NVIDIA announced the “Ultimate Countdown“towards their next gen Ampere GeForce graphics cards, but then there were the hints with numbers and more specifically: 21.
FREE GALLERY – 14 PHOTOS
After posting the tweet #UltimateCountdown, NVIDIA added: “21 Days. 21 Years” which means August 31st (21 days after the countdown was announced on August 10th) and it’s been 21 years since the first GeForce, the first GPU , the GeForce 256.
So much so that the number 21 runs that if you take ‘256’ and turn it through ’21’ you get 5376, which is the estimated number of CUDA cores on NVIDIA’s upcoming GA102 GPU.
My friend Usman Pirzada tweeted over at Wccftech “256 * 21 = 5376“but then the second I saw it, it clicked – so I tweeted:”5376 = 5 + 3 + 7 + 6 = 21“. NVIDIA might agree with us, but it seems that for some reason they are moving towards 21.
Let us remember that – 21 is important in many ways.
- 21 days until next-gen GeForce unveils.
- 21 years since the first GPU ever.
- 21 years since the first GeForce graphics card.
- The first GeForce graphics card was the GeForce 256.
- GeForce 256 was offered in 2 different VRAM options: SDR and DDR.
- The ‘256’ times ’21’ = 5376 and then 5 + 3 + 7 + 6 = 21.
- AMD has Navi 21 coming.
- Navi 21 will compete with Ampere.
21 seems NVIDIA very important.
That all those connections in ’21’ above, something about the most important of all: the nomenclature of the next gen-GeForce graphics card.
What if NVIDIA trolls us all with ’21’ and throws us off track, given that 21 seems to work for so many layers of this – and the actual reveal is GeForce RTX 21 series. I mean, it’s logical to see the jump from GeForce GTX 10 series to GeForce RTX 20 series.
It is a similar movement Samsung moved the Galaxy S10 to the Galaxy S20 – it was not the Galaxy S10 to Galaxy S11, but rather the Galaxy S20 that stepped out in the year 2020, and then the rumors Galaxy S21 is intended for next year. What if NVIDIA moved into the GeForce RTX 21 series for 2021 and beyond, as well as the next issue to GeForce RTX 20 series.
Another connection is that Samsung and NVIDIA are collaborating on 8nm GPUs for the first pair of Ampere GPUs over the next-gen GeForce RTX series (regardless of what nomenclature it launches with). Maybe NVIDIA took some inspiration from Samsung with their name scheme going forward.
Which do you prefer? Now these models are not confirmed, but I think we could see an “ULTIMATE” brand card, maybe a new “TITAN RTX ULTIMATE” or something along those lines. I personally would love to see the return of the ULTRA brand, but that’s just that I’m a nerd.
- GeForce RTX 2190 ULTIMATE
- GeForce RTX 2190 ULTRA
- GeForce RTX 2190
- GeForce RTX 2180 Ti
- GeForce RTX 2180
- GeForce RTX 2170
- (and so on)
Of …
- GeForce RTX 3090 ULTIMATE
- GeForce RTX 3090 ULTRA
- GeForce RTX 3090
- GeForce RTX 3080 Ti
- GeForce RTX 3080
- GeForce RTX 3070
- (and so on)
It seems that NVIDIA has almost only shuffled with the number 21, so maybe I’m wrong – but I’ll put this out there for the world to take. Fire away in the comments on social media, you can hit me personally on Twitter if you want to discuss the GeForce RTX 21 series.
- NVIDIA and TSMC issues over 7nm: NVIDIA shifted some of its weight to Samsung in 2019, but AMD soon began to see gains with its effencicy on 7nm TSMC and went back to Ampere. But AMD had secured the gigantic chunk of 7nm TSMC fab time with its various chips. This forced NVIDIA to split between Samsung (RTX 3080 and others) and then TSMC (higher-end RTX 3090, etc.).
- AMD beats all that 7nm up: AMD is gobbling up all that 7nm production at TSMC, between their Ryzen, Threadripper and EPYC Processors – then their Radeon RX series cards (both current and RDNA 2-based next gen cards) like the PlayStation 5 and Xbox Series X chips are all on display at 7nm at TSMC.
- GeForce RTX 3080 = 8nm with Samsung: This was not intended to happen, but it is – and NVIDIA has no choice but to take action. NVIDIA will use Samsung’s new 8nm node for at least 3 of its Ampere GPUs (you can do that read more about it here), but they will not have the power to completely reverse Big Navi.
- GeForce RTX 3090 = 7nm with TSMC: The GeForce RTX 3090 higher end will, however – but it will launch in Q1 2021. This card is what is referred to in NVIDIA as the Trump card.
- The Trump card from NVIDIA: GeForce RTX 3090 will be made at 7nm with TSMC, and this should be pretty much the monster indeed with the rumor being 45-60% faster than the RTX 2080 Ti. This will be the card that NVIDIA will hit very, very early 2021.
Image source.
Read more:
- Traversal coprocessor: We’ve had more leaks on NVIDIA’s next-gen GeForce RTX 3000 series than any other graphics card before, with an interesting “traversal coprocessor” on the new GeForce RTX 3080 and GeForce RTX 3090 graphics cards. you can read more about it here.
- NVCache: Ampere is meant to have what’s called NVCache, which would be NVIDIA’s own form of AMD’s HBCC (High Bandwidth Cache Controller, more about that here). NVCache would use your system RAM and SSD to optimize super-speed loading of gaming consoles as well as VRAM usage. you can read more about NVCache here.
- Tensor memory compression: NVCache is interesting, but Tensor Memory Compression will be on Ampere, and will most likely use Tensor Cores to compress and decompress items stored in VRAM. This could see a 20-40% reduction in VRAM usage, as more VRAM usage with higher textures in next-gen games and Tensor Memory Compression reduced that VRAM footprint by 20-40%.
- How fast is the GeForce RTX 3090?? Freaks fast according to rumors, with 60-90% more performance than the current Turing-based flagship GeForce RTX 2080 Ti. We could jump this enormous performance into titles of raytracing, but we’ll have to wait a little longer to see how much graphics power NVIDIA craps in these new cards. you can read more about those rumors here.
- Power hungry: In terms of power consumption, GA102 reports 230W – while 24GB GDDR6X (what we should see on the new Ampere-based TITAN RTX) consumes 60W of power. you can read more about it here.
- Production will begin shortly: NVIDIA is reported in the DVT (as Design Validation Test) range of its new GeForce RTX 3000 series graphics cards. Mass production kicks off report in August 2020, with a media event, benchmarks, and more in September 2020 as I predicted many months ago. More about this here.
I’ve already written about rumors of that NVIDIA’s next gen Ampere GPU architecture would be up to 75% faster than GPUs with current gen such as the Turing architecture, right after rumored that Ampere would offer 50% more performance at half the power of Turing. This is very crazy thing right now.
Not only that, but we have something called specifications on the designated GeForce RTX 3080 and GeForce RTX 3070 graphics cards, both of which will be powered by NVIDIA’s new Ampere GPU architecture.
We’ve heard that before Ampere would offer 50% more performance at half the power of Turing, who sent the hairs on my neck. Better yet, you can read about the leaky specs on the proposed Ampere-based GeForce RTX 3080 and GeForce RTX 3070 here.
Read more:
- NVIDIA GeForce RTX 3000 series announcement announced for September 9th
- NVIDIA stops reporting GeForce RTX 2070 SUPER graphics card
- NVIDIA GeForce RTX 3080 Ti: rumored launch September 17th
- AMD Big Navi: 5120 streaming processors plagued RDNA 2’s flagship
- AMD Radeon Instinct MI100 Acturus placed, NVIDIA Ampere Destroyer ?!
- GeForce RTX 3090: 50% faster than RTX 2080 Ti in early benchmarks
- NVIDIA A100 Almost benchmarked, is now the ‘fastest GPU ever recorded’
- Yeah, NVIDIA should release GeForce RTX 3000 series in September
- AMD Big Navi ‘NVIDIA Killer’ flagship card has 16 GB of memory
- NVIDIA GeForce RTX 3080 rumored: 20% faster than GeForce RTX 2080 Ti
- The next-gen Ampere DLSS 3.0 from NVIDIA could work on ANY game that uses TAA
- AMD is aiming for Big Navi launch for November as ‘show of strength’ for RDNA 2
- NVIDIA’s next-gen Ampere GeForce RTX 30 series: new 12-pin PCIe power
- AMD’s next-gen RDNA 2 rumored: 40-50% faster than GeForce RTX 2080 Ti
- NVIDIA stops making RTX 20 series, disasters after GeForce RTX 30 series
- GeForce RTX 3060 should cost $ 300- $ 400, here are some cool specs
- NVIDIA bundles Death Stranding with GeForce RTX graphics cards
- NVIDIA GeForce RTX 3070 Ti, RTX 3070 rumored specs: two sub $ 500 cards
- Do you want the next-gen Ampere GPU from NVIDIA now? You can have it … for $ 12,500
- NVIDIA’s Ampere GeForce RTX rumored: built on Samsung 8nm, not TSMC 7nm
- Say hello to the ASUS GeForce RTX 3080 Ti ROG STRIX, maybe
- NVIDIA’s new A100 PCIe accelerator: 40GB HBM2e memory, PCIe 4.0 tech
- This new leak of GeForce RTX 3090 has it at 26% faster than RTX 2080 Ti
- New GeForce RTX 3090 Leaks: 12GB GDDR6X at Crazy 21Gbps
- GeForce RTX 3080 and RTX 3090 rumored that ‘traversal coprocessor’ would pack
- The next-gen GeForce RTX 3080, RTX 3090 from NVIDIA to enter production shortly
- GeForce RTX 3090: GA102 consumes 230W, 24GB GDDR6 consumes 60W power
- NVIDIA rumored to use HUGE cooling pad on GeForce RTX 3080
- NVIDIA GeForce RTX 3080 cooler: rumored to cost $ 150 on its own
- GeForce RTX 3090 rumored: 24GB GDDR6X, would destroy RTX 2080 Ti
- NVIDIA GeForce RTX 3090 rumored: up to 60-90% faster than RTX 2080 Ti
- Check out these amazing renderers from NVIDIA’s next-gen GeForce RTX 3080
- This may be our first image of the GeForce RTX 3080 graphics card
- AMD and NVIDIA to launch both GPUs in September 2020
- NVIDIA is likely to drop the Tesla brand, too close to Elon Musk’s 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 specs: 54 billion transistors, 40GB HBM2, 7nm TSMC
.
