Nvidia's 3000 Series RTX GPU [3090s with different memory capacity]

Another thing to consider with regards to RAM size is that at the start of the current generation, the fastest GPU's available were sporting only 2 and 3GB vs the consoles 8. 10GB vs the new consoles 16GB is far better.

Shared pool for the consoles though, about 10 for next gen, about 2/3 for current base consoles.

aida64-gpgpu_geforce-cmk6q.png
aida64-gpgpu_geforce-f7j6i.png


https://www.pcgameshardware.de/Gefo...Tests/Raytracing-Vergleich-DLSS-Test-1357418/

Over 30TF for the 3080 it seems.
 
Notebook variants of the 20xx GPUs are heavily throttled to make them fit for a laptop. And even at the 80W limit of the MaxQ cards, the notebook cooling solutions required already start to be heavy and noisy. The most sensible solution would be a card capable of playing games maxed out at 720p, RT included, coupled with a high performance, high quality dedicated upscaling unit. But if the upscaling process has to run on the same shader units, as it happens now with Ampere, then the upscaling performance is tied to that of the whole GPU.

I know. But GA104 is more efficient than GA102 with GDDR6X. So notebooks will get 40%+ uplift with Ampere over Turing.
 
are they going to use eye tracking ? otherwise how do they know what part of the screen to render in high res

That's how VRS is used in VR.

Outside of VR, it doesn't use eye-tracking - there's a DF video on the topic:
 
It's almost funny how slow brute compute progress is. I have 3.5year old 1080ti and 3080 is not even double the perf. I suppose this is why we need creative solutions to create better graphics. DLSS, ray trace, mesh shaders, sampler feedback etc. Just adding more of the same doesn't work anymore, or at least it works very slowly.
 
It's almost funny how slow brute compute progress is. I have 3.5year old 1080ti and 3080 is not even double the perf. I suppose this is why we need creative solutions to create better graphics. DLSS, ray trace, mesh shaders, sampler feedback etc. Just adding more of the same doesn't work anymore, or at least it works very slowly.

I think with most games at 4k the 3080 is more than 2x faster than the 1080ti. Doom Eternal is close to 2x as fast at 1080p. I have a feeling that some games at 1080p may actually be cpu limited, even with a 10900K. 10900K seems to outperform the Ryzen 3900X by a significant margin with a 3080. Makes me wonder if the 10900K is even tapping it out. Doom Eternal scales VERY well with CPU and it's closer to 4K scaling at 1080p than other games.
 
It's almost funny how slow brute compute progress is. I have 3.5year old 1080ti and 3080 is not even double the perf. I suppose this is why we need creative solutions to create better graphics. DLSS, ray trace, mesh shaders, sampler feedback etc. Just adding more of the same doesn't work anymore, or at least it works very slowly.

To be fair, all those extra features cost transistors. If a game were made to make full use of everything Ampere had to offer I'm sure it'd be much more than 2x the overall performance of a 1080Ti.

We will likely see those features used much more this generation now that consoles support (some of) them too.

Pascal was truly awesome but as it enters its 5th year I think its really about to start showing its age.
 
To be fair, all those extra features cost transistors. If a game were made to make full use of everything Ampere had to offer I'm sure it'd be much more than 2x the overall performance of a 1080Ti.

We will likely see those features used much more this generation now that consoles support (some of) them too.

Pascal was truly awesome but as it enters its 5th year I think its really about to start showing its age.

Let's say those extra features cost something like 20% of the area. Adding 20% more compute wouldn't do much even with linear scaling. Using that 20% for ray tracing and dlss on the other hand does a lot. I really hope mesh shaders also turn out to be useful.
 
I think with most games at 4k the 3080 is more than 2x faster than the 1080ti. Doom Eternal is close to 2x as fast at 1080p. I have a feeling that some games at 1080p may actually be cpu limited, even with a 10900K. 10900K seems to outperform the Ryzen 3900X by a significant margin with a 3080. Makes me wonder if the 10900K is even tapping it out. Doom Eternal scales VERY well with CPU and it's closer to 4K scaling at 1080p than other games.

I based my statement on this tweet posted earlier in this thread:

 
I based my statement on this tweet posted earlier in this thread:



side by side comparison against 2080Ti

So the 3080 is substatially faster then the 2080TI in rasterization, while beig 5 degress celcius cooler. Thats not thinking about RT which obvously is going to be everywhere seeing next gen consoles games (basically ALL ps5 games have it).
It's also cheaper then a 2080Ti. I think its a very good release, and bodes well for what we get from AMD.
 
1000 CAD for 1 of these custom cards -- ahh decisions decisions. More memory isn't really going to solve things right?certainly will increase the price further I suspect.

Are people waiting for the super?
 
Back
Top