the future. At the same time, we are gathering feedback on usability and UX issues that are still open and need improvement.If you have any feedback or suggestions, we invite you to share your thoughts through our roadmap page, where you can vote for existing features in progress or suggest new ones. You can also reach out to ask questions in the forums.We’re constantly monitoring both channels and look forward to working together to further improve Unity’s performance and tooling.Install or learn more about the Memory Profiler 1.0.0 package. #everything #you #need #know #about

UNITY.COM

Everything you need to know about Memory Profiler 1.0.0

The new Memory Profiler – version 1.0.0 – delivers improvements to the profiling suite that offer a more complete view of memory usage and make it easier and faster to optimize performance or detect memory issues.We in the profiler team are very happy to announce that, as of September 2022, the Memory Profiler package moved from Experimental to Verified and is now accessible for everyone using the Unity 2022.2 Tech Stream.Optimizing memory is crucial for applications that run on devices with limited capabilities in order to prevent them from crashing out of memory. Moreover, if you’re planning to ship on multiple devices, your memory footprint might be in need of fine tuning to make sure you get the best out of each platform.The Memory Profiler is a tool that helps you with these challenges by allowing you to take a snapshot of your game and review its memory usage at a specific moment in time.By taking a “snapshot” of your game, you can see what is occupying most of its memory or identify problems that might cause applications to crash.Note: While Memory Profiler’s complete features and memory tracking are only available for 2022.2 onwards, you can take advantage of the new user interface (UI) to inspect older captures by opening Memory Profiler in an empty 2022+ project and importing older snapshots.Memory Profiler was originally released as an experimental package a couple of years ago, but this release includes a major overhaul of the UI and user experience (UX) to make it faster and easier to use.The original experimental package was used by some early adopters, and helped in many cases to identify memory issues. However, it was unwieldy, slow, and required expert knowledge in its first iteration.When we set up to work on a new version of the package, we had the following intentions in mind:Simplify the most common and important workflows, so that you would be able to identify and fix issues fasterProvide more complete and reliable information to make it easier to compare memory usage across devices (also using platform-specific profilers)Showcase information in a more accessible way, so that users without in-depth knowledge of memory can still effectively optimize applicationsTo achieve this, we’ve been working closely with the early adopters of the experimental package in order to understand their needs and pain points, and we have iterated several times on the design to make sure it provides the best UX possible.Thanks to their input, our team was able to greatly streamline and simplify the tool, reducing jargon and making it easier to inspect objects, details, and dependencies directly in the tool.Memory Profiler consists of three main sections that showcase its new and improved UI, which we’ll break down here:Snapshot listA Main section, which is itself divided into three workflow views (Summary, Unity Objects, and All of Memory)Selection DetailsThe Snapshot list is the area where you can capture and see snapshots that you have taken in your game, as well as high-level information about memory usage and date of capture. Here, you can capture snapshots either with the “Capture” button in the toolbar or – if no snapshot is available – through the snapshot button in the Main section.You can also see whether memory usage is changing across snapshots and can get a sense of whether project performance is improving or not.Finally, you can also use Snapshot list to select whether you want to look at single snapshots or compared snapshots (more information about this below).This is the core area of Memory Profiler, divided into three workflow views: Summary, Unity Objects, and All of Memory. Each of these views is accessible from the tabs in the top of the Main section.The Summary view is the default view that opens as you load or capture a snapshot. It provides information on how much memory you are using, how much is “resident” on the device, and how much is committed but not currently on device. It also provides information about how memory is distributed across categories, to simplify choosing where to start your investigation.Finally, Summary view provides entry points for more detailed analysis, such as the “Top Unity Objects in Memory” or the “Managed memory breakdown.”The Unity Objects view is where you will likely spend the most time. It lists all the main types of objects loaded in memory, such as textures, shaders, fonts, and so on. Usually, this is where you can identify objects that are too big, don’t need to be there because they have been loaded mistakenly, or were kept in memory by some hanging reference.The All of Memory view enables you to see all memory, divided by category – Native, Managed, Graphics, or Executables. You can see the data captured by the Memory Profiler and investigate more specific aspects to see what is composing your project’s memory.Selection details offers more information about items selected in the Main section, including descriptions of different categories of memory and potential references or details of selected objects.This section offers more information to help you understand what you are looking at, and makes it easier to determine why it is in memory. Moreover, if you have the project open in the Editor, you will have the possibility to “ping” objects and inspect assets directly in the Scene or Project window.Each of the views described above is available for inspecting single snapshots or for comparing them.If you want to compare a snapshot, you can select “Compare” in the snapshot list, then select a second snapshot from your list. Being in the compare tab of the snapshot list will automatically turn all views into Compare mode.In Compare mode, the Summary view provides the two captures’ memory breakdown, side by side, so you can see the main differences in memory usage between the snapshots (which will be named “A” and “B”).In the Unity Objects and All of Memory views, there is a new, dedicated UI that allows you to see how different memory categories have changed in size or the amount of objects contained in them.If you select a category in the top table, you will be able to inspect individual differences for snapshot A and B in the tables below.If you want to go back to inspecting a single snapshot, you can select “Single” again in the snapshot list. The second snapshot will stay latently loaded so you can quickly switch between the two, as needed.In the coming months, we hope to provide you with further resources and information on how you can effectively optimize memory. Stay tuned for more blog posts covering how to identify common problems in memory, as well as a deep dive into how memory works in Unity.For more general resources on how to optimize application performance, we recommend the following:Memory in Unity manualThe Ultimate guide to profiling Unity gamese-bookOptimize your console and PC game performancee-bookOptimize your mobile game performance e-bookThis redesign of the Memory Profiler integrates input from many early adopters of the experimental version of this package. While this version brings major improvements in both the reliability and usability of this tool, our work is far from done.We are also currently working on better understanding workflows related to reducing fragmentation, so that we might be able to help you with this aspect in the future. At the same time, we are gathering feedback on usability and UX issues that are still open and need improvement.If you have any feedback or suggestions, we invite you to share your thoughts through our roadmap page, where you can vote for existing features in progress or suggest new ones. You can also reach out to ask questions in the forums.We’re constantly monitoring both channels and look forward to working together to further improve Unity’s performance and tooling.Install or learn more about the Memory Profiler 1.0.0 package.

·120 مشاهدة

الرجاء تسجيل الدخول , للأعجاب والمشاركة والتعليق على هذا!
NVIDIA شارك رابطًا

2025-05-19 20:06:40 ·

@ChaosGroup Enscape & @Autodesk VRED just leveled up with DLSS 4! Enjoy sharper visuals + faster performance for real-time design & 3D visuali...

@ChaosGroup Enscape & @Autodesk VRED just leveled up with DLSS 4! Enjoy sharper visuals + faster performance for real-time design & 3D visualization. Learn more:
#chaosgroup #enscape #ampamp #autodesk #vred

⚡ @ChaosGroup Enscape & @Autodesk VRED just leveled up with DLSS 4! 🚀 Enjoy sharper visuals + faster performance for real-time design & 3D visuali...
⚡ @ChaosGroup Enscape & @Autodesk VRED just leveled up with DLSS 4! 🚀Enjoy sharper visuals + faster performance for real-time design & 3D visualization. 🔥Learn more: #chaosgroup #enscape #ampamp #autodesk #vred

X.COM

⚡ @ChaosGroup Enscape & @Autodesk VRED just leveled up with DLSS 4! 🚀 Enjoy sharper visuals + faster performance for real-time design & 3D visuali...

⚡ @ChaosGroup Enscape & @Autodesk VRED just leveled up with DLSS 4! 🚀Enjoy sharper visuals + faster performance for real-time design & 3D visualization. 🔥Learn more: https://www.nvidia.com/en-us/geforce/news/125-dlss-4-multi-frame-gen-games-more-announced-computex-2025/#:~:text=Chaos%20Enscape%20%26%20Autodesk,NVIDIA%20Studio

·39 مشاهدة

الرجاء تسجيل الدخول , للأعجاب والمشاركة والتعليق على هذا!
Business Insider شارك رابطًا

2025-05-19 19:36:19 ·

Jamie Dimon opens the door to bitcoin, warns against stagflation in wide-ranging remarks to investors

Jamie Dimon

Tom Williams/CQ-Roll Call, Inc via Getty Images

2025-05-19T19:21:31Z

d

Read in app

This story is available exclusively to Business Insider
subscribers. Become an Insider
and start reading now.
Have an account?

JPMorgan CEO Jamie Dimon addressed a range of topics at the firm's Investor Day meeting on Monday.
He said the bank will allow investors to buy bitcoin, while warning against stagflation.
He sounded dour on the economy but hopeful on the potential for a regulatory reset.

Jamie Dimon isn't a fan of bitcoin, but he plans to start offering it to clients of JPMorgan Chase nonetheless."We are going to allow you to buy it," Dimon said at the bank's annual presentation for investors on Monday. "We're not going to custody it. We're going to put it in statements for clients.""I don't think you should smoke. But I defend your right to smoke," he said in explaining his position. The bitcoin comments came as the JPMorgan CEO, often considered Wall Street's elder statesman, took the stage to answer questions from investors and research analysts. In the roughly 40-minute session, he touched on a range of topics, from the economy to what he expects from Trump's regulators. Dimon sounded a dour note on the economy, saying he thinks the risk of stagflation is "two times" higher than many think, and making dire predictions on credit as an investment class. "I think the worst one for a bank and for most companies is stagflation," he continued, warning: "I think the odds of that are probably two times what the market thinks."He also said the bank had lost some commercial opportunities as a result of Trump's trade war. "We've lost business because of that," he said in response to an analyst's question.He sounded upbeat, however, when it came to the president's regulatory agenda. "I think that the Secretary of Treasury, the president of the United States, the new head of the OCC, the new head of the CFPB, Michelle Bowman at Federal Reserve, and the SEC have all made it clear that they want to fix some of the things they think are broken," he said. "I think they'll accomplish some of that. Some will take longer than others, but they all want to do it."He called on regulators to consider lightening regulations for publicly traded companies, which he said have been halved since the 1990s, from 8,000 to 4,000. "We're driving companies out of the public marketplace because of expensive reporting, litigation, cookie-cutter approaches to boards, compensation, and litigation," he said. "I would love to be a private company," he added.Dimon also raised questions about the rapid expansion of investments in credit, including through funds raised to make nonbank loans, or private credit. "I don't like making forecasts," Dimon said, "but I am not a buyer of credit today. I think credit today is a bad risk," he said, adding, "I think that people who haven't been through major downturns are missing the point about what can happen in credit." As interest rates rise and economic conditions soften, the risk of credit defaults rises, sometimes leaving borrowers strapped for cash and lenders struggling to recoup capital.Earlier in the day, Marianne Lake, JPMorgan's CEO of consumer and community banking, said the firm was "closely monitoring the whole ecosystem" of lending but not giving up in spite of warning signs. "The environment is very challenging for home lending and auto," she said, adding, "but we remain committed."Have a tip? Contact this reporter via email at ralexander@businessinsider.com or SMS/Signal at 561-247-5758. Use a personal email address and a nonwork device; here's our guide to sharing information securely.

Recommended video
#jamie #dimon #opens #door #bitcoin

Jamie Dimon opens the door to bitcoin, warns against stagflation in wide-ranging remarks to investors
Jamie Dimon Tom Williams/CQ-Roll Call, Inc via Getty Images 2025-05-19T19:21:31Z d Read in app This story is available exclusively to Business Insider subscribers. Become an Insider and start reading now. Have an account? JPMorgan CEO Jamie Dimon addressed a range of topics at the firm's Investor Day meeting on Monday. He said the bank will allow investors to buy bitcoin, while warning against stagflation. He sounded dour on the economy but hopeful on the potential for a regulatory reset. Jamie Dimon isn't a fan of bitcoin, but he plans to start offering it to clients of JPMorgan Chase nonetheless."We are going to allow you to buy it," Dimon said at the bank's annual presentation for investors on Monday. "We're not going to custody it. We're going to put it in statements for clients.""I don't think you should smoke. But I defend your right to smoke," he said in explaining his position. The bitcoin comments came as the JPMorgan CEO, often considered Wall Street's elder statesman, took the stage to answer questions from investors and research analysts. In the roughly 40-minute session, he touched on a range of topics, from the economy to what he expects from Trump's regulators. Dimon sounded a dour note on the economy, saying he thinks the risk of stagflation is "two times" higher than many think, and making dire predictions on credit as an investment class. "I think the worst one for a bank and for most companies is stagflation," he continued, warning: "I think the odds of that are probably two times what the market thinks."He also said the bank had lost some commercial opportunities as a result of Trump's trade war. "We've lost business because of that," he said in response to an analyst's question.He sounded upbeat, however, when it came to the president's regulatory agenda. "I think that the Secretary of Treasury, the president of the United States, the new head of the OCC, the new head of the CFPB, Michelle Bowman at Federal Reserve, and the SEC have all made it clear that they want to fix some of the things they think are broken," he said. "I think they'll accomplish some of that. Some will take longer than others, but they all want to do it."He called on regulators to consider lightening regulations for publicly traded companies, which he said have been halved since the 1990s, from 8,000 to 4,000. "We're driving companies out of the public marketplace because of expensive reporting, litigation, cookie-cutter approaches to boards, compensation, and litigation," he said. "I would love to be a private company," he added.Dimon also raised questions about the rapid expansion of investments in credit, including through funds raised to make nonbank loans, or private credit. "I don't like making forecasts," Dimon said, "but I am not a buyer of credit today. I think credit today is a bad risk," he said, adding, "I think that people who haven't been through major downturns are missing the point about what can happen in credit." As interest rates rise and economic conditions soften, the risk of credit defaults rises, sometimes leaving borrowers strapped for cash and lenders struggling to recoup capital.Earlier in the day, Marianne Lake, JPMorgan's CEO of consumer and community banking, said the firm was "closely monitoring the whole ecosystem" of lending but not giving up in spite of warning signs. "The environment is very challenging for home lending and auto," she said, adding, "but we remain committed."Have a tip? Contact this reporter via email at ralexander@businessinsider.com or SMS/Signal at 561-247-5758. Use a personal email address and a nonwork device; here's our guide to sharing information securely. Recommended video #jamie #dimon #opens #door #bitcoin

WWW.BUSINESSINSIDER.COM

Jamie Dimon opens the door to bitcoin, warns against stagflation in wide-ranging remarks to investors

Jamie Dimon Tom Williams/CQ-Roll Call, Inc via Getty Images 2025-05-19T19:21:31Z Save Saved Read in app This story is available exclusively to Business Insider subscribers. Become an Insider and start reading now. Have an account? JPMorgan CEO Jamie Dimon addressed a range of topics at the firm's Investor Day meeting on Monday. He said the bank will allow investors to buy bitcoin, while warning against stagflation. He sounded dour on the economy but hopeful on the potential for a regulatory reset. Jamie Dimon isn't a fan of bitcoin, but he plans to start offering it to clients of JPMorgan Chase nonetheless."We are going to allow you to buy it," Dimon said at the bank's annual presentation for investors on Monday. "We're not going to custody it. We're going to put it in statements for clients.""I don't think you should smoke. But I defend your right to smoke," he said in explaining his position. The bitcoin comments came as the JPMorgan CEO, often considered Wall Street's elder statesman, took the stage to answer questions from investors and research analysts. In the roughly 40-minute session, he touched on a range of topics, from the economy to what he expects from Trump's regulators. Dimon sounded a dour note on the economy, saying he thinks the risk of stagflation is "two times" higher than many think, and making dire predictions on credit as an investment class. "I think the worst one for a bank and for most companies is stagflation," he continued, warning: "I think the odds of that are probably two times what the market thinks."He also said the bank had lost some commercial opportunities as a result of Trump's trade war. "We've lost business because of that," he said in response to an analyst's question.He sounded upbeat, however, when it came to the president's regulatory agenda. "I think that the Secretary of Treasury, the president of the United States, the new head of the OCC, the new head of the CFPB, Michelle Bowman at Federal Reserve, and the SEC have all made it clear that they want to fix some of the things they think are broken," he said. "I think they'll accomplish some of that. Some will take longer than others, but they all want to do it."He called on regulators to consider lightening regulations for publicly traded companies, which he said have been halved since the 1990s, from 8,000 to 4,000. "We're driving companies out of the public marketplace because of expensive reporting, litigation, cookie-cutter approaches to boards, compensation, and litigation," he said. "I would love to be a private company," he added.Dimon also raised questions about the rapid expansion of investments in credit, including through funds raised to make nonbank loans, or private credit. "I don't like making forecasts," Dimon said, "but I am not a buyer of credit today. I think credit today is a bad risk," he said, adding, "I think that people who haven't been through major downturns are missing the point about what can happen in credit." As interest rates rise and economic conditions soften, the risk of credit defaults rises, sometimes leaving borrowers strapped for cash and lenders struggling to recoup capital.Earlier in the day, Marianne Lake, JPMorgan's CEO of consumer and community banking, said the firm was "closely monitoring the whole ecosystem" of lending but not giving up in spite of warning signs. "The environment is very challenging for home lending and auto," she said, adding, "but we remain committed."Have a tip? Contact this reporter via email at ralexander@businessinsider.com or SMS/Signal at 561-247-5758. Use a personal email address and a nonwork device; here's our guide to sharing information securely. Recommended video

·79 مشاهدة

الرجاء تسجيل الدخول , للأعجاب والمشاركة والتعليق على هذا!
InformationWeek شارك رابطًا

2025-05-19 19:35:26 ·

Don’t Automate the Wrong Thing: Lessons from Building Agentic Hiring Systems

Pankaj Khurana, VP Technology & Consulting, RocketMay 19, 20253 Min ReadElenaBs via Alamy StockFor the past four years, I’ve been building AI-powered tools that help recruiters do their job better. Before that, I was a recruiter myself -- reading resumes, making calls, living the grind. And here’s one thing I’ve learned from straddling both worlds: In hiring, automating the wrong thing can quietly erode everything that makes your process work. As engineering leaders, we’re constantly told to streamline and optimize. Move fast. But if you automate the wrong step -- like how candidates are filtered, scored, or messaged -- you might be replacing good human judgment with rigid shortcuts. And often, you won’t notice the damage until weeks later, when engagement plummets or teams stop trusting your system. The Allure of Automation Hiring is messy. Resumes come in all shapes. Job descriptions are vague. Recruiters are overworked. AI seems like a godsend. We start by automating outreach. Then scoring. Then matching. Eventually, someone asks: can this whole thing run without a person? But here’s the rub: many hiring decisions are deeply contextual. Should a product manager with a non-traditional background be fast-tracked for a high-growth SaaS role? That’s not a “yes/no” the system can decide for you. Early on at Rocket, we made that mistake. Our scoring engine prioritized resumes based solely on skills overlap. It was fast -- but completely off for roles that required nuance. We had to pause, rethink, and admit: “This isn’t working like we hoped.” Related:What Agentic Systems Do Well I’m not anti-automation. Far from it. But it has to be paired with human review. We found that agentic systems -- AI tools with autonomy to assist but not decide -- were far more effective. Think copilots, not autopilots. For example, our system can: Suggest better phrasing for job descriptions Flag resumes that match roles 80% or more Recommend outreach templates based on role and tone But it never auto-rejects or sends messages without review. The AI suggests; the recruiter decides. That balance makes all the difference. Lessons Learned: Where Automation Fails One of our biggest missteps? Automating outreach too heavily. We thought sending personalized AI-written emails at scale would boost response rates. It didn’t. Candidates sensed something off. The emails looked polished but felt cold. Engagement dropped. We eventually went back to having humans rewrite the AI drafts. That one shift nearly doubled our positive response rate. Why? Because candidates want to feel seen -- not sorted. Related:A CIO’s Checklist: What Not to Automate If you’re leading an AI initiative in hiring, here’s a checklist we now swear by: Don’t automate decisions that impact trust. Rejections, scores, hiring calls? Keep a human in the loop. Avoid automating tasks with high context needs. A great candidate might not use trendy buzzwords. That doesn’t make them a bad fit. Be careful with candidate-facing automation. Generic outreach harms brand perception. Do automate the repetitive stuff. Parsing, meeting scheduling, draft -- automate those and give time back to your team. Human-AI Collaboration Wins We saw the best outcomes when recruiters felt like they had an assistant -- not a competitor. Here’s one quick story: A recruiter used our AI to shortlist 10 profiles for a hard-to-fill GTM analyst role. She reviewed five, adjusted the messaging tone slightly, and got two responses in under a day. Same tools -- different mindset. Feedback loops mattered too. We built in ways for users to rate suggestions. The model kept improving -- and more importantly, people trusted it more. Final Thought: Think Like a System Designer If you’re building AI into your hiring stack, go beyond automation. Think augmentation. Don’t just ask, “Can this task be automated?” Instead, ask, “If I automate this, what do we lose in context, empathy, or nuance?” Related:Agentic hiring systems can deliver speed and scale -- but only if we let people stay in control of what matters most. About the AuthorPankaj KhuranaVP Technology & Consulting, RocketPankaj Khurana is VP of Technology & Consulting at Rocket, an AI-driven recruiting firm. He has over 20 years of experience in hiring and tech and has led the development of agentic hiring tools used by top US startups. See more from Pankaj KhuranaWebinarsMore WebinarsReportsMore ReportsNever Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.SIGN-UPYou May Also Like
#dont #automate #wrong #thing #lessons

Don’t Automate the Wrong Thing: Lessons from Building Agentic Hiring Systems
Pankaj Khurana, VP Technology & Consulting, RocketMay 19, 20253 Min ReadElenaBs via Alamy StockFor the past four years, I’ve been building AI-powered tools that help recruiters do their job better. Before that, I was a recruiter myself -- reading resumes, making calls, living the grind. And here’s one thing I’ve learned from straddling both worlds: In hiring, automating the wrong thing can quietly erode everything that makes your process work. As engineering leaders, we’re constantly told to streamline and optimize. Move fast. But if you automate the wrong step -- like how candidates are filtered, scored, or messaged -- you might be replacing good human judgment with rigid shortcuts. And often, you won’t notice the damage until weeks later, when engagement plummets or teams stop trusting your system. The Allure of Automation Hiring is messy. Resumes come in all shapes. Job descriptions are vague. Recruiters are overworked. AI seems like a godsend. We start by automating outreach. Then scoring. Then matching. Eventually, someone asks: can this whole thing run without a person? But here’s the rub: many hiring decisions are deeply contextual. Should a product manager with a non-traditional background be fast-tracked for a high-growth SaaS role? That’s not a “yes/no” the system can decide for you. Early on at Rocket, we made that mistake. Our scoring engine prioritized resumes based solely on skills overlap. It was fast -- but completely off for roles that required nuance. We had to pause, rethink, and admit: “This isn’t working like we hoped.” Related:What Agentic Systems Do Well I’m not anti-automation. Far from it. But it has to be paired with human review. We found that agentic systems -- AI tools with autonomy to assist but not decide -- were far more effective. Think copilots, not autopilots. For example, our system can: Suggest better phrasing for job descriptions Flag resumes that match roles 80% or more Recommend outreach templates based on role and tone But it never auto-rejects or sends messages without review. The AI suggests; the recruiter decides. That balance makes all the difference. Lessons Learned: Where Automation Fails One of our biggest missteps? Automating outreach too heavily. We thought sending personalized AI-written emails at scale would boost response rates. It didn’t. Candidates sensed something off. The emails looked polished but felt cold. Engagement dropped. We eventually went back to having humans rewrite the AI drafts. That one shift nearly doubled our positive response rate. Why? Because candidates want to feel seen -- not sorted. Related:A CIO’s Checklist: What Not to Automate If you’re leading an AI initiative in hiring, here’s a checklist we now swear by: Don’t automate decisions that impact trust. Rejections, scores, hiring calls? Keep a human in the loop. Avoid automating tasks with high context needs. A great candidate might not use trendy buzzwords. That doesn’t make them a bad fit. Be careful with candidate-facing automation. Generic outreach harms brand perception. Do automate the repetitive stuff. Parsing, meeting scheduling, draft -- automate those and give time back to your team. Human-AI Collaboration Wins We saw the best outcomes when recruiters felt like they had an assistant -- not a competitor. Here’s one quick story: A recruiter used our AI to shortlist 10 profiles for a hard-to-fill GTM analyst role. She reviewed five, adjusted the messaging tone slightly, and got two responses in under a day. Same tools -- different mindset. Feedback loops mattered too. We built in ways for users to rate suggestions. The model kept improving -- and more importantly, people trusted it more. Final Thought: Think Like a System Designer If you’re building AI into your hiring stack, go beyond automation. Think augmentation. Don’t just ask, “Can this task be automated?” Instead, ask, “If I automate this, what do we lose in context, empathy, or nuance?” Related:Agentic hiring systems can deliver speed and scale -- but only if we let people stay in control of what matters most. About the AuthorPankaj KhuranaVP Technology & Consulting, RocketPankaj Khurana is VP of Technology & Consulting at Rocket, an AI-driven recruiting firm. He has over 20 years of experience in hiring and tech and has led the development of agentic hiring tools used by top US startups. See more from Pankaj KhuranaWebinarsMore WebinarsReportsMore ReportsNever Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.SIGN-UPYou May Also Like #dont #automate #wrong #thing #lessons

WWW.INFORMATIONWEEK.COM

Don’t Automate the Wrong Thing: Lessons from Building Agentic Hiring Systems

Pankaj Khurana, VP Technology & Consulting, RocketMay 19, 20253 Min ReadElenaBs via Alamy StockFor the past four years, I’ve been building AI-powered tools that help recruiters do their job better. Before that, I was a recruiter myself -- reading resumes, making calls, living the grind. And here’s one thing I’ve learned from straddling both worlds: In hiring, automating the wrong thing can quietly erode everything that makes your process work. As engineering leaders, we’re constantly told to streamline and optimize. Move fast. But if you automate the wrong step -- like how candidates are filtered, scored, or messaged -- you might be replacing good human judgment with rigid shortcuts. And often, you won’t notice the damage until weeks later, when engagement plummets or teams stop trusting your system. The Allure of Automation Hiring is messy. Resumes come in all shapes. Job descriptions are vague. Recruiters are overworked. AI seems like a godsend. We start by automating outreach. Then scoring. Then matching. Eventually, someone asks: can this whole thing run without a person? But here’s the rub: many hiring decisions are deeply contextual. Should a product manager with a non-traditional background be fast-tracked for a high-growth SaaS role? That’s not a “yes/no” the system can decide for you. Early on at Rocket, we made that mistake. Our scoring engine prioritized resumes based solely on skills overlap. It was fast -- but completely off for roles that required nuance. We had to pause, rethink, and admit: “This isn’t working like we hoped.” Related:What Agentic Systems Do Well I’m not anti-automation. Far from it. But it has to be paired with human review. We found that agentic systems -- AI tools with autonomy to assist but not decide -- were far more effective. Think copilots, not autopilots. For example, our system can: Suggest better phrasing for job descriptions Flag resumes that match roles 80% or more Recommend outreach templates based on role and tone But it never auto-rejects or sends messages without review. The AI suggests; the recruiter decides. That balance makes all the difference. Lessons Learned: Where Automation Fails One of our biggest missteps? Automating outreach too heavily. We thought sending personalized AI-written emails at scale would boost response rates. It didn’t. Candidates sensed something off. The emails looked polished but felt cold. Engagement dropped. We eventually went back to having humans rewrite the AI drafts. That one shift nearly doubled our positive response rate. Why? Because candidates want to feel seen -- not sorted. Related:A CIO’s Checklist: What Not to Automate If you’re leading an AI initiative in hiring, here’s a checklist we now swear by: Don’t automate decisions that impact trust. Rejections, scores, hiring calls? Keep a human in the loop. Avoid automating tasks with high context needs. A great candidate might not use trendy buzzwords. That doesn’t make them a bad fit. Be careful with candidate-facing automation. Generic outreach harms brand perception. Do automate the repetitive stuff. Parsing, meeting scheduling, draft -- automate those and give time back to your team. Human-AI Collaboration Wins We saw the best outcomes when recruiters felt like they had an assistant -- not a competitor. Here’s one quick story: A recruiter used our AI to shortlist 10 profiles for a hard-to-fill GTM analyst role. She reviewed five, adjusted the messaging tone slightly, and got two responses in under a day. Same tools -- different mindset. Feedback loops mattered too. We built in ways for users to rate suggestions. The model kept improving -- and more importantly, people trusted it more. Final Thought: Think Like a System Designer If you’re building AI into your hiring stack, go beyond automation. Think augmentation. Don’t just ask, “Can this task be automated?” Instead, ask, “If I automate this, what do we lose in context, empathy, or nuance?” Related:Agentic hiring systems can deliver speed and scale -- but only if we let people stay in control of what matters most. About the AuthorPankaj KhuranaVP Technology & Consulting, RocketPankaj Khurana is VP of Technology & Consulting at Rocket, an AI-driven recruiting firm. He has over 20 years of experience in hiring and tech and has led the development of agentic hiring tools used by top US startups. See more from Pankaj KhuranaWebinarsMore WebinarsReportsMore ReportsNever Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.SIGN-UPYou May Also Like

·100 مشاهدة

الرجاء تسجيل الدخول , للأعجاب والمشاركة والتعليق على هذا!
Computer Weekly شارك رابطًا

2025-05-19 19:33:19 ·

Trump visit bolsters Saudi AI

Saudi Arabia’s artificial intelligenceambitions were boosted following US president Donald Trump’s recent trip to the Middle East.
The Trump administration’s import tariffs and AI chip export restrictions appear to have been behind the decision by Saudi Arabian datacentre operator DataVolt to sign a major server deal with US firm Supermicro. The company’s selection of the US firm for its bn investment in AI datacentre technology will secure graphics processing unithardware currently seen as the best way to provide the accelerated processing power AI workloads require.
“Partnering with Supermicro guarantees us a US-made supply chain for critical GPU systems and positions DataVolt to accelerate our investment plans,” said Rajit Nanda, CEO of DataVolt.
Supermicro said the collaboration would fast-track delivery of its ultra-dense GPU platforms, storage and rack-based systems for DataVolt’s hyperscale gigawatt-class renewable and net-zero green AI datacentre facilities.
The hardware uses liquid cooling, which Supermicro said reduces power costs by up to 40%, which allows datacentres to run more efficiently with lower power usage effectiveness.
“We are excited to collaborate with DataVolt to bring our advanced AI systems featuring the latest direct liquid cooling technology, powered by local renewable, sustainable and net-zero green technology,” said Charles Liang, president and CEO of Supermicro.
Supermicro claims its latest AI infrastructure is able to lower the total cost of ownership by up to 20%.

Along with its Saudi AI datacentre build, Datavolt, Google, Oracle, Salesforce, AMD and Uber have said they are committed to the investment of bn in cutting-edge transformative technologies in both the US and Saudi Arabia.
Oracle said it plans to invest bn over the next 10 years to deliver cloud and AI technology to Saudi Arabia. “Thanks to the decisive actions and strong leadership of President Trump and his administration, Oracle is providing the world’s most advanced cloud and AI technology to Saudi Arabia,” said Safra Catz, CEO at Oracle. “Our expanded partnership with the kingdom will create new opportunities for its economy and deliver better health outcomes for its people.”
As part of Trump’s visit, Humain, a subsidiary of Saudi Arabia’s Public Investment Fund, also announced a bn deal with US chipmaker AMD to develop 500 megawatts of open, scalable, resilient and cost-efficient AI infrastructure that spans Saudi Arabia and the US.
Through the collaboration, Humain will oversee end-to-end delivery, including hyperscale datacentres, sustainable power systems and global fibre interconnects, while AMD has said it will be providing its AI compute processors and the AMD ROCm open software ecosystem.
Tareq Amin, CEO of Humain, described the initiative as “an open collaboration to the world’s innovators”. “We are democratising AI at the compute level, ensuring that access to advanced AI is limited only by imagination, not by infrastructure,” he said.

about AI in Saudi Arabia

Saudi puts bn into AI as experts debate next steps: The kingdom’s Leap 2025 tech show is the backdrop for huge investment, plus debate over the future of artificial intelligence as a productivity tool but which can also potentially undermine human society.
Saudi Arabia struggling to reach global leadership in deeptech: Petrostate monarchy trying to build surrogate industry made of foreign startups because own ecosystem is too immature.

A number of US tech firms have made major investments in Saudi this year, driven by the kingdom’s ambitions to boost AI. Data for the Saudi Data and AI Authority from Accenture shows that generative AIhas the potential to elevate its gross domestic productby approximately bn by augmenting and automating nearly a third of all jobs.
In February, Accenture said it was collaborating with Google Cloud to accelerate the adoption of cloud and GenAI capabilities in Saudi Arabia to support the KSA’s local data, operational and software sovereignty needs. As part of this work, Google and Accenture announced a joint centre of excellence for GenAI.
During the Leap 2025 conference, Salesforce said it would be investing m in Saudi, with the focus again on AI. The company announced a regional head office in Riyadh, and as part of this expansion, Salesforce said it was working with Amazon Web Servicesto deliver its Hyperforce, next-generation platform architecture to the kingdom.
The work with AWS means Salesforce’s global customers are able to run workloads locally through a distributed public cloud infrastructure that Salesforce said complies with local regulations.
Commenting on Salesforce’s plans, Abdullah Alswaha, minister of communications and information technology, said: “We look forward to seeing Salesforce expand its presence here and welcome the investment in AI that will drive unprecedented innovation and operational efficiency, supporting the realisation of Saudi Arabia’s Vision 2030 goals.”
While the White House has linked Trump’s visit to the investments major tech firms are making in Saudi Arabia, the kingdom faces a number of challenges as it fleshes out its AI ambitions.
According to a March 2025 paper from the Carnegie Endowment for International Peace, Vision 2030 represents a future-orientated vision to make religion less of a focal point of its society, particularly with regards to how Saudi Arabia is perceived internationally. But the paper notes that private-sector employers face the challenge of hiring Saudi citizens to fill high-skilled jobs. “This skills gap will only worsen as the kingdom targets a wide range of highly skilled sectors, including AI startups, green technology, datacentres, e-gaming, fintech and EV production,” it says.
#trump #visit #bolsters #saudi

Trump visit bolsters Saudi AI
Saudi Arabia’s artificial intelligenceambitions were boosted following US president Donald Trump’s recent trip to the Middle East. The Trump administration’s import tariffs and AI chip export restrictions appear to have been behind the decision by Saudi Arabian datacentre operator DataVolt to sign a major server deal with US firm Supermicro. The company’s selection of the US firm for its bn investment in AI datacentre technology will secure graphics processing unithardware currently seen as the best way to provide the accelerated processing power AI workloads require. “Partnering with Supermicro guarantees us a US-made supply chain for critical GPU systems and positions DataVolt to accelerate our investment plans,” said Rajit Nanda, CEO of DataVolt. Supermicro said the collaboration would fast-track delivery of its ultra-dense GPU platforms, storage and rack-based systems for DataVolt’s hyperscale gigawatt-class renewable and net-zero green AI datacentre facilities. The hardware uses liquid cooling, which Supermicro said reduces power costs by up to 40%, which allows datacentres to run more efficiently with lower power usage effectiveness. “We are excited to collaborate with DataVolt to bring our advanced AI systems featuring the latest direct liquid cooling technology, powered by local renewable, sustainable and net-zero green technology,” said Charles Liang, president and CEO of Supermicro. Supermicro claims its latest AI infrastructure is able to lower the total cost of ownership by up to 20%. Along with its Saudi AI datacentre build, Datavolt, Google, Oracle, Salesforce, AMD and Uber have said they are committed to the investment of bn in cutting-edge transformative technologies in both the US and Saudi Arabia. Oracle said it plans to invest bn over the next 10 years to deliver cloud and AI technology to Saudi Arabia. “Thanks to the decisive actions and strong leadership of President Trump and his administration, Oracle is providing the world’s most advanced cloud and AI technology to Saudi Arabia,” said Safra Catz, CEO at Oracle. “Our expanded partnership with the kingdom will create new opportunities for its economy and deliver better health outcomes for its people.” As part of Trump’s visit, Humain, a subsidiary of Saudi Arabia’s Public Investment Fund, also announced a bn deal with US chipmaker AMD to develop 500 megawatts of open, scalable, resilient and cost-efficient AI infrastructure that spans Saudi Arabia and the US. Through the collaboration, Humain will oversee end-to-end delivery, including hyperscale datacentres, sustainable power systems and global fibre interconnects, while AMD has said it will be providing its AI compute processors and the AMD ROCm open software ecosystem. Tareq Amin, CEO of Humain, described the initiative as “an open collaboration to the world’s innovators”. “We are democratising AI at the compute level, ensuring that access to advanced AI is limited only by imagination, not by infrastructure,” he said. about AI in Saudi Arabia Saudi puts bn into AI as experts debate next steps: The kingdom’s Leap 2025 tech show is the backdrop for huge investment, plus debate over the future of artificial intelligence as a productivity tool but which can also potentially undermine human society. Saudi Arabia struggling to reach global leadership in deeptech: Petrostate monarchy trying to build surrogate industry made of foreign startups because own ecosystem is too immature. A number of US tech firms have made major investments in Saudi this year, driven by the kingdom’s ambitions to boost AI. Data for the Saudi Data and AI Authority from Accenture shows that generative AIhas the potential to elevate its gross domestic productby approximately bn by augmenting and automating nearly a third of all jobs. In February, Accenture said it was collaborating with Google Cloud to accelerate the adoption of cloud and GenAI capabilities in Saudi Arabia to support the KSA’s local data, operational and software sovereignty needs. As part of this work, Google and Accenture announced a joint centre of excellence for GenAI. During the Leap 2025 conference, Salesforce said it would be investing m in Saudi, with the focus again on AI. The company announced a regional head office in Riyadh, and as part of this expansion, Salesforce said it was working with Amazon Web Servicesto deliver its Hyperforce, next-generation platform architecture to the kingdom. The work with AWS means Salesforce’s global customers are able to run workloads locally through a distributed public cloud infrastructure that Salesforce said complies with local regulations. Commenting on Salesforce’s plans, Abdullah Alswaha, minister of communications and information technology, said: “We look forward to seeing Salesforce expand its presence here and welcome the investment in AI that will drive unprecedented innovation and operational efficiency, supporting the realisation of Saudi Arabia’s Vision 2030 goals.” While the White House has linked Trump’s visit to the investments major tech firms are making in Saudi Arabia, the kingdom faces a number of challenges as it fleshes out its AI ambitions. According to a March 2025 paper from the Carnegie Endowment for International Peace, Vision 2030 represents a future-orientated vision to make religion less of a focal point of its society, particularly with regards to how Saudi Arabia is perceived internationally. But the paper notes that private-sector employers face the challenge of hiring Saudi citizens to fill high-skilled jobs. “This skills gap will only worsen as the kingdom targets a wide range of highly skilled sectors, including AI startups, green technology, datacentres, e-gaming, fintech and EV production,” it says. #trump #visit #bolsters #saudi

WWW.COMPUTERWEEKLY.COM

Trump visit bolsters Saudi AI

Saudi Arabia’s artificial intelligence (AI) ambitions were boosted following US president Donald Trump’s recent trip to the Middle East. The Trump administration’s import tariffs and AI chip export restrictions appear to have been behind the decision by Saudi Arabian datacentre operator DataVolt to sign a major server deal with US firm Supermicro. The company’s selection of the US firm for its $20bn investment in AI datacentre technology will secure graphics processing unit (GPU) hardware currently seen as the best way to provide the accelerated processing power AI workloads require. “Partnering with Supermicro guarantees us a US-made supply chain for critical GPU systems and positions DataVolt to accelerate our investment plans,” said Rajit Nanda, CEO of DataVolt. Supermicro said the collaboration would fast-track delivery of its ultra-dense GPU platforms, storage and rack-based systems for DataVolt’s hyperscale gigawatt-class renewable and net-zero green AI datacentre facilities. The hardware uses liquid cooling, which Supermicro said reduces power costs by up to 40%, which allows datacentres to run more efficiently with lower power usage effectiveness. “We are excited to collaborate with DataVolt to bring our advanced AI systems featuring the latest direct liquid cooling technology (DLC-2), powered by local renewable, sustainable and net-zero green technology,” said Charles Liang, president and CEO of Supermicro. Supermicro claims its latest AI infrastructure is able to lower the total cost of ownership by up to 20%. Along with its Saudi AI datacentre build, Datavolt, Google, Oracle, Salesforce, AMD and Uber have said they are committed to the investment of $80bn in cutting-edge transformative technologies in both the US and Saudi Arabia. Oracle said it plans to invest $14bn over the next 10 years to deliver cloud and AI technology to Saudi Arabia. “Thanks to the decisive actions and strong leadership of President Trump and his administration, Oracle is providing the world’s most advanced cloud and AI technology to Saudi Arabia,” said Safra Catz, CEO at Oracle. “Our expanded partnership with the kingdom will create new opportunities for its economy and deliver better health outcomes for its people.” As part of Trump’s visit, Humain, a subsidiary of Saudi Arabia’s Public Investment Fund (PIF), also announced a $10bn deal with US chipmaker AMD to develop 500 megawatts of open, scalable, resilient and cost-efficient AI infrastructure that spans Saudi Arabia and the US. Through the collaboration, Humain will oversee end-to-end delivery, including hyperscale datacentres, sustainable power systems and global fibre interconnects, while AMD has said it will be providing its AI compute processors and the AMD ROCm open software ecosystem. Tareq Amin, CEO of Humain, described the initiative as “an open collaboration to the world’s innovators”. “We are democratising AI at the compute level, ensuring that access to advanced AI is limited only by imagination, not by infrastructure,” he said. Read more about AI in Saudi Arabia Saudi puts $15bn into AI as experts debate next steps: The kingdom’s Leap 2025 tech show is the backdrop for huge investment, plus debate over the future of artificial intelligence as a productivity tool but which can also potentially undermine human society. Saudi Arabia struggling to reach global leadership in deeptech: Petrostate monarchy trying to build surrogate industry made of foreign startups because own ecosystem is too immature. A number of US tech firms have made major investments in Saudi this year, driven by the kingdom’s ambitions to boost AI. Data for the Saudi Data and AI Authority from Accenture shows that generative AI (GenAI) has the potential to elevate its gross domestic product (GDP) by approximately $42.3bn by augmenting and automating nearly a third of all jobs. In February, Accenture said it was collaborating with Google Cloud to accelerate the adoption of cloud and GenAI capabilities in Saudi Arabia to support the KSA’s local data, operational and software sovereignty needs. As part of this work, Google and Accenture announced a joint centre of excellence for GenAI. During the Leap 2025 conference, Salesforce said it would be investing $500m in Saudi, with the focus again on AI. The company announced a regional head office in Riyadh, and as part of this expansion, Salesforce said it was working with Amazon Web Services (AWS) to deliver its Hyperforce, next-generation platform architecture to the kingdom. The work with AWS means Salesforce’s global customers are able to run workloads locally through a distributed public cloud infrastructure that Salesforce said complies with local regulations. Commenting on Salesforce’s plans, Abdullah Alswaha, minister of communications and information technology, said: “We look forward to seeing Salesforce expand its presence here and welcome the investment in AI that will drive unprecedented innovation and operational efficiency, supporting the realisation of Saudi Arabia’s Vision 2030 goals.” While the White House has linked Trump’s visit to the investments major tech firms are making in Saudi Arabia, the kingdom faces a number of challenges as it fleshes out its AI ambitions. According to a March 2025 paper from the Carnegie Endowment for International Peace, Vision 2030 represents a future-orientated vision to make religion less of a focal point of its society, particularly with regards to how Saudi Arabia is perceived internationally. But the paper notes that private-sector employers face the challenge of hiring Saudi citizens to fill high-skilled jobs. “This skills gap will only worsen as the kingdom targets a wide range of highly skilled sectors, including AI startups, green technology, datacentres, e-gaming, fintech and EV production,” it says.

·122 مشاهدة

الرجاء تسجيل الدخول , للأعجاب والمشاركة والتعليق على هذا!

الصفحات المعززة

انضم إلينا

اللغات

NASA Engineers Revive Long-Dormant Thrusters on Voyager 1, the Farthest Spacecraft From Earth, in a 'Miracle Save'

NVIDIA and Microsoft Accelerate Agentic AI Innovation, From Cloud to PC

Windows 10 emergency update KB5061768 fixes BitLocker boot loops

$8000* Disaster Prebuilt PC - Corsair &amp; Origin Fail Again

Apple Magic Mouse vs. Logitech MX Master 3S: Should You Finally Ditch Your Dinky Magic Mouse?

Everything you need to know about Memory Profiler 1.0.0

⚡ @ChaosGroup Enscape &amp; @Autodesk VRED just leveled up with DLSS 4! 🚀 Enjoy sharper visuals + faster performance for real-time design &amp; 3D visuali...

Jamie Dimon opens the door to bitcoin, warns against stagflation in wide-ranging remarks to investors

Don’t Automate the Wrong Thing: Lessons from Building Agentic Hiring Systems

Trump visit bolsters Saudi AI

$8000* Disaster Prebuilt PC - Corsair & Origin Fail Again

⚡ @ChaosGroup Enscape & @Autodesk VRED just leveled up with DLSS 4! 🚀 Enjoy sharper visuals + faster performance for real-time design & 3D visuali...