Wednesday, January 22nd 2025
NVIDIA GeForce RTX 5090 3DMark Performance Reveals Impressive Improvements
The RTX 50-series gaming GPUs have the gaming community divided. While some appreciate the DLSS 4 and MFG technologies driving impressive improvements in FPS through AI wizardry, others are left disappointed by the seemingly poor improvements in raw performance. For instance, when DLSS and MFG are taken out of the equation, the RTX 5090, RTX 5080, and RTX 5070 are around 33%, 15%, and 20% faster than their predecessors respectively in gaming performance. That said, VideoCardz has tapped into its sources, and revealed the 3DMark scores for the RTX 5090 GPU, and the results certainly do appear to exceed expectations.
In the non-ray traced Steel Nomad test at 4K, the RTX 5090 managed to score around 14,133 points, putting it roughly 53% ahead of its predecessor. In the Port Royal test, which does utilize ray tracing, the RTX 5090 raked in 36,667 points - a 40% improvement over the RTX 4090. The results are much the same in the older Time Spy and Fire Strike tests as well, indicating at roughly a 31% and 38% jump in performance respectively. Moreover, according to the benchmarks, the RTX 5090 appears to be roughly twice as powerful as the RTX 4080 Super. Of course, synthetic benchmarks do not entirely dictate gaming performance, and VideoCardz clearly mentions that gaming performance (without MFG) will witness a substantially more modest improvement. There is no denying that Blackwell's vastly superior memory bandwidth is helping a lot with the synthetic tests, with the 33% extra shaders doing the rest of the work.
Source:
VideoCardz
In the non-ray traced Steel Nomad test at 4K, the RTX 5090 managed to score around 14,133 points, putting it roughly 53% ahead of its predecessor. In the Port Royal test, which does utilize ray tracing, the RTX 5090 raked in 36,667 points - a 40% improvement over the RTX 4090. The results are much the same in the older Time Spy and Fire Strike tests as well, indicating at roughly a 31% and 38% jump in performance respectively. Moreover, according to the benchmarks, the RTX 5090 appears to be roughly twice as powerful as the RTX 4080 Super. Of course, synthetic benchmarks do not entirely dictate gaming performance, and VideoCardz clearly mentions that gaming performance (without MFG) will witness a substantially more modest improvement. There is no denying that Blackwell's vastly superior memory bandwidth is helping a lot with the synthetic tests, with the 33% extra shaders doing the rest of the work.


75 Comments on NVIDIA GeForce RTX 5090 3DMark Performance Reveals Impressive Improvements
Nvidia are milking Pros with their AI chips and Quadro GPUs, but $2000 for a Gaming GPU is a lot. I can definitely buy one but it's not cheap for sure. And that's the only true 4K card of the RTX 50 series. The 5080 is not even beating the 4090...
$400 more is too much honestly. In 2022 we had just got out of the pandemic and there were some chip shortages still but the 4090 was $100 more even though TSMC 4N was much more expensive than Samsung 8nm and the 4090 had 52% more CUDA Cores than the 3090 Ti and 60% more than the 3090 and they increased the L2 Cache from 6MB to 72MB too, so I think the 4090 was actually more expensive to manufacture than the 5090 back then, but RDNA3 was supposed to be very competitive so Nvidia had to go almost all-in. The big plus of the 5090 are the 512-bit bus, GDDR7 memory and the 2-slot cooler but I don't think it's more expensive than the 4090 was back then.
So to give you the short answer, it's not a maxxed out die, there is room for improvement so long as TSMC can refine the process enough to make more perfect wafers.
Each GB200 GPU has 192GB of HBM3E with a 4096-bit bus and a Memory Bandwidth of 8TB/s, whereas the 5090 has a GB 202 chip with 32GB GDDR7 on a 512-bit bus and a memory Bandwidth of ~1.8TB/s.
On the other hand the QUADRO GPUs (made for Workstations and Professionals) will use the same GB202 chip as the 5090. But they will probably pack either 64GB or 96GB of GDDR7 and have a 190SM enabled (2 disabled for better yields).
Anyhow, if they have enough leftover chips that the enterprise is not in a rush to buy, they could sell those to consumers with a smaller profit, but I don't think it'll be the case.The Quadro/Tesla GPUs are also AI chips, fwiw. Albeit less capable than a GB200, they're way cheaper and make for great inference devices, or even training if your model is not that big.
I guess it gets a bit ambiguous, since the highest-end chip (those x100/x200 ones) often have a somewhat different architecture from the rest of the lineup (with FP64 and HBM, as you said, and lacking on the RT cores and any other graphics-relevant components), but one could also mean "full die" as in the full GB202 die.
But I don't think we'll see a full GB202 or almost like 190SM (out of 192) on a Consumer GPU this generation. They want to keep the best chips for Workstations and AI and charge a Premium for it (~$7000+ for Quadro and GB200 are $30k-40k)
Same chips, and often same configs, just different form factor for different uses.Given how we didn't see this happening with the AD102 either, I totally agree.No, the only Ampere on TSMC 7 was the A100, which is not really comparable to the GA102 and below, all of which were on Samsung 8nm (be it quadro or tesla).
Anyhow, bar the x100 chips, yeah, the rest of the lineup shares the chips among the geforce/quadro/tesla lineup.
some leaks point that x90 top bin dies will become 96GB Pro GPU for AI, clam shell design with 3gb memory modules.
You can even desolder the chip from one of those GPUs and solder into the board from another, this has been done more than once.
Given that it's also open source, you can easily see that there's no distinction among the different binnings of each chip, their HAL and other related functions make no distinction between those. Have those as examples:
github.com/NVIDIA/open-gpu-kernel-modules/blob/54d69484da940413ba6017df9cf43bed8b75da46/src/nvidia/src/kernel/gpu/gsp/arch/ada/kernel_gsp_ad102.c
github.com/NVIDIA/open-gpu-kernel-modules/blob/54d69484da940413ba6017df9cf43bed8b75da46/src/nvidia/generated/g_fbsr_nvoc.c#L117
Nouveau, the open source driver upstream driver, is pretty useless for most cases.
FWIW, even on windows you can add extra stuff from the quadro/tesla drivers into the geforce ones, and vice-versa. I've done so while helping the community with the vGPU mods for the Turing GPUs.