NVIDIA was all systems go for its Ampere GPU architecture (which I exclusively unveiled all the way back in April 2018) unveiling at its own GPU Technology Conference (GTC) last week, but it was canceled over the coronavirus COVID-19 outbreak. This meant the Ampere announcement was also delayed.
But now Bloomberg's Tai Kim has written a new article titled 'NVIDIA's Biggest Pleasant Surprise May Be Yet to Come', where he goes into the fact that NVIDIA is kinda flourishing over the coronavirus outbreak. In positive news about coronavirus, more gamers are at home, and more games are being played right now than pretty much ever before in the history of gaming.
NVIDIA's gaming segment (GeForce graphics cards) represents close to half of its revenue, so with more gamers playing games at home that is only a good thing for NVIDIA. Kim writes that NVIDIA management said they've seen a 50% surge in "total gaming hours from its installed base as many students and workers were staying-at-home".
According to my sources and own analysis of the market, NVIDIA was going to launch its next-gen Ampere GPU architecture at GTC 2020. It most likely would've had some higher-end Quadro RTX offerings, and then the GeForce RTX 3000 series later this year -- the same game plan for its Turing launch in 2018, where the Turing-based Quadro RTX graphics cards launched before the GeForce RTX graphics cards.
NVIDIA Ampere GPUs
· GA102 - 84 SMs / 5376 CUDA cores / 12GB GDDR6 / 384-bit bus - 40% faster than RTX 2080 Ti
· GA103 - 60 SMs / 3840 CUDA cores / 10GB GDDR6 / 320-bit bus - 10% faster than RTX 2080 Ti
· GA104 - 48 SMs / 3072 CUDA cores / 8GB GDDR6 / 256-bit bus - 5% slower than RTX 2080 Ti
NVIDIA Ampere GA100 Specs
· 8192 CUDA cores @ 2GHz (2.2GHz boost)
· 1024 Tensor Cores
· 130 RT Cores
· 48GB of HBM2e memory @ 1.2GHz
· 300W TDP
· TSMC 7nm+
· 36 TFLOPs peak output
I've already written about rumors that NVIDIA's next-gen Ampere GPU architecture would be up to 75% faster than current-gen GPUs such as the Turing architecture, right after rumors that Ampere would offer 50% more performance at half the power of Turing. This is pretty crazy stuff right there.
Not only that, but we've got some rumored specs on the purported GeForce RTX 3080 and GeForce RTX 3070 graphics cards, which will both be powered by NVIDIA's new Ampere GPU architecture.
We've already heard that Ampere would offer 50% more performance at half the power of Turing, which sent the hairs on my neck standing up. Better yet, you can read about the leaked specs on the purported Ampere-based GeForce RTX 3080 and GeForce RTX 3070 right here.
GA103 (GeForce RTX 3080)
· 10/20GB GDDR6
· 320-bit memory interface
· 60 SMs
· 3480 CUDA cores
GA104 (GeForce RTX 3070)
· 8/16GB GDDR6
· 256-bit memory interface
· 48 SMs
· 3072 CUDA cores
AMD also just revealed its next-gen RDNA 2 / Navi 2X plans, something we referred to as Big Navi until just recently. Those specs are monstrous, but they pale in comparison to what NVIDIA has planned for its flagship HPC destined Ampere GA100 beast.
AMD RDNA 2 / Navi 2X / Big Navi Specs
· 7nm node (confirmed)
· RDNA 2 architecture (confirmed)
· 80 compute units
· 5120 stream processors
· 17.5 TFLOPs compute performance
· Hardware ray tracing support (confirmed)