CPU Consideration No. 1: Goals and UpgradesIt's important to set a goal for what you want to achieve with a change of desktop CPUs. You want better performance, but are you looking for gaming performance, specifically? Is it for a system you will use for day-to-day work? Or do you simply have an older computer sitting around that you'd like to spruce up for internet use and watching videos? We can't stress enough the importance of knowing what you want out of your system, to make sure you don't end up with a disappointing amount of horsepower in the end, whether that's due to overspending or underspending. This will also largely help you to determine if you want to upgrade your PC or build a new one.(Credit: Michael Justin Allen Sexton)First, we should be clear that upgrading is often a viable option, especially if you just want the system for light-duty use like web browsing. Technically, even antique PCs based on AMD's Phenom, Phenom II, and FX series are still suitable for this task, as are Intel's Core 2 Duo and Core 2 Quad systems, but these relics would certainly show a good deal of slowdown if you pushed them very hard. These aforementioned chips certainly aren't worth upgrading or purchasing now—far better options are available for the money.If your computer is relatively modern, though, it might be better for you to upgrade your current PC instead of buying a new one. This makes your choice of processor far simpler, as there will be a limited number of chips that will work with your current system. Upgrading isn't always an option, as you might already have the best CPU that your current motherboard supports. But as long as that's not the case, upgrading can be the most affordable way to end up with a faster PC, as you won't need to purchase other components.As to whether it's worthwhile to upgrade your system, a good rule of thumb is to consider the age of the system you're using. If that system was made within the last 10 years and doesn't already have the fastest CPU available for it (or close to it), upgrading may get you the performance boost you desire. Nothing older than this is remotely worth upgrading, as it becomes more practical to save for a newer system instead. And even with a system that's around 10 years old, it's only worth considering if you have something like an old Intel Core i3 or worse, and you are upgrading to an old Core i7—but this is also only worth considering if you can find that latter processor for cheap.A quick search shows that for older PCs based on Intel's LGA 1155 socket, which was the platform used by Intel's second- and third-gen Core "Sandy Bridge" and "Ivy Bridge" processors (now more than a decade old), you can buy second-hand Core i7-3770 processors for around $30 to $50 from take-your-chances sources like eBay. This processor today would lag behind a modern Core i3, but it would still be capable of web browsing or even use in a low-end gaming PC. Again, we must stress, however, that this sort of upgrade only makes sense if you have one of the slowest CPUs available for that old platform and are upgrading to one of the fastest ones. If you had a Sandy Bridge Core i5, it would make more sense to save for a new system instead of upgrading, since the ceiling for the best chip available on that platform isn't much higher.Upgrading more recent systems also follows this same logic, i.e. it only makes sense to upgrade if you are making a substantial jump on the same platform—and you can do so cheaply. Even on an almost new system, we would hesitate to upgrade from, say, an AMD Ryzen 5 9600X to an AMD Ryzen 7 9700X, or from an Intel Core Ultra 5 245K to an Intel Core Ultra 9 285K. It's not that you would not see a noticeable and measurable performance boost (our reviews show that there clearly would be one), but the gains you receive for the cost of upgrading often don't make it worthwhile. This makes it exceedingly important that you buy the right processor from the start, rather than waiting and trying to upgrade later.If your goal is to simply fix up your old PC for web browsing or low-end gaming, then you may want to stop reading here and see if an affordable upgrade option is available for your system. Each motherboard, including the one in your PC, has a list of supported processors that you should be able to find on the manufacturer's product page. If none of the processors supported on your current motherboard is affordably priced or a significant step up, then just forget about upgrading and start thinking about building or buying a new system instead. If your goal was instead to get a faster system for work or high-end gaming, or if upgrading just won't cut it for you, then keep reading.System Planning 101: How to Pick a Motherboard, a CPU, and RAM That Work TogetherIf you're going to build a new system, then you'll need to pick mostly new parts. If you had an older system, you might be able to re-use a few items, notably the power supply, case, or storage device, but everything else you'll likely have to buy. Some parts, like the ones we just listed that you could reuse, are compatible with a wide range of systems and can last for several generations. But motherboards, CPUs, and main system memory (RAM) are the opposite.(Credit: Joseph Maldonado)These items will only last for a few generations, at most. All motherboards have a socket that the CPU is designed to be mounted in, and at the chip maker's level, that socket's design has to be changed every so often to allow for the addition of new features. As a result, motherboards only support CPUs that will fit in (and are explicitly compatible with) their sockets, and the same is true vice versa for CPUs. (Sometimes a chip will physically fit, but it isn't supported.) RAM, similarly, has a custom set of slots that it mounts on a PC motherboard, and these slots only support one broad type of RAM and won't work with any others. (The latest types are known as DDR4 and DDR5.)(Credit: Joseph Maldonado)Due to sockets and, thus, motherboard platforms changing every few years, you will typically need to buy a new motherboard and, possibly, a new set of RAM if you want to buy a new CPU that succeeds your current one by more than a few years. The only time this won't be the case is if you are upgrading on the same platform, like in the possible scenarios discussed above. If you upgrade frequently, your old RAM might remain compatible with your new system. In that vein, when buying parts, you'll want to make sure the parts you get will be compatible with each other. CPU sockets are typically numbered; for mainstream CPUs bought by most consumers, the latest are AMD's AM4 and AM5, and Intel's LGA 1700 and LGA 1851, and both motherboards and processors will be identified by this number to make picking a matching pair easy. Still, you'll want to make certain that a given CPU is explicitly supported by a given board; verify that compatibility on the board maker's website.(Credit: Joseph Maldonado)For several years now, RAM has also followed a simple numbering scheme as we've progressed from DDR to DDR2, DDR3, DDR4, and now DDR5. None of these will work in RAM slots made for the other, and this is also detailed on the motherboard specs page, which again makes finding a matching compatible part relatively easy.Power, Now! Don't Build With Plans of UpgradingOften, people consider building a PC with plans to upgrade it in the future. The idea is that if you get a motherboard and CPU now and can later upgrade to a faster CPU, this might help you avoid having to build another computer and save costs down the road the next time you feel your PC slowing down.The problem with this train of thought? It rarely works out as planned. Intel, over the last decade, has changed to a new CPU socket every two or three years, typically once every two CPU generations. AMD postpones changing sockets longer, and it only recently introduced its AM5-socketed motherboards as a successor to its AM4 motherboards. AM4 was first launched in 2017, giving it a five-year reign during which time AMD pushed out four primary generations of processors for the platform. We don't know how long AM5 will be AMD's main platform at this time, but there's good reason to believe it will be used for roughly the same period as AM4 was.(Credit: Joseph Maldonado)Even with AMD's extended platform lifespan, you aren't overly likely to upgrade processors on the same motherboard. Compatibility remains a question mark even with AM4, as not all processors made for the platform are supported on all motherboards. AMD made efforts, especially toward the end of AM4's reign of dominance, to encourage board makers to push out updated BIOS versions for older motherboards to extend support for their newest AM4 CPUs to older motherboards. But this effort still relied on AMD's board partners to implement the new BIOS versions on dozens or even hundreds of boards. Long story short: You cannot buy a motherboard today and be absolutely confident it will support processors released a few years down the road, even if they are released for the same socket as the motherboard you are buying today.Even if you can upgrade the CPU on an existing board, strong reasons may arise why you might not want to. Newer motherboards with updated chipsets have their inherent benefits. These might include things like support for faster RAM, faster USB and storage connections, faster PCI Express slots and underlying buses, faster internet support, and myriad other potential benefits. Plus, after you upgrade, you'll be left with an orphaned CPU to then resell or put into a new motherboard. It starts to make a lot more sense to just think about selling (or giving away) your old PC or CPU/motherboard combo, and building new, most of the time, when you need something faster.If you are buying into an Intel platform, this is doubly true. Unless you buy a low-end CPU, to begin with, you aren't likely to want to upgrade in just one or two years before Intel introduces a whole new platform. Though there is some merit in the idea of doing just that (buying a low-end CPU to upgrade next year when high-end prices drop), it still raises overall costs, as you now have a low-end CPU you paid for in addition to a high-end CPU. You'd be better off just buying that high-end CPU to begin with and enjoy it that much longer.All in all, though some may view platforms that are aging out or about to be replaced as dead ends, it's best to view all platforms in this same frame of mind. That's not to say that none of them is worth buying (that's not true), but upgradability shouldn't be first on your mind when selecting a motherboard and CPU to use in your new PC. Some things can be easily upgraded later, like the RAM, storage, or graphics card, but this is not an advantage that motherboards or processors have.When you do buy a system, unless you are getting an exceptional deal, you should still buy into the newest platform available with the best processor you can afford that fits your needs. Currently, Intel's newest platform is the LGA 1700 platform, and AMD's (as mentioned above) is AM5. There's no harm in getting an AM4 system or a last-gen Intel LGA 1200 system, but, unless you're getting those steeply discounted from their launch prices, you're going to get a better system and more bang for your buck by going with what's newest.Buying Basics: Four Key Concepts to Know About CPUsNow that we've covered the basic considerations of whether to upgrade or buy new and the other hardware you'll need to consider when buying a CPU, let's talk about what differentiates one processor from another. A multitude of factors need to be considered here, but the most important ones are the microarchitecture, core count, thread count, and clock speed.MicroarchitectureThis is by far the most important aspect of any CPU, and indeed the same is true for just about any piece of technology. It's the design and internal blueprints of how the device is built and what makes it work. Due to differences in architecture, you can have multiple CPUs that operate at the same speed but still perform drastically differently. It might be helpful for you to picture each architecture in a similar vein as you might consider different makes of cars or airplanes. In the same way that you can have multiple cars with their engines running at 2,000rpm and get drastically different performance and speeds out of the cars, the same is true for CPUs.(Credit: Joseph Maldonado)Judging architectures is exceedingly difficult, as they are incredibly complex, with billions of transistors and a multitude of other external factors that bear on their performance. To gain an idea of how different architectures perform, you should read our reviews that touch on the subject and also compare processors utilizing different architectures against each other.(Credit: Michael Justin Allen Sexton)An important detail about microarchitectures that you should know, and that is easy to understand, is how to identify products based on different architectures. This is surprisingly easy if you learn the numbering systems used by AMD and Intel. If we take AMD's Ryzen 9 9950X as an example, for desktop chips the numbers break down like this...Recommended by Our Editors9 = Generational number. AMD's microarchitectures of recent years are named "Zen," followed by a number. All modern AMD desktop processors that start with a "9" utilize the Zen 5 microarchitecture, and all processors that start with a "7" utilize the Zen 4 microarchitecture. All modern AMD processors that start with a "5" utilize the Zen 3 microarchitecture. AMD has no desktop processors that begin with a "6"; these were mobile only. There were a few desktop Ryzen CPUs that begin with an "8", and these also run the Zen 4 microarchitecture. 9 = Product tier. AMD makes Ryzen 3, 5, 7, and 9 processors, with this numeral included after "Ryzen" in the product name and as the second digit in the product number.5 = Sub-product tier placement.0 = Sub-product tier placement. Rarely used.X = "Extreme" or higher-end variant of a processor. An AMD chip may or may not have a non-"X" variant.(Note that AMD is moving to a wholly different processor numbering scheme for its very latest mobile CPUs. See the details here; the guidelines above will not apply to them.)Traditionally, Intel used a similar coding scheme for its desktop processors that's still partly in use. Let's take an Intel high-end CPU, the Core i9-14900K, as an example...14 = Generational number. Number "14" is assigned to Intel's 14th Gen "Raptor Lake Refresh" processors. 13 = 13th Gen "Raptor Lake." 12 = 12th Gen "Alder Lake." 11 = 11th Gen "Rocket Lake." 10 = 10th Gen "Comet Lake."9 = Product tier. Intel makes Core i3, i5, i7, and i9 processors, which are included after the "Core" in the product name and as the third digit in the product number (or the second digit, in processors older than 10th Gen).0 = Sub-product tier placement.Second 0 = Sub-product tier placement.K = Indicates a higher-end variant of a processor with overclocking support. Several other "suffix" letters are used by Intel to indicate other differences, but the most common ones you will encounter are "K" (indicating overclockability), "F" (indicating that the CPU has no integrated graphics processor), and "KF" (indicating both aspects).With Intel's new "Arrow Lake" desktop processors, Intel has adopted a new naming scheme that diverges somewhat from previous generations, but it remains similar. The Core i9, i7, and i5 of previous generations are gone, replaced by Core Ultra 9, Core Ultra 7, and Core Ultra 5 families, respectively. The numbering scheme after these new names has been reduced to three digits, plus optional letter suffixes, which remain largely unchanged from previous generations. Here's an example of how the new naming scheme works, using the flagship Core Ultra 9 285K chip as an example.2 = Generational number. The number "2" is assigned to Intel's Arrow Lake processors. No desktop chips were assigned the number "1," and no additional desktop chips have used this naming sequence yet.8 = Relative performance differentiator. A higher number here general suggests higher performance in relation to other processors in the same product line. All Ultra 9 CPUs have an "8" here, while all Ultra 7 processors have a "6" here. Ultra 5 CPUs can have a "4," "3" or "2" in this position. This number is not an absolute, however, as ultra-efficiency models like the Ultra 9 285T could potentially be outpaced by chips like the Ultra 7 265K that support higher power ratings.5 = This digit currently holds little meaning; all Arrow Lake desktop processors so far have a "5" as the third digit. We'll see if that changes down the line.K = Indicates a higher-end variant of a processor with overclocking support. Several other "suffix" letters are used by Intel to indicate other differences, but the most common ones you will encounter are "K" (indicating overclockability), "F" (indicating no integrated graphics processor), and "KF" (indicating both aspects). Finally, the "T" suffix is used for desktop processors that are optimized for power efficiency and lower heat output. Some Arrow Lake desktop chips will have no letter at all at the end, terminating just in the "5." Using these numbers, you can compare processors of the same generation within a vendor's own line in terms of relative performance with some accuracy. It's always best to check reviews when possible and compare other details about the processor to more accurately compare CPUs.Core CountInside all mainstream desktop processors today are multiple CPU cores. In the past, processors only had one CPU core, but as technology has improved, more cores have been manufactured into processor silicon to increase performance. Each CPU core operates as a semi-independent component inside the processor and is capable of completing tasks.The advantage of having more CPU cores is simply that you can get more work done at a time. It wouldn't be too much of a stretch to think of the CPU cores as workers and the processor as the building in which they work.(Credit: Joseph Maldonado)Traditionally, all of these CPU cores would have been identical, but this changed starting with Intel's 12th Gen Alder Lake processors. Intel now employs two different types of CPU cores inside most of its desktop processors. Processors based on this design have what Intel calls "P-cores," which are built on a high-performance microarchitecture. Alongside the "P-cores," Intel also adds "E-cores," which utilize a different microarchitecture that enables these cores to be physically smaller and more energy-efficient.If we return to our comparison of CPU cores to employees, you could think of Intel's P-cores as higher-level workers with larger offices who can complete more work due to their more extensive experience and larger workspaces. At the same time, the E-cores could be thought of as lower-level workers with smaller offices to work in and lesser skill sets. They may get less done in a given period of time, but you can cram more of them in for the money, and they take up less space.Though the E-cores are slower, they still significantly improve the processor's performance. AMD hasn't adopted a similar scheme yet, which means all of the cores in an AMD processor are identical.(Credit: Michael Justin Allen Sexton)Core count contributes greatly to a processor's overall performance, but this alone does not determine whether one processor is faster than the other. It's entirely possible for a quad-core processor to be faster than an octa-core processor, and vice versa.MultithreadingWork that needs to be performed on the CPU cores comes into the processor in a somewhat chaotic fashion. Some processors take the work orders as they come in and simply proceed to work on them in the order they were received. Processors that do this implement what we know as an In-Order execution design. This has been shown to hamper a processor's overall performance.Each work order sent to the processor requires instruction information and raw data for the work to be completed. When an In-Order CPU core goes to work on a task, it must have both of these; otherwise, it will sit and wait while the needed instruction information or data is fetched.Processors that implement an "Out-of-Order" or "OoO" execution design largely circumvent this issue by re-ordering tasks as they come in. They can place work orders that have everything they need to run ahead of work orders that don't. Inevitably, some work orders still get to the CPU cores without everything they need, though, and this still leads to a stall while the required data gets fetched.That is, unless a processor implements simultaneous multithreading (SMT, for short). Essentially, what this technology does is open up a second line for work orders to come into the CPU. The processor isn't able to work on two work orders at once, so when everything is running smoothly, the processor continues to run through work orders from each line in order, back and forth. When a stall occurs, however, SMT technology enables the processor to set the stalled work order aside and work on items in the other line until the required data is fetched.This technology significantly reduces processor stalls and drastically improves performance. Processors that implement SMT will show as having two threads for each CPU core that supports it. This is why an AMD processor with eight cores and SMT technology will be touted as a 16-thread processor. The same is true for Intel processors, with the most notable difference being that Intel calls SMT "Hyper-Threading" on its processors.Just like with core count, thread count doesn't tell you enough to determine which processor is best, but it can give you an idea as to which processor is better in a given line. A processor with more threads may well have a performance advantage over one that has fewer threads supported, in applications that can take advantage of the technology. But as we said with the core count, all of these factors need to be taken into consideration to know for sure.Clock SpeedThe last and easiest-to-understand key defining characteristic of a processor is its clock speed. This directly relates to a processor's overall speed and is measured in hertz. Processors today are so fast that this is typically reported in gigahertz (GHz).(Credit: AMD)A processor's clock speed is sometimes reported as the total number of operations that it can perform at any given second. For example, a one gigahertz processor can theoretically perform 1,000,000,000 operations each second. Modern CPUs operate at multiple GHz with some, like Intel's Core i9-14900K, peaking at 6.0GHz.In truth, this description is inaccurate, as some operations require multiple clock cycles (multiple hertz) to complete, and this is where architecture comes in, coming full circle. When comparing processors that are part of the same generation and product line, it's safe to think the one with the most cores and the highest clock speed will perform the best. Comparing across different architectures and product lines, however, this is not always the case.On modern processors, you'll often see a base clock listed, as well as a "Boost Clock" or "Turbo Clock." You can essentially ignore the base clock listing if you see either of these other clocks listed. A base clock is a processor's true baseline speed, but modern CPUs are designed to increase their clock speed to a point, as long as the right conditions are met; those conditions are based on thermal and power-draw limitations.Modern processors run at these elevated clock speeds most of the time when under a heavy load, which is why these boosted numbers are far more important for determining performance than the base clock. It is also possible to increase the peak clock speed yourself on some processors by what is called overclocking, but that's for another guide.To overclock, you'll need a lot more specific information on the topic, beyond the scope of this article. You'll also need special hardware for overclocking (an appropriate motherboard, and robust CPU cooling). We point out which processors can overclock in our reviews, and while talking about specific products on this page, so you will know which to buy if you want to try your hand at overclocking. But you should make sure you do plenty of research on the topic first, as overclocking can be hazardous to your components.How to Buy the Right CPU: Final Buying AdviceThink carefully about your computing needs and pick a processor that fits them. If you have extra cash to spare and want to be extra safe, buy one a little better than your target chip.One thing we would recommend against is buying the most expensive processor that money can buy or that you can afford without thinking through the decision. Yes, more-expensive processors are typically better, in general—AMD and Intel charge more for them for a reason—but that doesn't mean you need that level of performance.No matter how performant a processor you buy, you will eventually want to upgrade to something newer and faster. Though you may be able to prolong the life of your PC by buying a newer and faster CPU, it might make better sense to opt for upgrading more frequently.The rate at which technology has been improving makes it likely that, in five years or so, you'll be able to buy a midrange processor that will be just as fast—if not faster than—today's very fastest processors. If you buy or build a new midrange PC every five years or so, you'll likely have a faster computer than someone who is still running a five-year-old PC with a high-end processor.Upgrading more often may not be better financially in the long term, as there is a definite cost and set of diminishing returns involved, but it does get you the improvements that come with changing platforms. As PC components eventually fail, it can also help to avoid unexpected downtime from old parts breaking. For further guidance, check our processor reviews for more details on each CPU. Source: https://me.pcmag.com/en/processors/14418/the-best-cpus-for-2023 #the #best #cpus #for #uae #and #saudi #arabia

News Hogwarts Legacy Trailer Shows Switch vs Switch 2 Graphics Comparison WB Games' open world action RPG will be a launch title for the Nintendo Switch 2 when the console releases worldwide on June 5. Posted By Shubhankar Parijat | On 14th, May. 2025 The Nintendo Switch 2 is set to launch with a couple of some pretty major heavy-hitting games in under a month’s time, but at the same time, its launch lineup is also going to boast several third party ports and re-releases. One of those is going to be Hogwarts Legacy, for which a new gameplay trailer has been uploaded to offer glimpses of how much better its visuals are on the Switch 2 compared to the Switch.  The trailer showcases a number of shots of the game, showing both indoor and outdoor environments, and the difference is, quite frankly, massive. From the lighting and the reflections to the density and detail in the environments and the draw distances, the open world action RPG is looking like a significant technical and graphical leap over the game’s Nintendo Switch version. Check out the comparison gameplay trailer below.  Hogwarts Legacy is available on PS5, Xbox Series X/S, PS4, Xbox One, Nintendo Switch, and PC, and launches for Nintendo Switch 2 on June 5.  Tagged With: Elden Ring: Nightreign Publisher:Bandai Namco Developer:FromSoftware Platforms:PS5, Xbox Series X, PS4, Xbox One, PCView More DOOM: The Dark Ages Publisher:Bethesda Softworks Developer:id Software Platforms:Xbox Series X, PS5, PCView More Death Stranding 2: On the Beach Publisher:Sony Developer:Kojima Productions Platforms:PS5View More Amazing Articles You Might Want To Check Out! PS5 Hits 77.8 Million Units Shipped Worldwide The console shipped 2.8 million units worldwide in the final quarter of the 2024-25 fiscal year, with total sh... Hogwarts Legacy Trailer Shows Switch vs Switch 2 Graphics Comparison WB Games' open world action RPG will be a launch title for the Nintendo Switch 2 when the console releases wor... Clair Obscur: Expedition 33 Dev Says a Nintendo Switch 2 Port “Could be Interesting” French studio Sandfall Interactive could be open to bringing its acclaimed JRPG to Nintendo's upcoming next-ge... The Super Mario Bros. Movie Sequel Could be Titled Super Mario World Nintendo, Universal, and Illumination's next Super Mario movie could be named after the beloved 1991 2D sidesc... The Wait for Grand Theft Auto 6’s Next Trailer is Going to Feel Excruciatingly Long GTA 6's newest trailer was a great one, but it has left us wanting more- which we're not getting for a while y... The Texas Chain Saw Massacre Winds Down Support, Won’t Receive Any More DLC The next big patch arrives this month, but the asymmetrical multiplayer title will remain playable and purchas... View More Source: https://gamingbolt.com/hogwarts-legacy-trailer-shows-switch-vs-switch-2-graphics-comparison #hogwarts #legacy #trailer #shows #switch #graphics #comparison

In this tutorial, we lean hard on Together AI’s growing ecosystem to show how quickly we can turn unstructured text into a question-answering service that cites its sources. We’ll scrape a handful of live web pages, slice them into coherent chunks, and feed those chunks to the togethercomputer/m2-bert-80M-8k-retrieval embedding model. Those vectors land in a FAISS index for millisecond similarity search, after which a lightweight ChatTogether model drafts answers that stay grounded in the retrieved passages. Because Together AI handles embeddings and chat behind a single API key, we avoid juggling multiple providers, quotas, or SDK dialects. !pip -q install --upgrade langchain-core langchain-community langchain-together faiss-cpu tiktoken beautifulsoup4 html2text This quiet (-q) pip command upgrades and installs everything the Colab RAG needs. It pulls core LangChain libraries plus the Together AI integration, FAISS for vector search, token-handling with tiktoken, and lightweight HTML parsing via beautifulsoup4 and html2text, ensuring the notebook runs end-to-end without additional setup. import os, getpass, warnings, textwrap, json if "TOGETHER_API_KEY" not in os.environ: os.environ["TOGETHER_API_KEY"] = getpass.getpass("🔑 Enter your Together API key: ") We check whether the TOGETHER_API_KEY environment variable is already set; if not, it securely prompts us for the key with getpass and stores it in os.environ. The rest of the notebook can call Together AI’s API without hard‑coding secrets or exposing them in plain text by capturing the credentials once per runtime. from langchain_community.document_loaders import WebBaseLoader URLS = [ "https://python.langchain.com/docs/integrations/text_embedding/together/", "https://api.together.xyz/", "https://together.ai/blog" ] raw_docs = WebBaseLoader(URLS).load() WebBaseLoader fetches each URL, strips boilerplate, and returns LangChain Document objects containing the clean page text plus metadata. By passing a list of Together-related links, we immediately collect live documentation and blog content that will later be chunked and embedded for semantic search. from langchain.text_splitter import RecursiveCharacterTextSplitter splitter = RecursiveCharacterTextSplitter(chunk_size=800, chunk_overlap=100) docs = splitter.split_documents(raw_docs) print(f"Loaded {len(raw_docs)} pages → {len(docs)} chunks after splitting.") RecursiveCharacterTextSplitter slices every fetched page into ~800-character segments with a 100-character overlap so contextual clues aren’t lost at chunk boundaries. The resulting list docs holds these bite-sized LangChain Document objects, and the printout shows how many chunks were produced from the original pages, essential prep for high-quality embedding. from langchain_together.embeddings import TogetherEmbeddings embeddings = TogetherEmbeddings( model="togethercomputer/m2-bert-80M-8k-retrieval" ) from langchain_community.vectorstores import FAISS vector_store = FAISS.from_documents(docs, embeddings) Here we instantiate Together AI’s 80 M-parameter m2-bert retrieval model as a drop-in LangChain embedder, then feed every text chunk into it while FAISS.from_documents builds an in-memory vector index. The resulting vector store supports millisecond-level cosine searches, turning our scraped pages into a searchable semantic database. from langchain_together.chat_models import ChatTogether llm = ChatTogether( model="mistralai/Mistral-7B-Instruct-v0.3", temperature=0.2, max_tokens=512, ) ChatTogether wraps a chat-tuned model hosted on Together AI, Mistral-7B-Instruct-v0.3 to be used like any other LangChain LLM. A low temperature of 0.2 keeps answers grounded and repeatable, while max_tokens=512 leaves room for detailed, multi-paragraph responses without runaway cost. from langchain.chains import RetrievalQA qa_chain = RetrievalQA.from_chain_type( llm=llm, chain_type="stuff", retriever=vector_store.as_retriever(search_kwargs={"k": 4}), return_source_documents=True, ) RetrievalQA stitches the pieces together: it takes our FAISS retriever (returning the top 4 similar chunks) and feeds those snippets into the llm using the simple “stuff” prompt template. Setting return_source_documents=True means each answer will return with the exact passages it relied on, giving us instant, citation-ready Q-and-A. QUESTION = "How do I use TogetherEmbeddings inside LangChain, and what model name should I pass?" result = qa_chain(QUESTION) print("n🤖 Answer:n", textwrap.fill(result['result'], 100)) print("n📄 Sources:") for doc in result['source_documents']: print(" •", doc.metadata['source']) Finally, we send a natural-language query through the qa_chain, which retrieves the four most relevant chunks, feeds them to the ChatTogether model, and returns a concise answer. It then prints the formatted response, followed by a list of source URLs, giving us both the synthesized explanation and transparent citations in one shot. Output from the Final Cell In conclusion, in roughly fifty lines of code, we built a complete RAG loop powered end-to-end by Together AI: ingest, embed, store, retrieve, and converse. The approach is deliberately modular, swap FAISS for Chroma, trade the 80 M-parameter embedder for Together’s larger multilingual model, or plug in a reranker without touching the rest of the pipeline. What remains constant is the convenience of a unified Together AI backend: fast, affordable embeddings, chat models tuned for instruction following, and a generous free tier that makes experimentation painless. Use this template to bootstrap an internal knowledge assistant, a documentation bot for customers, or a personal research aide. Check out the Colab Notebook here. Also, feel free to follow us on Twitter and don’t forget to join our 90k+ ML SubReddit. Asif RazzaqWebsite |  + postsBioAsif Razzaq is the CEO of Marktechpost Media Inc.. As a visionary entrepreneur and engineer, Asif is committed to harnessing the potential of Artificial Intelligence for social good. His most recent endeavor is the launch of an Artificial Intelligence Media Platform, Marktechpost, which stands out for its in-depth coverage of machine learning and deep learning news that is both technically sound and easily understandable by a wide audience. The platform boasts of over 2 million monthly views, illustrating its popularity among audiences.Asif Razzaqhttps://www.marktechpost.com/author/6flvq/Agent-Based Debugging Gets a Cost-Effective Alternative: Salesforce AI Presents SWERank for Accurate and Scalable Software Issue LocalizationAsif Razzaqhttps://www.marktechpost.com/author/6flvq/A Step-by-Step Guide to Deploy a Fully Integrated Firecrawl-Powered MCP Server on Claude Desktop with Smithery and VeryaXAsif Razzaqhttps://www.marktechpost.com/author/6flvq/OpenAI Releases HealthBench: An Open-Source Benchmark for Measuring the Performance and Safety of Large Language Models in HealthcareAsif Razzaqhttps://www.marktechpost.com/author/6flvq/PrimeIntellect Releases INTELLECT-2: A 32B Reasoning Model Trained via Distributed Asynchronous Reinforcement Learning Source: https://www.marktechpost.com/2025/05/14/step-by-step-guide-to-build-a-fast-semantic-search-and-rag-qa-engine-on-web-scraped-data-using-together-ai-embeddings-faiss-retrieval-and-langchain/ #stepbystep #guide #build #fast #semantic #search #and #rag #engine #webscraped #data #using #together #embeddings #faiss #retrieval #langchain

The award-winning painter and illustrator explains how to improve your technical skill set with colour and composition. Source: https://www.creativebloq.com/art/traditional-art/greg-manchesss-10-tips-to-improve-colour-and-composition-in-your-art #discover #greg #manchess039s #tips #for #more #dynamic #paintings

In the quiet streets of Tottenham, London, a modest 1970s clergy house has been reimagined as an environmentally conscious family home, cloaked in the textured warmth of cork. Known as Nina’s House, the transformation was led by design studio Nina+Co in partnership with ROAR Architects, demonstrating how regenerative materials and thoughtful design can be used to turn a passive facade into an engaging and high-performing community landmark. At the heart of this “deep retrofit” is cork, a material considered as poetic in its physical presence as it is powerful in its technical performance. Once cold and uninspiring, the building has been made to radiate warmth, curiosity, and intention, as it has been covered entirely in expanded cork panels. Dark, mottled, and tactile, this natural cladding is made to invite the hands of schoolchildren, the questions of neighbors, and the admiration of passersby. Designer: Nina+Co and ROAR Architects The use of cork in architecture represents a shift toward biomaterials that are both high-performing and sensorially engaging. As a material, cork combines sustainability with durability, offering a renewable alternative to synthetics while delivering superior insulation. It reduces the building’s embodied carbon and enhances its energy efficiency, making it a strategic material choice for climate-resilient design. Cork, which is harvested from the bark of cork oak trees, is known as a renewable and biodegradable material with strong insulation properties. In Nina’s House, expanded cork was chosen for its regenerative farming practices, zero synthetic binders, and minimal build-up requirements; no plastic membranes or toxic adhesives were included. Its low-impact manufacturing process uses only heat to expand the cork, further emphasizing its role in environmentally responsible construction. According to Nina Woodcroft, founder of Nina+Co, cork stands out as a natural material that is not only effective but also appeals to the senses. She emphasized her fascination with cork’s scent, feel, and sustainable story, noting that this project used cork for both function and visual impact. The house, she explained, was intentionally wrapped in a breathable and insulating material to offer both environmental performance and gentle street presence. Cork’s application at Nina’s House challenges the misconception that natural materials are limited to rustic or niche aesthetics. Instead, it demonstrates how cork can serve as a contemporary architectural finish, one that ages gracefully and contributes to long-term resilience. Rather than extend the home, the existing footprint was maximized. The ground floor layout was opened up, and the small garage and porch were brought into the thermal envelope, so that a sense of flow and cohesion was created. Budget was directed toward infrastructure: the gas supply was removed, underfloor heating and an air-source heat pump were installed, airtightness was improved, and all glazing was replaced with high-performance alternatives, including suspended film technology and triple glazing. Expanded cork was made to play a starring role both outside and in. On the facade, it is used as an insulating skin; inside, cork can also be seen in flooring, while walls have been finished with lime-based plaster containing cork granules. In areas where external insulation was not possible, these lime plasters were applied to provide an internal thermal buffer. The decision to leave some internal surfaces raw and sealed with a non-toxic glaze was made both for stylistic and economical reasons, so that practicality could be blended with beauty. Cork’s versatility enables it to be used beyond cladding. Its acoustic properties make it ideal for interior finishes, especially in homes that aim to be peaceful and meditative. It also helps regulate humidity and temperature indoors, supporting healthier living environments. Throughout the house, reclaimed and local materials have been used to reflect a low-waste, high-value design ethos. Pine floorboards that were removed for structural work were reworked into ceiling cladding. A statement kitchen island, designed by Woodcroft, was crafted by local artisans from a London Plane tree that had been felled in Soho Square, an urban tree that was given new life by the organization Fallen & Felled. The kitchen also features recycled plastic countertops, by which the home’s dedication to circularity is echoed. All joinery was made ultra-locally using British Douglas Fir tri-ply board, thereby showing how sustainable design can be used to support local economies while reducing carbon impact. The cork cladding is expected to last between 15 to 25 years. When its surface has weathered, lime render will be applied over it, so that its life can be extended while its appearance is subtly altered. This long-term approach reflects a preference for resilience over superficial finishes. Cork’s biodegradability also ensures that it can return safely to the earth at the end of its lifecycle, reinforcing the circular logic behind its use. In total, Nina’s House measures 112 m² (1,206 ft²), modest in scale but grand in vision. Its open-plan layout has been designed to support a young family’s dynamic lifestyle, with zones for connection and quiet being offered. Movement, gathering, and even dancing are encouraged, so that sustainability can be felt less like a constraint and more like freedom.The post This 1970s House Gets a Sustainable Cork Makeover And It’s Disrupting London Architecture first appeared on Yanko Design. Source: https://www.yankodesign.com/2025/05/13/this-1970s-house-gets-a-sustainable-cork-makeover-and-its-disrupting-london-architecture/?utm_source=rss&utm_medium=rss&utm_campaign=this-1970s-house-gets-a-sustainable-cork-makeover-and-its-disrupting-london-architecture #this #1970s #house #gets #sustainable #cork #makeover #and #its #disrupting #london #architecture

Why it matters: Efforts to reverse-engineer source code from retro console games to create native PC ports have intensified in recent years. Most completed projects have brought high resolutions, modding capabilities, and other enhancements to titles from the Nintendo 64 era and earlier. However, recent breakthroughs have begun to make more advanced platforms, like the GameCube and Xbox 360, increasingly accessible. Mario Party 4 is set to become the first GameCube title to receive a full decompilation, paving the way for unofficial PC ports. Unlike emulation, reverse engineering retro games offers modders significantly more flexibility when making modifications or adapting classics to new platforms. As MattKC Bytes explains in the video below, the decompilation of Mario Party 4 is over 99 percent complete, with only a few complex functions remaining. While decompilation alone doesn't produce a PC port, it is a critical first step – and a port of Mario Party 4 with online multiplayer is already in development. Mario Party 4 might seem like an obscure choice for the first fully decompiled GameCube game, but it was easier to reverse engineer than others because developer Hudson Soft originally compiled it without optimizations. Modders are also working on more popular GameCube titles, including Super Smash Bros. Melee, Metroid Prime, and The Legend of Zelda: Wind Waker. Meanwhile, an unofficial PC port of the Xbox 360 title Sonic Unleashed has recently been completed. // Related Stories Although emulators like Dolphin have long allowed fans to play these classics on modern platforms, porting decompiled code enables much more extensive enhancements – akin to mods for open-source PC games like Doom and Quake. In a recent video (below), YouTuber Nerrel explained how ongoing efforts allow games like the original Super Smash Bros. to run at 240 frames per second and open the door to major gameplay tweaks for The Legend of Zelda: Majora's Mask. Games such as Super Mario 64, Star Fox 64, and The Legend of Zelda: Ocarina of Time have already received PC ports that enable high resolutions and framerates. Additionally, a related project is attempting to revive Dinosaur Planet, which was never released. Modders are even experimenting with adding ray tracing to classic games, similar to NVIDIA's RTX Remix. A modding tool released about a year ago has helped streamline the recompilation process, though the community still faces numerous technical challenges. As a result, the timeline for future PC ports remains uncertain. Importantly, Nintendo hasn't struck down decompilation projects with lawsuits yet. Modders normally reverse engineer games using clean room tactics to avoid legal issues, and playing a PC port requires a copy of the original title. Masthead credit: VGR Remix Source: https://www.techspot.com/news/107905-modders-use-reverse-engineering-bring-mario-party-4.html #modders #use #reverse #engineering #bring #mario #party #more #gamecube #games #follow

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Productivity platform Notion is betting on large language models (LLMs) powering more of its new enterprise capabilities, including building OpenAI’s GPT-4.1 and Anthropic’s Claude 3.7 into their dashboard. Even as both OpenAI and Anthropic start building productivity features into their respective chat platforms, bringing these LLMs into a separate service shows how competitive the space is.  Notion announced its new all-in-one AI toolkit inside the Notion workspace today, including AI meeting notes, enterprise search, research mode and the ability to switch between GPT-4.1 and Anthropic’s Claude 3.7. One of the new features lets users chat with LLMs inside the Notion workspace and switch between models. Right now, Notion only supports GPT-4.1 and Claude 3.7. The idea is to reduce window and context switching. The company said early adopters of the new feature include OpenAI, Ramp, Vercel and Harvey.  Model mixing and fine-tuning Notion built the features with a mix of LLMs from OpenAI and Claude, and its own models.  The move away from pure reasoning models makes the model choice interesting for Notion. GPT-4.1 is technically not a reasoning model, while Claude 3.7 is a hybrid model— capable of acting as a regular LLM and a reasoning model.  Reasoning models are having a moment, though many warn that these models can still sometimes lie. However, while reasoning models like OpenAI’s o3 (and yes, Claude 3.7 Sonnet) take their time to answer, going through different scenarios, they are not the best for quick thinking and data-gathering tasks. Many productivity tasks, like meeting transcriptions or searches for task data, don’t need the power of a reasoning model behind them.  Sarah Sachs, Notion AI Engineering Lead, told VentureBeat in an email that the company aimed for features that didn’t sacrifice accuracy, safety and privacy, along with responding to queries in the speed enterprises need.  “In order to achieve a low-latency experience, we fine-tuned the models with internal usage and feedback from trusted testers, in order to make the AI specialized in Notion retrieval tasks,” Sachs said. “This setup helps Notion AI understand business needs, give relevant answers, serve customers with sub-second latency, and keep customer data safe and compliant.” Sachs said hosting and building with different models allows users to “pick the option that best fits their needs— whether that’s a more conversational tone, better coding capabilities, or faster response times.” AI meeting notes and more Notion AI for Work tracks and transcribes meetings for users, especially if they added Notion to their calendar, and it can listen in on calls.  Users can use Notion for enterprise search by connecting apps like Slack, Microsoft Teams, GitHub, Google Drive, Sharepoint and Gmail. Sachs said Notion AI will search an organization’s internal documents, databases and the connected apps.  The enterprise search results, along with other uploaded documents or a web search, allow Notion users to access the new Research Mode. It will draft documents directly from Notion while “analyzing all of your sources—plus the web—and think through the best response.” Notion also added both GPT-4.1 and Claude 3.7 as chat options. OpenAI noted that Notion users can chat with GPT-4.1 on the workspace and create a Notion template directly from the conversation. Sachs said the company is working on adding more models to its chat feature.  Subscribers to Notion’s Business and Enterprise plans with the Notion AI add-on get immediate access to the new features.   Compete with the model providers Even though Notion users can access both Anthropic and OpenAI on the platform, Notion still has to compete with model providers.  OpenAI’s Deep Research has been hailed as a game-changer for agentic retrieval augmented generation (RAG). Google also has its version of Deep Research. And Anthropic can search the internet for you.  Not to mention, Notion needs to compete with other platforms that already leverage AI. The meeting space is chock full of companies tracking, transcribing, summarizing and pulling insights from calls with AI. However, Notion’s big selling point is that it has all these capabilities on one single platform. Enterprises can use all those different services but live outside their chosen productivity platform. Notion said having all of these features in one place, with one all-in-one pricing, will save enterprises from subscribing to different platforms.  Daily insights on business use cases with VB Daily If you want to impress your boss, VB Daily has you covered. We give you the inside scoop on what companies are doing with generative AI, from regulatory shifts to practical deployments, so you can share insights for maximum ROI. Read our Privacy Policy Thanks for subscribing. Check out more VB newsletters here. An error occured. Source: https://venturebeat.com/ai/notion-bets-big-on-integrated-llms-adds-gpt-4-1-and-claude-3-7-to-platform/ #notion #bets #big #integrated #llms #adds #gpt41 #and #claude #platform

The Nintendo Switch 2 is set to launch with a couple of some pretty major heavy-hitting games in under a month’s time, but at the same time, its launch lineup is also going to boast several third party ports and re-releases. One of those is going to be Hogwarts Legacy, for which a new gameplay trailer has been uploaded to offer glimpses of how much better its visuals are on the Switch 2 compared to the Switch.  The trailer showcases a number of shots of the game, showing both indoor and outdoor environments, and the difference is, quite frankly, massive. From the lighting and the reflections to the density and detail in the environments and the draw distances, the open world action RPG is looking like a significant technical and graphical leap over the game’s Nintendo Switch version. Check out the comparison gameplay trailer below.  Hogwarts Legacy is available on PS5, Xbox Series X/S, PS4, Xbox One, Nintendo Switch, and PC, and launches for Nintendo Switch 2 on June 5.  Source: https://gamingbolt.com/hogwarts-legacy-trailer-shows-switch-vs-switch-2-graphics-comparison #hogwarts #legacy #trailer #shows #switch #graphics #comparison

Tennis players know the struggle of finding consistent practice partners who match their skill level and availability. Traditional ball machines offer repetitive feeds but lack the dynamic interaction that makes tennis truly engaging and effective for skill development. While these machines can help with basic shot repetition, they fail to simulate the unpredictability, movement, and strategic elements that define real match play, leaving serious players with a significant gap in their training options. Enter Acemate, the revolutionary tennis robot that transforms solo practice into an authentic rally experience. Unlike conventional ball machines that simply shoot balls in predetermined patterns, Acemate actively moves across the court, tracks your shots with 4K binocular vision, and returns balls with human-like groundstrokes that respond to your play style. This breakthrough technology creates a training experience that closely mimics actual match conditions, allowing players to develop not just technical skills but also the tactical awareness and decision-making abilities essential for competitive success. Designer: Acemate Click Here to Buy Now: $1499 $2329 ($830 off). Hurry, only 28/290 left! Raised over $1,001,000. The robot’s 4K binocular camera system represents a quantum leap beyond the monocular vision found in other training devices. By capturing depth and speed similar to human eyes, Acemate achieves centimeter-level accuracy in tracking both the ball and your movements across the court. This precision enables the robot to position itself optimally for returns while analyzing your technique in real time, providing an experience that feels remarkably natural and responsive compared to the static drills of traditional practice methods. Players of all levels will appreciate Acemate’s versatile training modes designed to target specific aspects of the game. The Baseline Rally mode delivers customizable groundstrokes that keep you moving laterally and testing your consistency from the back of the court. Mini-court Rally offers gentler exchanges perfect for beginners or warm-up sessions, while Net Rally mode focuses on volleys with controlled feeds that enhance your reactions and hand-eye coordination at the net, addressing the full spectrum of skills needed for complete player development. Advanced players seeking higher-intensity training will love the Pro-level Simulation mode, which recreates iconic rally patterns inspired by ATP and Grand Slam matches. This feature allows you to experience the rhythm and intensity of professional exchanges, pushing your footwork, shot selection, and stamina to new heights. For serve practice, Acemate functions as both ball boy and coach, delivering balls directly to your hand while measuring serve speed and tracking placement accuracy, eliminating the tedious ball collection that typically interrupts valuable practice time. This versatility is thanks to Acemate’s mobility system, featuring four Mecanum wheels that allow omnidirectional movement, enabling the robot to glide across the court with surprising agility. This movement capability means you’ll be challenged to hit shots to different court positions just as you would against a human opponent, improving your court coverage and shot placement under realistic conditions. The robot adjusts its positioning based on your shots, creating the dynamic back-and-forth that makes tennis so engaging and developing the spatial awareness that static training simply cannot provide. The companion app transforms your smartphone or Apple Watch into a powerful training command center, offering real-time performance metrics and customization options from ball velocity to landing spots to shot types. After each session, receive comprehensive reports detailing your performance, as well as actionable insights that guide your improvement with the precision of professional coaching. The robot’s compact form and retractable handle make it surprisingly portable, and compatibility with all court surfaces without special setups maximizes your valuable practice time. With Acemate, your tennis practices transform from monotonous ball feeding into dynamic, responsive training that mirrors match conditions. Players gain a consistent, tireless partner that adapts to their level while providing detailed performance analytics previously available only to professionals. This breakthrough technology finally solves tennis’s fundamental training challenge, allowing players to develop complete game awareness and tactical skills even when hitting solo on the court. Click Here to Buy Now: $1499 $2329 ($830 off). Hurry, only 28/290 left! Raised over $1,001,000.The post Ace Your Tennis Training with Acemate’s Revolutionary Rally Robot first appeared on Yanko Design. Source: https://www.yankodesign.com/2025/05/13/ace-your-tennis-training-with-acemates-revolutionary-rally-robot/?utm_source=rss&utm_medium=rss&utm_campaign=ace-your-tennis-training-with-acemates-revolutionary-rally-robot #ace #your #tennis #training #with #acemates #revolutionary #rally #robot

There's nothing more satisfying than hunting demons while performing brutal glory kills to style on them in Doom: The Dark Ages. Although the game runs buttery smooth on a range of hardware configurations, some technical issues may result in the game crashing or getting stuck on a black screen on launch. While these issues will be addressed and fixed with an upcoming patch sooner or later, here are some fixes you can try in the meantime to resolve them. Source: https://gamerant.com/fix-doom-the-dark-ages-crashing-black-screen-error-pc/ #how #fix #crashing #ampamp #black #screen #errors #doom #the #dark #ages