The Hot Take: Interesting.
Nvidia has just demoed its Neural Texture Compression technique again at a GTC talk, where it showed VRAM usage dropping from 6.5 GB to just 970 MB in a scene. NTC uses a neural network to decompress textures instead of standard block-based compression, reducing texture size and VRAM usage while also improving final image quality.
Read the full article
The Hot Take: Following intels steps on the Arc? Also, how much space are the pre-compiled shaders going to consume of diskspace?
NVIDIA has introduced a new beta feature called Auto Shader Compilation, or ASC, through the latest NVIDIA App update, and it targets a familiar pain point in modern PC gaming: long initial loading phases and shader compilation stutter in DirectX 12 titles.
Read the full article
The Hot Take: Let's milk the architecture untit the pleebs scream, beg and plead for a new architecture... All while ringing out as much cash from the Ai market......
There’s an easier way: A manufacturer could simply release the most expensive gaming graphics card in the series, and the market would eventually settle down. For NVIDIA, however, that moment seems to be a long time coming. Since early February, reports have been circulating that an even more powerful Blackwell model—positioned above the GeForce RTX […]
Source
Read the full article
The Hot Take: You will get mid-range only and love it.... Probably all while renting it I assume.
Nvidia wants a slice of every laptop sold, not just the ones with a chunky discrete GPU.
The firm is lining up “exclusive” laptop system-on-chips for consumers this year, barging into a market long owned by Intel and AMD while trying to cash in on the AI PC hype.
The pitch is that Nvidia has ignored the huge integrated CPU-and-GPU segment, even though it ships bucketloads of graphics chips for gaming and workstations.
Nvidia chief executive Jensen Huang said: “There’s 150 million laptops sold per year, and Nvidia’s market largely targets gaming and workstation markets where discrete GPUs are used. And we’re very successful there. There’s an entire segment of the market where the CPU and the GPU are integrated. And that segment has been largely unaddressed by Nvidia today.”
He said that entire segment of the market is quite rich, large, and underserved today, with state-of-the-art, world-class GPUs like Nvidia’s.
The big idea leans hard on on-device AI, with CPU vendors repackaging product lines around NPUs such as Intel’s NPU and AMD XDNA, and Nvidia fancies itself as the obvious third wheel.
It is pushing the envelope by pairing silicon with software, dropping its open-source model stack, Nemotron, alongside laptop SoCs to ride the edge AI frenzy.
If Nvidia stuffs enough consumer machines with its own silicon, it can bake “on-device AI” features in as defaults and grab a bigger cut of whatever edge AI turns into.
This would give Nvidia an edge that Intel and AMD “cannot achieve”, because they are not building foundation models, they are just selling the compute.
If edge AI really does hit the predicted $160 billion valuation by 2030, then Nvidia could be on to something.
On the silicon side, the rumour mill says Nvidia is building ARM-based laptop chips with MediaTek, following the shape of its GB10 SuperChip used in the DGX Spark mini-AI supercomputer.
The Nvidia and MediaTek pairing is not new, since they have already collaborated in automotive via the “Dimensity Auto” line with RTX GPU IP bolted in.
Two consumer SKUs are expected, codenamed “N1X” and “N1”, with the latter pitched as the weaker of the two, and both have appeared on public benchmarks.
The architecture is tipped to use “ARM foundations” because power efficiency matters in laptops and MediaTek lives on ARM anyway.
There is speculation that Nvidia could co-design ARM IP to stand out from other ARM laptop plays, such as the Fruity Cargo Cult Apple and Qualcomm.
If Nvidia follows the GB10 pattern, it could use ARM v9.2, but that is still guesswork.
Process rumours point to TSMC 3nm, and the leaked CPU numbers for the bigger N1X suggest a 20-core cluster at 2.81GHz base with a 4GHz boost.
The weaker N1 is expected to land in eight or 12-core setups.
In graphics, the integrated RTX chunk is expected to be Blackwell-based, and early chatter claims a 6,144-CUDA-core layout with 48 SMs.
Despite that headline figure, it is still a mobile part, with leaks suggesting up to 120W TDP, putting it in the same power bracket as AMD Strix Halo and Intel Lunar Lake.
The Geekbench OpenCL numbers being waved around put “Nvidia N1X (6144 Cores)” at 46,361, miles behind “RTX 5070 Desktop (6144 Cores)” at 185,269.
Memory support is expected to include LPDDR5X, with up to one petaflop of FP4 AI compute.
Nvidia is even rumoured to be eyeing handhelds later, since it cannot resist chasing the whole gaming market once it smells blood.
It is not stopping at ARM, either, since it is said to be working on an x86 laptop chip through its partnership with Intel, which would give it a foot in both camps.
That ambition runs straight into supply reality, with DRAM tight and TSMC capacity reportedly fully booked, so consumer dreams may lose to data centre margins.
The expectation is that if the N1X and N1 show up at Computex in early June 2026, early availability may be limited due to a stretched supply chain.
Dell and Lenovo are said to be gearing up for designs, hinting that OEMs are curious, even as they brace for pricing and volume drama.
Pricing is still foggy, but the piece puts the N1X laptops in a rough $1,500 to $2,000 range, depending on configuration.
Read the full article
The Hot Take: I'm glad, but will be actually be able to afford or get any in our hands? Also what games are we going to need this for, as game releases have definitely stagnated along with the market.
NVIDIA's GeForce RTX 60 Series GPUs will be powered by the Rubin architecture, which exists only for AI and data center use thus far. The Rubin CPX, for example, is built around NVIDIA's GR212 chips, but new information shared with YouTuber RedGamingTech claims the RTX 60 Series chips will be the GR202 (RTX 6090), GR203 (RTX 6080), and GR205
Read the full article
The Hot Take: Uh oh, Ai king looks to be in trouble.
U.S. senators Elizabeth Warren (D-Mass.) and Jim Banks (R-Ind.) told Commerce Secretary Howard Lutnick that he should suspend all active export licenses to China for Nvidia AI chips, saying that Nvidia's most advanced AI GPUs are being diverted into the country despite Jensen Huang's assurances.
Read the full article
The Hot Take: Nvidia starting to feel the heat of competition and see those $ evaporate as they try other vendors.
Nvidia is preparing to launch a new chip designed to speed up AI responses, breaking with its long-running habit of flogging the same processor for every job.
Nvidia chief executive Jensen Huang is expected to unveil a chip focused on “inference”, meaning running models rather than training them.
According to people familiar with the plans for GTC next week, the chip is the first new product to emerge from December’s $20bn deal to hire the founders of Groq, a start-up building “language processing units” tuned for high-speed answers to complex AI queries.
Three months after that deal, Nvidia is expected to debut a Groq-based LPU to sit alongside its forthcoming flagship Vera Rubin graphics processing unit. It is part of a product family meant to head off challengers and meet new kinds of AI applications.
The move lands as the world’s most valuable company gets grief from start-ups and customers, such as Google, all busy cooking up their own AI chips. This week, Meta announced a new family of four inference-focused processors.
One Silicon Valley venture investor said: “We are entering an interesting phase that is not ‘Nvidia dominant’,”
For the past three years, Nvidia’s $4.5tn market capitalisation has been built on its GPUs, which have become the backbone of generative AI. They train models such as the ones behind OpenAI’s ChatGPT.
Huang has insisted that a single system can handle training and then run the chatbots and coding tools built on top. Big Tech has spent hundreds of billions deploying these boxes while funding their own specialised silicon.
But the growing sophistication of AI tools, including “agentic” coding systems, is pushing Huang to ditch the mantra that one GPU fits every workload.
The Groq deal was worth about $20bn, according to people familiar with the transaction, making it one of the biggest deals in Nvidia’s 33-year history. It includes licensing and the hiring of key talent, including Groq founder and former Google chip executive Jonathan Ross.
Groq, which had been working with Samsung to manufacture its products, previously bragged that its LPUs were faster and more efficient than Nvidia’s GPUs for inference. Nvidia clearly listened.
Nvidia’s flagship Blackwell and Rubin systems lean on high-bandwidth memory to cope with the massive data loads that AI models fling around. But HBM is expensive and in increasingly short supply as SK Hynix and Micron struggle to keep up with demand.
The Groq-style chip will use SRam rather than the dynamic Ram used for HBM, according to people familiar with Nvidia’s plans, because SRam is more available and better suited to speeding up AI “reasoning” tasks.
Bank of America reckons that by 2030, inference will account for 75 per cent of AI data centre spending, up from about 50 per cent last year, and it expects a “broadened AI portfolio” at GTC.
Read the full article
The Hot Take: Good choice getting close to the Ai market king, but that's a double edge sword. I have a feeling this will just hasten Intel's acquisition by Nvidia. Nvidia just dropping a large chuck of ARM which doesn't make sense seeing they're Grace CPU is an ARM SoC and their NX1 is supposed to be another ARM SoC too. Time will tell.
<p><a href="https://wccftech.com/intel-to-show-up-at-nvidia-gtc-at-the-perfect-time/"><img width="728" height="546" src="https://cdn.wccftech.com/wp-content/uploads/2025/10/G1H6wsLaQAER8gz-2-728x546.jpeg" alt="Man wearing a grey sweater with intel. logo in an indoor setting."></a></p><p>Intel is now coming to NVIDIA's GTC mega-event, not just a guest this time, but rather the company will play an important role in dictating the future of NVIDIA's compute capabilities. Intel's Server CPU Constraints Are Going to Get a Lot More Aggressive, Following Their Collaboration With NVIDIA For those unaware, this year's GTC is expected to feature several major announcements that will influence NVIDIA and its supply chain partners, particularly Intel, which will also get the spotlight. NVIDIA and Intel entered into a $5 billion agreement a few months ago, in which both companies agreed to work together in […]</p><p>Read full article at <a href="https://wccftech.com/intel-to-show-up-at-nvidia-gtc-at-the-perfect-time/">https://wccftech.com/intel-to-show-up-at-nvidia-gtc-at-the-perfect-time/</a></p>
Read the full article