Where did you get info on Blackwell's clocks? It seems to be well hidden to disclose it as far away from presentation as possible, clearly they are not great, perhaps even regression.
Where did you get info on Blackwell's clocks? It seems to be well hidden to disclose it as far away from presentation as possible, clearly they are not great, perhaps even regression.48 vs 56 SMs and Blackwell's clocks are only barely higher.
Nvidia updated their site with 5000 series specs: https://www.nvidia.com/en-me/geforce/graphics-cards/50-series/rtx-5070-family/Where did you get info on Blackwell's clocks? It seems to be well hidden to disclose it as far away from presentation as possible, clearly they are not great, perhaps even regression.
AMD will copy it soon and later Intel. So get to know multifake frames well.😛I don't know maybe I'm old school but there is something I don't like about this frame generation. Perhaps I just don't know enough about these new Blackwell GPUs yet.
Ad102 was chungus yea.Nah, the 4090 was great. I’m talking about the architecture itself. Not gimped vram and SMs on lower tiers
yeah same node, can't do magic.
SolidQ was saying AMD already did it (not officially). Nvidia taking credit for innovation yet again? 🤣AMD will copy it soon and later Intel. So get to know multifake frames well.😛
The 5090 being CPU limited I could believe. But the others? Nah...Now that low precision has been pushed to the limit, the bottleneck now is a battle of CPU draw calls and memory bandwidth.
You can only push fake frames so far without reaching a physical limit.
Cache can also help, but it matters less for transformer style compute, though I assume we are at 128MB L2 now.
Likethree things I don’t like, nothing is interesting from a RTX 40 user standpoint.
12GB on $500 cards is not good. Finally, it’s not N3E. So it’s gets a skip from me
But this would insert lag into the gaming experience? Gaming by nature has to be realtime.
I don't like predicted, because it's not a future looking operation. They are interpolations between two known frames. You are always figuring out positions in between two frame you already have.I'm under the assumption that using your analogy it only needs to predict what the screen will look like for 3 frames. So instead of computing those 3 frames, they are predicted. Assuming you are running 100fps, that's 3ms or 1 frame per ms. The 4th frame is then computed. Since only 3 ms has transpired unless there was a complete scene change the difference between the 3rd predicted frame what that actual computed 3rd frame would have looked like will be basically nothing, so then the moving from frame 3 (predicted) to frame 4 (computed) would be smooth.
I would assume that the faster the hardware can natively run the game the more predicted frame can be used. So if the game is running natively at 60fps maybe only 1 frame can be predicted between computed frames. But if it's running 150fps, then perhaps 2 or 3 frames can be predicted.
And AI!But GDDR7 might make up for it in apps and games.
Yeah but it has better more performant SM.Should be using N4P tho, which should be allowing for a decent boost in clocks. It appears to be design driven. Remember that 40 series cards were doing 2.7 easy at stock.
what even is the other option?looks like 9800X3D is being used now in benchmarks.
but 9800X3D is better(tm) and available now(r).The classic 7800X3D.
Joking, right?available now(r).
Yes, but it's still worth pointing out since the tests could have been done months ago before that was the case.but 9800X3D is better(tm) and available now(r).
you can buy n-CCD V$ in 9684X.And tell Lisa I don't like her that much anymore after she didn't give me dual CCD V-cache.
Hell no, these are always recent. Benchmarketing and comp analysis guys never sleep before launch.Yes, but it's still worth pointing out since the tests could have been done months ago before that was the case.
Yeah but it has better more performant SM.
But it does. stuff below GB202 seems pretty good?Doesn't seem like it based upon the benchmarks NV provided.