We found out about NVCache back in May 2020 with some Ampere leaks, with NVIDIAs supposed NVCache acting like HBCC– HBCC being AMDs own High Bandwidth Cache Controller that debuted with Vega.
VIEW GALLERY – 3 IMAGES
Sony is using some genuinely next-gen storage technology inside of its PlayStation 5 console, with a deep dive on that here. Microsoft will be doing the same with their Xbox Series X console, however what about the PC? AMD will surely have something awesome up their sleeve with RDNA 2– while NVIDIA has NVCache.
Well, back in the May leaks about Ampere and NVCache we heard that it “leverages both your DDR & & SSD for improved load times & & VRAM”. The brand-new leaks from Moores Law is Dead tease that NVIDIAs upcoming NVCache on Ampere cards in the GeForce RTX 3000 series family will “dynamically use bandwidth from SSDs, VRAM, and DDR for multiple tasks at the very same time”.
Theres also Tensor Memory Compression that on Ampere, and will supposedly utilize Tensor Cores to both compress and decompress items that are kept in VRAM. This could see a 20-40% decrease in VRAM use, or more VRAM use with higher textures in next-gen games and Tensor Memory Compression decreasing that VRAM footprint by 20-40%.
NVIDIA could have some Holy Trinity level mixes of functions and innovations between DLSS 3.0, NVCache, and Tensor Memory Compression. All of these when signed up with forces with a more powerful, more unhinged Ampere GPU, could result in among the most interesting GeForce releases ever.
Also, right when AMD is about to strike when the iron is hot with RDNA 2 and its “NVIDIA Killer”, more on that here.
More reading:
Ive already discussed reports that NVIDIAs next-gen Ampere GPU architecture would depend on 75% faster than current-gen GPUs such as the Turing architecture, right after reports that Ampere would use 50% more efficiency at half the power of Turing. This is pretty crazy stuff right there.
Not only that, but weve got some reported specifications on the purported GeForce RTX 3080 and GeForce RTX 3070 graphics cards, which will both be powered by NVIDIAs new Ampere GPU architecture.
Weve currently heard that Ampere would provide 50% more performance at half the power of Turing, which sent the hairs on my neck standing up. Much better yet, you can check out the dripped specifications on the purported Ampere-based GeForce RTX 3080 and GeForce RTX 3070 right here.
Traversal coprocessor: We have actually had more leaks on NVIDIAs next-gen GeForce RTX 3000 series than any family of graphics cards prior to it, with a fascinating “traversal coprocessor” on the brand-new GeForce RTX 3080 and GeForce RTX 3090 graphics cards. Freaking quick according to rumors, with 60-90% more performance than the existing Turing-based flagship GeForce RTX 2080 Ti. Production starts soon: NVIDIA is reportedly in the DVT (or Design Validation Test) range of its new GeForce RTX 3000 series graphics cards.
Much more reading:
Traversal coprocessor: We have had more leakages on NVIDIAs next-gen GeForce RTX 3000 series than any household of graphics cards prior to it, with an intriguing “traversal coprocessor” on the brand-new GeForce RTX 3080 and GeForce RTX 3090 graphics cards. How quick is the GeForce RTX 3090? Freaking quick according to reports, with 60-90% more efficiency than the present Turing-based flagship GeForce RTX 2080 Ti. Power hungry: As for power usage, GA102 supposedly utilizes 230W– while 24GB of GDDR6X (which we ought to see on the brand-new Ampere-based TITAN RTX) takes in 60W of power. Production begins quickly: NVIDIA is supposedly in the DVT (or Design Validation Test) range of its brand-new GeForce RTX 3000 series graphics cards.
NVIDIAs next-gen Ampere GeForce RTX 30 series: new 12-pin PCIe power AMDs next-gen RDNA 2 report: 40-50% faster than GeForce RTX 2080 Ti NVIDIA stops making RTX 20 series, ramps to GeForce RTX 30 series GeForce RTX 3060 should cost $300-$ 400, heres some leaked specifications NVIDIA bundles Death Stranding with GeForce RTX graphics cards NVIDIA GeForce RTX 3070 Ti, RTX 3070 rumored specifications: two sub $500 cards Want NVIDIAs next-gen Ampere GPU now? You can have it … for $12,500 NVIDIAs Ampere GeForce RTX report: constructed on Samsung 8nm, not TSMC 7nm Say hey there to the ASUS GeForce RTX 3080 Ti ROG STRIX, maybe NVIDIAs new A100 PCIe accelerator: 40GB HBM2e memory, PCIe 4.0 tech This new GeForce RTX 3090 leakage has it at 26% faster than RTX 2080 Ti New GeForce RTX 3090 leakages: 12GB GDDR6X at ridiculous 21Gbps GeForce RTX 3080 and RTX 3090 rumored to load traversal coprocessor NVIDIAs next-gen GeForce RTX 3080, RTX 3090 to enter production soon GeForce RTX 3090: GA102 consumes 230W, 24GB GDDR6 takes in 60W power NVIDIA reported to use HUGE cooling block on GeForce RTX 3080 NVIDIA GeForce RTX 3080 cooler: rumored to cost $150 on its own GeForce RTX 3090 report: 24GB GDDR6X, would wipe out RTX 2080 Ti NVIDIA GeForce RTX 3090 reports: as much as 60-90% faster than RTX 2080 Ti Check out these amazing renders of NVIDIAs next-gen GeForce RTX 3080 This might be our first image of the GeForce RTX 3080 graphics card AMD and NVIDIA to both launch next-gen GPUs in September 2020 NVIDIA apparently drops Tesla brand, too near Elon Musks Tesla NVIDIA amps up battle against COVID-19 with Ampere-based supercomputer NVIDIA DGX A100: 8 x A100 Ampere GPUs, AMD CPU, 15TB NVMe SSD NVIDIA Ampere A100 specifications: 54 billion transistors, 40GB HBM2, 7nm TSMC
