Tuesday, September 20th 2022
NVIDIA Project Beyond GTC Keynote Address: Expect the Expected (RTX 4090)
NVIDIA just kicked off the GTC Autumn 2022 Keynote address that culminates in Project Beyond, the company's launch vehicle for its next-generation GeForce RTX 40-series graphics cards based on the "Ada" architecture. These are expected to nearly double the performance over the present generation, ushering in a new era of photo-real graphics as we inch closer to the metaverse. NVIDIA CEO Jensen Huang is expected to take center-stage to launch these cards.
15:00 UTC: The show is on the road.
15:00 UTC: AI remains the center focus, including how it plays with gaming.
15:01 UTC: Racer X is a real-time interactive tech demo. Coming soon.
15:02 UTC: Future games will be simulations, not pre-baked- Jensen Huang
15:03 UTC: This is seriously good stuff (RacerX). It runs on a single GPU, in real-time, uses RTX Neural Rendering
15:05 UTC: Ada Lovelace is a huge GPU
15:06 UTC: 76 billion transistors, over 18,000 shaders. 76 billion transistors, Micron GDDR6X memory. Shader execution reordering is major innovation, as big as out-of-order execution for CPUs, gains up to 25% in-game performance. Ada built on TSMC 4 nm, using 4N, a custom process designed in together with NVIDIA.
There's a new streaming multiprocessor design, with a total of 90 TFLOPS. Power efficiency is doubled over Ampere.
Ray Tracing is on the third generation now, with 200 RT TFLOPS and twice the triangle intersection speed.
Deep Learning AI uses 4th gen Tensor Cores, 1400 TFLOPS, "Optical Flow Accelerator"
15:07 UTC: Shader Execution Reordering similar to the one we saw with Intel Xe-HPG
15:08 UTC: Several new hardware-accelerated ray tracing innovations with 3rd gen RTX.
15:09 UTC: DLSS 3 is announced. It brings with it several new innovations, including temporal components, and Reflex latency optimizations. Generates new frames without involving the graphics pipeline.
15:11 UTC: Cyberpunk 2077 to get DLSS 3 and SER. 16 times increase in effective performance using DLSS 3 vs. DLSS 1. MS Flight Simulator to get DLSS 3 support
15:13 UTC: Portal RTX, a remaster just like Quake II RTX, available from November, created with Omniverse RTX Remix.
15:14 UTC: Ada offers a giant leap in total performance. Everything has been increased 40 -> 90 TFLOPS shader, 78 -> 200 TFLOPS RTX, 126 -> 300 TFLOPS OFA, 320 -> 1400 TFLOPS Tensor.
15:17 UTC: Power efficiency is more than doubled, but power goes up to 450 W now.
15:18 UTC: GeForce RTX 4090 will be available on October 12, priced at $1600. It comes with 24 GB GDDR6X and is 2-4x faster than RTX 3090 Ti.
15:18 UTC: RTX 4080 is available in two versions, 16 GB and 12 GB. The 16 GB version starts at $1200, the 12 GB at $900. 2-4x faster than RTX 3080 Ti.
15:19 UTC: New pricing for RTX 30-series, "for mainstream gamers", RTX 40-series "for enthusiasts".
15:19 UTC: "Ada is a quantum leap for gamers"—improved ray tracing, shader execution reordering, DLSS 3.
15:20 UTC: Updates to Omniverse
15:26 UTC: Racer X demo was built by a few dozen artists in just 3 months.
15:31 UTC: Digital twins would play a vital sole in product development and lifecycle maintenence.
15:31 UTC: Over 150 connectors to Omniverse.
15:33 UTC: GDN (graphics delivery network) is the new CDN. Graphics rendering over the Internet will be as big in the future as streaming video is today.
15:37 UTC: Omniverse Cloud, a planetary-scale GDN
15:37 UTC: THOR SuperChip for automotive applications.
15:41 UTC: NVIDIA next-generation Drive
15:02 UTC: Future games will be simulations, not pre-baked- Jensen Huang
There's a new streaming multiprocessor design, with a total of 90 TFLOPS. Power efficiency is doubled over Ampere.
Ray Tracing is on the third generation now, with 200 RT TFLOPS and twice the triangle intersection speed.
Deep Learning AI uses 4th gen Tensor Cores, 1400 TFLOPS, "Optical Flow Accelerator"
15:26 UTC: Racer X demo was built by a few dozen artists in just 3 months.
333 Comments on NVIDIA Project Beyond GTC Keynote Address: Expect the Expected (RTX 4090)
Germany 4k EUR? Am not sure if that is true. My friend is a nurse and she gets measly 2K-2.4K EUR. Imagine there are a lot more less lucrative professions. Well in Poland that is considered a lot but yet in Germany you pay a lot for other things as well. Everything is way more expensive there.
The data from 2021 show that Polish salary on average is around 90720PLN while German people get 47.700 EUR average data from 2020. Dont get me wrong but that is all gross. Now if you consider the tax rates in both of those countries you will know your data is totally inaccurate. In Poland you have 12% and 34% if you exceed 120k by the salary number given earlier Polish person gets 80k PLN per year after tax income in Germany you've got 14% to 42% for 9.985 – 58.596 euros by the number given above you are left with 26kEUR. All of them are calculated as average but it also depends which tax category you will have in Germany.
Just by looking at the numbers you know Germans dont have it that sweet. Working in Germany and living in Poland totally different perspective. You have to consider that things in Germany are way more expensive than in Poland.
It shows a comparison between native and DLSS 3. Portal goes up to 6x (or almost that, I guess about 5.6x).
So if i interpret correctly the graphs, around 5.6X (3090Ti native 4K vs 4090 4K DLSS performance (1080p) + frame interpolation= DLSS 3.0)
or around 2.8X (3090Ti 4K DLSS performance (1080p) vs 4090 4K DLSS performance (1080p) + frame interpolation= DLSS 3.0))
117/5.6=20-21fps but for 3090Ti 4K native not for 4090!
I expect the games on the left side of the slide are CPU-limited, that is why only the interpolation is causing a 2x increase in framerate. Anything above 2x means that lowering the rendering resolution gives a boost, after which the framerate is doubled by DLSS 3.
The bottom of the slide only mentions that the resolution is 4K with DLSS Performance and the interpolation enabled.
cdn.wccftech.com/wp-content/uploads/2022/09/NVIDIA-Ada-Lovelace-GPU-GeForce-RTX-4090-RTX-4080-Series-Graphics-Cards-_7-1480x833.jpg
I can't wait for independent testing because if indeed is 4090 only, then Nvidia claims for example that in Cyberpunk max RT when you enable DLSS 3.0 your frame rate quadruples!
I wonder, can it be that good in games that push raytracing hard?
Anyway in classic raster, since jensen claimed 4090 will be 2X vs 3090Ti, i expect it to be less (-10% -15% less, so 1.8X 1.7X).
I think jensen in 3080 launch had said also 2X vs 2080 but the actual difference at launch was around 1.65X not 2X, so the logic assumption is around 1.65X this time also!
Original for comparison.
About price, there have been significant price cuts in all 30x0 products and AMD GPUs. Who wants the newest high-end card, barely on the market, has to pay early adopter price and could run into early adopter problems. Enough said. I will wait six months, to see real performance, impact of new AMD series, likely second revision of AIBs, consolidated market price. Until then, a nice time for all, don't forget to enjoy, what you already have.
Maybe I'm getting old or have played so many games but I can no longer play games with no RT or simple baked lighting, flying objects and flat image.
To be fair there are games with no RT that managed to capture the environment, not correctly but in a way that it doesn't trigger the player's brain.Only those who play competitive games and only, do not care about RT.
It's the holy grail in the single player games.
The performance will always be something that we have to work around when new techs are added. When GeForce 3 Ti introduced pixel shader 1, the performance was bad. Should we have stayed with the MXs because they were faster?(showing flat shxxxxxxty surfaces everywhere...)
Only games with very static environments like the Last of us 2 pull off baked lighting pretty decently but I'm personally not a huge fan of games that linear and even then it would benefit from RT.
Flagships/high end gpu should not only be able to do RT but do it well. Midrange I can still give a pass for now.
In the GTX era, 104/204 dies were always under 200 W on the initial launch. Even the 3070 was only 220 W, and that was 393 mm2.
These cards will undervolt like a dream. You can probably cut the power consumption by 50% and only lose 20% performance or less. If the prices ever get changed to normal, these cards will be very attractive.
On top of this, AD104 has 38 billion transistors in that area, vs. 17 billion for GA104 - so not only are the transistors running much faster, but there are twice as many of them, in a smaller overall area.It's always fun when people have nothing to back up their "arguments" and instead resort to name-calling and personal attacks. It really, really isn't my problem that you are clearly incapable of seeing the ideological undertones of your own statements. Because they are there, whether you like it or not. Clear as day to anyone with even the most minute ability to interpret writing. This isn't me being offended, it's me telling you that you're blind to the implications of your own beliefs and words, and attempting to inform you that there might be things that you're missing in what you're saying. Your initial responses made it seem like you didn't really want to be expressing these kinds of things, so maybe you should try and reconsider those beliefs, and fix that instead?
Coolers will have to have good contact to avoid hotspots, that's for sure. Other than that, my only worry is the heat passed on to the rest of the case. GPU temps should be fine, I think.