For this week's curated picks of architecture and design competitions listed on Bustler, we are featuring four calls seeking the world's most innovative industrial, product, and graphic designs, excellent projects across architectural, interior, construction product, and landscape design, proposals for the renewal of a traditional tenement-style apartment building in Milan, and the best ideas that redefine skyscraper design.For the complete directory of newly listed competitions, click here.

While Nintendo has priced the upcoming Switch 2 at $449.99, the company has been quiet about the tariffs announced by the United States. In light of the tariffs, however, Nintendo has also put pre-orders for the consoles, which were supposed to go live over the weekend, on pause. It is worth noting that, while Nintendo is yet to make any announcements about the matter, it is likely that the Switch 2 will get some sort of increase to its price tag. This is because of the console’s manufacturing process taking place in Vietnam, where the US has levied a 46 percent tariff. This would significantly raise the price of the console in the country. While the US has more recently announced a pause on the tariff plan for some countries, the Switch 2 will still have to contend with a 10 percent tariff in the country. According to Bloomberg, Bernstein analyst Robin Zhu believes that Nintendo will likely go ahead with its $449.99 price tag if the 10 percent tariffs were the only thing to deal with. However, Zhu expects Nintendo to raise the price by quite a bit if the 46 percent tariffs are levied. “Nintendo probably keeps pricing at $450 and just takes the hit on margin,” said Zhu of the 10 percent tariffs. “At 46% Vietnam tariffs, I expected them to raise by $50 to $100.” Former Nintendo PR managers Kit Ellis and Krysta Lang have noted that Nintendo would benefit quite a bit if it were to “eat the cost,” earning the company quite a bit of good will amongst fans in light of the controversial high price tags for the console and its games. “They can do it, they can eat it, it would not make a huge dent in them,” said Ellis. Yang responded by noting that the high costs of games like Mario Kart World has already lost Nintendo a fair bit of trust with its fans. “They have lost so much trust already with the current negative sentiment around price, and around what they’re doing with the games and charging the high price for Mario Kart World, this would repair some goodwill with their current fan base,” said Yang. When it comes to software, Nintendo has been steadfast about its new $80 price tag for Mario Kart World. In an interview, Nintendo of America VP of Product and Player Experience spoke about how the company believes that Mario Kart World will definitely be worth its price. He also spoke about the value offered by the Switch 2 Edition releases for The Legend of Zelda: Breath of the Wild and The Legend of Zelda: Tears of the Kingdom. Despites pre-orders for the console being halted for the time being, the Nintendo Switch 2 is slated to hit store shelves on June 5. The console will also be available in a bundle alongside Mario Kart World, priced at $499.99. Among several games available for the title at launch will also be the Nintendo Switch 2 Welcome Tour, which will explain all the features of the console to players, and will be priced at $9.99.

A New Satellite Will Map the Carbon Content of Rainforests From Space, and It’s Set to Launch This Month The European Space Agency’s new probe, Biomass, will spend five years orbiting the planet and gathering radar imagery of forests across multiple continents Biomass will monitor the Earth's tropical forests over the next five years. ESA / ATG medialab under CC-BY-SA 3.0 IGO After more than ten years of development, the European Space Agency (ESA) is preparing to launch a satellite that will map the carbon content of the world’s rainforests. The probe, named Biomass, was recently shipped to French Guiana, where it’s slated to lift off from Europe’s Spaceport on April 29 aboard a 230-ton Vega-C rocket. Over the next five years, the spacecraft will orbit the Earth and monitor forests’ above-ground biomass—think trunks and large branches—as well as the height of trees in Africa, Asia and South America. Those measurements, captured with the help of a first-of-its-kind radar, can then serve as a proxy for tracking stored carbon. “What the mission will do, effectively, is weigh the forests it studies,” says physicist Shaun Quegan, the leader of the Biomass science team, to Robin McKie at the Observer. “We know half that weight must be made up of carbon. So, we are going to be able to weigh the carbon content of the world’s tropical forests from space and, crucially, work out how much these are changing over time.” “We will then know the balance of carbon that is flowing to and from the atmosphere,” Quegan adds. “That is enormously important.” Meet Biomass Watch on Forests are valuable carbon sinks—they alone sequester a quarter of human-generated carbon emissions annually. But their ability to store carbon is weakening because of deforestation and environmental degradation: A 2020 study led by researchers at the University of Leeds in England found that rainforests could take up one-third less carbon in the 2010s than in the 1990s. The new probe will allow scientists to understand exactly how carbon levels are fluctuating, teach scientists more about the carbon cycle and map areas of deforestation. “We need to know the health of our tropical forests,” explains Simonetta Cheli, ESA’s director of earth observation programs, to the Observer. “We need to know the quality and diversity of its vegetation and the amount of carbon stored there. To get that information, we are going to create 3D images of them—from the top of the forest canopy to the roots of its trees.” Barely any sunlight reaches the floor of a tropical rainforest, but the space-based probe will be able to image all the way to the ground. That’s because it will carry a new P-band synthetic aperture radar, an imaging system that has never been used by a satellite before. The radar signals sent by the instrument can penetrate through dense forest canopies to visualize what’s below. “By pushing the boundaries of remote sensing, Biomass not only advances our understanding of global forests and carbon cycles but also demonstrates the transformative potential of space-based innovation in tackling Earth’s most pressing environmental challenges,” Cheli adds in a statement. The Biomass satellite, shown at Airbus in Toulouse, France, before it was shipped to French Guiana ESA / S. Corvaja The probe’s capabilities can be transferred to other research areas as well, according to a statement from the ESA. Its radars can be used to observe the structure and motion of ice sheets, for instance, or create models of other areas covered by dense vegetation. In dry areas, it could peer underground. Scientists hope to use the data from Biomass to create new climate change and carbon cycle models by integrating it with artificial intelligence. “That will tell us what is likely to happen in future,” says Quegan to the Observer. “It will tell us what we are up against.” That information can also help inform policy decisions on forest management and climate mitigation. Now, engineers have finished fueling the satellite, and they will be making final preparations before its scheduled launch this month. “It’s marvelous to see the satellite standing proud today, and I wish to thank our industrial partners for all they have done,” Cheli says in a statement. Once Biomass is in orbit, it will collect and deliver “much-needed data to advance Earth science and our understanding of the carbon cycle.” Get the latest stories in your inbox every weekday.

Storycraft -- the AI-powered game platform where players create their own interactive story worlds -- has raised a $3 million.Read More

Agonalea Games secures $2m in funding from Mika Capital Mobile developer will use investment to support work on its upcoming free-to-play title Punko.io Image credit: Agonalea Games News by Sophie McEvoy Staff Writer Published on April 11, 2025 Buenos Aires-based mobile developer Agonalea Games has raised $2 million in a seed funding round backed by Mika Games. The studio will use the investment to continue development and publish its free-to-play mobile game Punko.io. Agonalea Games was founded in 2018 by Pedro Aira (former Sensor Tower product manager) and Vanina Fregoti (former Pixowl project manager). It has a portfolio of eight original games and has contributed to more than ten titles. "We are investing in Agonalea Games not just with capital, but with strategic support to help them build innovative, standout mobile games," said Mika Games founder Nick Katselapov. This marks the third investment from Mika Games, which previously backed studios including Guli Games and Hot Siberians. The investment firm itself is backed by venture capital firm Flint Capital. Mika Games raised $10 million in its initial funding round in April 2024, and intends to secure a further $50 million. In addition to investing in developers, Mika Games also offers developers support in marketing, game design, and monetisation. "83% of mobile games fail within three years, and a significant factor contributing to this is the lack of financing and business guidance," said Katselapov. "We see a huge market gap here, which presents an immense opportunity that we can capitalise on. Our industry experience as both investors and operators positions us well to select talented teams and ambitious games."

This week the University of Texas at Austin announced the founding of a new game industry academic department following a $5 million donation from Saber Interactive co-founder and CEO Matt Karch on behalf of the Karch family. The program, dubbed the "Karch Gaming Institute," will be dedicated to "shaping the next generation of creative leaders" in the video game industry.The program will operate under UT Austin's Moody College of Communication. To the best of our knowledge the donation, one of the first video game studies programs to be named on behalf of a major industry donor. It's also another instance of Karch boosting his profile after taking Saber Interactive out from under Embracer Group in a deal worth $247 million.For comparison, Sony Interactive Entertainment donated $3 million to the University of Southern California's Gerald A. Lawson fund in 2022.The institute's first course offerings expand on the existing game design offerings at UT Austin. "Business of Gaming: From Concept to Console" will explore game development through a business lens from pitching to release, and "Gaming Usability Lab Fundamentals" will educate students on the purpose of usability labs in the video game industry.Both courses seem appropriate for an institute bearing Karch's name. The Saber CEO is the surprisingly rare high-level executive with direct experience in game development, going from the trenches of TimeShift to the leader of the one game company that seems to have escaped the implosion of Embracer relatively unsinged.Related:Karch's largesse is a boon for UT Austin's prospects in teaching students the ins and outs of the game industry. It's also another noteworthy play for an executive who seemingly to relishes more and more time in the public eye.Saber Interactive's CEO has a habit of defending billionaires & criticizing "agendas" in gamesCommentary from various video game C-suite execs has become a relatively subdued affair since the heady moments of the 2000s. Gone are the high-profile stunts like former Xbox Boss Peter Moore's Halo 2 tattoo, more common now are subdued comments on quarterly conference calls. The most high-profile examples of public executive noisemaking today are the friendly press appearances of Microsoft Gaming CEO Phil Spencer and former Activision Blizzard CEO Bobby Kotick's aggressive defense of his reputation following his departure from the company.Karch's vocal—often unprompted—hot takes go against the grain. In April 2024 he said critics were being "too hard" on Embracer and its CEO Lars Wingefors. In May he told Game Developer that not speaking up in defense of Wingefors would be "worse" than "saying something publicly bad."Related:After the successful launch of Saber's Warhammer 40,000: Space Marine II in September 2024, an user claiming to be Karch left a comment on the YouTube channel of far-right personality Asmongold, stating that he saw games that made him "want to cry with their overblown attempts at messaging or imposing morals on gamers" during his time as Embracer's chief operating officer.A Saber Interactive PR representative said the company had "no comment" on whether the account belonged to Karch. On his LinkedIn page, Karch posted that whoever wrote the comment seemed "pretty sharp."He seemed to offer some context for his outspokenness in an interview with Game File's Stephen Totilo (which took place during a flight on his private jet), saying independent developers like Tim Sweeney "fight the good fight" because they "have chips on their shoulders that are...often bigger than other parts of their body. It's because we've all been taken advantage of."Now with an entire "gaming institute" bearing his surname, Karch certainly seems interested in keeping his name in the public eye. If Saber's star continues to rise, it's likely we'll hear more from the opinionated exec.Related:And with multimillion dollar donations benefitting public institutions, many in the industry may be more likely to listen close to what he says.

Apple may finally be ready to roll out its smarter, more personalized Siri features before the 2025 holiday season, according to The New York Times. The publication cited three sources with knowledge of Apple’s plan to release a virtual assistant “in the fall” that can edit and send photos to a friend on request — features that were supposed to arrive in iOS 18. This is the earliest prospective rollout timeline we’ve seen so far. Apple spokesperson Jacqueline Roy said in a statement to Daring Fireball in March that the company expects to start releasing its upgraded Siri features “in the coming year.” Meanwhile, Bloomberg’s Mark Gurman said Apple’s AI department believes a ”true modernized, conversational version of Siri” won’t be ready until 2027 “at best” due to difficulties with development. Some of the delays were reportedly caused by leadership issues within the company. John Giannandrea, who formerly led AI and Siri at Apple, was replaced in March after CEO Tim Cook “lost confidence” in his capabilities according to a report from Bloomberg, while The Information reports that senior Apple director Robby Walker and software executive Sebastien Marineau-Mes butted heads over who should oversee Siri’s upgrade project. Several former employees from Apple’s AI and machine learning (AI/ML) group singled Walker out to the publication as lacking the ambition or risk-taking necessary to revamp Siri, and that the team had been dubbed “AIMLess” by internal engineers. The New York Times reports that earlier setbacks for the project occurred in 2023 when Cook’s efforts to double the team’s budget for AI chips were reportedly dashed by Luca Maestri, Apple’s finance chief. Maestri reportedly “reduced the increase to less than half that” and instead directed the team to make their existing chips — 50,000 of which were more than five years old — more efficient. Sources with knowledge of Cook’s request told the publication that this was “far fewer” than the hundreds of thousands of chips being purchased by competitors like Microsoft, Amazon, Google, and Meta.

Recent advancements in LLMs have significantly enhanced their reasoning capabilities, particularly through RL-based fine-tuning. Initially trained with supervised learning for token prediction, these models undergo RL post-training, exploring various reasoning paths to arrive at correct answers, similar to how an agent navigates a game. This process leads to emergent behaviors such as self-correction, often called the “aha moment,” where models begin revising their mistakes without explicit instruction. While this improves accuracy, it also results in much longer responses, increasing token usage, computational costs, and latency. Despite assumptions that longer outputs equate to better reasoning, research shows mixed results—some improvements are seen, but excessively lengthy answers can also reduce performance, indicating diminishing returns. Researchers are exploring ways to balance reasoning quality and efficiency to address this. Methods include using smaller, faster models, applying prompt engineering to reduce verbosity, and developing reward-shaping techniques encouraging concise yet effective reasoning. One notable approach is long-to-short distillation, where models learn from detailed explanations and are trained to produce shorter yet accurate answers. Using these techniques, models like Kimi have demonstrated competitive performance even against larger models like GPT-4 while consuming fewer tokens. Studies also highlight the concept of “token complexity,” showing that problems require a minimum token threshold for accurate resolution, and prompt strategies aimed at conciseness often fall short of this optimal point. Overall, the findings emphasize the importance of developing more efficient reasoning methods without compromising performance. Researchers from Wand AI challenge the belief that longer responses inherently lead to better reasoning in large language models. Through theoretical analysis and experiments, they show that this verbosity is a by-product of RL optimization rather than a necessity for accuracy. Interestingly, concise answers often correlate with higher correctness, and correct responses are shorter than incorrect ones. They propose a two-phase RL training approach: The first phase enhances reasoning ability, while the second enforces conciseness using a small dataset. This method reduces response length without sacrificing accuracy, offering improved efficiency and performance with minimal computational cost. Longer responses do not always lead to better performance in language models. RL post-training tends to reduce response length while maintaining or improving accuracy, especially early in training. This counters the belief that long reasoning chains are necessary for correctness. The phenomenon is tied to “deadends,” where excessively long outputs risk veering off-course. Analyzing language tasks as Markov Decision Processes reveals that RL minimizes loss, not length, and longer outputs only arise when rewards are consistently negative. A two-phase RL strategy—first on hard problems, then on solvable ones—can boost reasoning while eventually promoting conciseness and robustness. The two-phase RL strategy led to notable performance gains across different model sizes. Training on varying difficulty levels showed that easier problems helped models shorten responses while maintaining or improving accuracy. A second RL phase using just eight math problems produced more concise and robust outputs across benchmarks like AIME, AMC, and MATH-500, with similar trends seen in STEM tasks from MMLU. Even minimal RL post-training improved accuracy and stability under low-temperature sampling. Furthermore, models without prior RL refinement, such as Qwen-Math-v2.5, showed large accuracy boosts—up to 30% from training on only four math problems. In conclusion, the study presents a two-phase RL post-training method that improves reasoning and conciseness in language models. The first phase enhances accuracy, while the second focuses on shortening responses without sacrificing performance. Applied to R1 models, this approach reduced response length by over 40% while maintaining accuracy, especially at low temperatures. The findings reveal that longer answers are not inherently better and that targeted RL can achieve concise reasoning. The study also highlights that even minimal RL training can greatly benefit non-reasoning models, emphasizing the value of including moderately solvable problems and carefully tuning PPO parameters. Check out the Paper. All credit for this research goes to the researchers of this project. Also, feel free to follow us on Twitter and don’t forget to join our 85k+ ML SubReddit. Sana HassanSana Hassan, a consulting intern at Marktechpost and dual-degree student at IIT Madras, is passionate about applying technology and AI to address real-world challenges. With a keen interest in solving practical problems, he brings a fresh perspective to the intersection of AI and real-life solutions.Sana Hassanhttps://www.marktechpost.com/author/sana-hassan/RoR-Bench: Revealing Recitation Over Reasoning in Large Language Models Through Subtle Context ShiftsSana Hassanhttps://www.marktechpost.com/author/sana-hassan/T* and LV-Haystack: A Spatially-Guided Temporal Search Framework for Efficient Long-Form Video UnderstandingSana Hassanhttps://www.marktechpost.com/author/sana-hassan/Unveiling Attention Sinks: The Functional Role of First-Token Focus in Stabilizing Large Language ModelsSana Hassanhttps://www.marktechpost.com/author/sana-hassan/RARE (Retrieval-Augmented Reasoning Modeling): A Scalable AI Framework for Domain-Specific Reasoning in Lightweight Language Models

For a limited time, Amazon is offering a Hoto 3.6V Electric Screwdriver for $35.99 after 40% off coupon code "40SZOFEK". Make surenot to clip any of the coupons on the product page, otherwise this code will not apply and you'll be stuck with a lesser discount. This is a great, compact screwdriver for anyone who tinkers with small electronics and deals with dozens of tiny screws, but wants something stronger than one of those "mini precision" screwdrivers.40% Off Hoto 3.6V 4Nm Electric Driver for $35.99Hoto 3.6V 4Nm Electric ScrewdriverHoto sells a few electric screwdriver models, and this is their most powerful model. Most of their electric screwdrivers are very small - about the thickness of a pen flashlight - and deliver about 0.2Nm-0.5Nm of max torque. This particular model feels heftier - about the thickness of a regular flashlight - and delivers up to 4Nm of max torque. It also has a higher rotational speed of 220RPM (vs 150RPM on the smaller models) and a far bigger 1,500MaH battery (vs 300mAh or less on the smaller models). It's more versatile since it will handle bigger and tougher jobs, however that means it's also powerful enough to strip screws, so you'll have to be more careful.Although this screwdriver isn't the smallest, it's still compact and weighs in at just over half a pound. It fits neatly in the provided aluminum box, which also houses a selection of 12 different bits. In case you lose any and need replacements, the screwdriver is compatible with standard sized bits. There's even a handy LED ring to illuminate whatever it is you're driving and it will recharge over USB Type-C.Why Should You Trust IGN's Deals Team?IGN's deals team has a combined 30+ years of experience finding the best discounts in gaming, tech, and just about every other category. We don't try to trick our readers into buying things they don't need at prices that aren't worth buying something at. Our ultimate goal is to surface the best possible deals from brands we trust and our editorial team has personal experience with. You can check out our deals standards here for more information on our process, or keep up with the latest deals we find on IGN's Deals account on Twitter.Eric Song is the IGN commerce manager in charge of finding the best gaming and tech deals every day. When Eric isn't hunting for deals for other people at work, he's hunting for deals for himself during his free time.

The sky is on fire. A streak of silver cuts through the clouds. A towering alien arrives to silently watch. Galactic is coming. Even today, those moments that Jack Kirby and Stan Lee created for 1966’s Fantastic Four #48 – 50 still amaze readers, still set the standard for all superhero epics. So with Fantastic Four: First Steps pitting Marvel‘s first family against the Devourer of Worlds, expectations couldn’t be higher. First Steps director Matt Shakman knows about these expectations too, and is doing what he can to meet them. “I didn’t want to just use motion-capture for Galactus,” Shakman recently told Entertainment Weekly. “I wanted to actually have someone there embodying the part.” Yes, that’s right, Galactus won’t just be a CGI creation voiced by Ralph Ineson’s deep English brogue. He will be a person on the set, played by Ineson in a costume around that deep, marvelous English brogue. Of course given that Galactus is the size of a building, but must still interact with heroes Reed Richards (Pedro Pascal) and Sue Storm (Vanessa Kirby), this approach presents problems. “We’ve built an entire costume for him, and we’ve done a lot of photography testing to figure out,” admitted Shakman. “How do you make sure that the scale is correct? How do you film Mount Rushmore?” Difficult as the process surely is for Shakman and his team, the fans certainly appreciate it. The previous attempt at visualizing Galactus in 2007’s Fantastic Four: Rise of the Silver Surfer remains one of the most embarrassing moments in superhero movie history. For all the warnings provided by the Silver Surfer (voiced by Laurence Fishburne and physically performed by Doug Jones) about the threat posed by Galactus, the being only appears in the form of a cloud, which only takes the shape of his signature helmet. Or perhaps he is a great Kirby-esque being hidden inside a cloud? It’s intentionally vague. Although irritating, the Galactus cloud fell in line with the approach taken by most superhero movies of that era. Use the names and basic concepts from the comics, but strip away everything else. Thus we get Batman in body armor blowing up buildings full of people and the X-Men in black leather. All of that changed when Kevin Feige inaugurated the MCU in 2008, making movies that embraced comic book wackiness. More recently, however, Marvel’s garnered a reputation for shoddy aesthetics, making movies and television shows that fall far short of the source material’s visuals despite spending millions and overworking animators past ethical limits. For that reason, it’s easy to imagine that the Galactus who appears in the finished version of First Steps would be much like the Galactus who shows up at the end of the sizzle reel footage leaked last summer. Giant and comics-accurate, complete with the improbable purple headpiece designed by Kirby back in the mid-’60s, and all rendered in CGI. It’s easy to imagine a final stand-off not unlike those in Captain America: Brave New World or Shang-Chi and the Legend of the Ten Rings, with actors making faces as they float weightlessly through badly-composited artificial environments. To be sure, that might still be the case. But at least Shakman’s making Galactus a real person in a real suit on a real set. One of many, in fact, according to the EW report. Marvel has built practical sets for Reed’s high-tech lab and for Yancy Street, the Thing’s New York neighborhood. Ebon Moss-Bachrach wears a costume as the Thing, and the team’s robot assistant H.E.R.B.I.E. is a remote controlled machine that interacts with the actors. Honestly, it wouldn’t take much to improve on the previous four Fantastic Four live action films, all of which have their shortcomings. But the combination of a comic book accurate Galactus and practical sets gives us hope that the World Devourer’s arrival will be as terrifying as it has been in our imaginations. Fantastic Four: First Steps releases in theaters on July 25, 2025. Join our mailing list Get the best of Den of Geek delivered right to your inbox!