Friday, January 24th 2025
New Leak Reveals NVIDIA RTX 5080 Is Slower Than RTX 4090
A set of newly leaked benchmarks has revealed the performance capabilities of NVIDIA's upcoming RTX 5080 GPU. Scheduled to launch alongside the RTX 5090 on January 30, the GPU was spotted on Geekbench under OpenCL and Vulkan benchmark tests—and based on the performance, it might not make it among the best graphics cards. The tested device was an MSI-branded RTX 5080 labeled as model MS-7E62. This setup had AMD's Ryzen 7 9800X3D processor, which many consider one of the best CPUs for gaming. It also included an MSI MPG 850 Edge TI Wi-Fi motherboard and 32 GB of DDR5-6000 memory.
The benchmark results show that the RTX 5080 scored 261,836 points in Vulkan and 256,138 points in OpenCL tests. Compared to the RTX 4080, its previous version, the RTX 5080 has a 22% boost in Vulkan performance and a small 6.7% gain in OpenCL. Reddit user TruthPhoenixV found that on the Blender Open Data platform, the GPU got a median score of 9,063.77. This score is 9.4% higher than the RTX 4080 and 8.2% better than the RTX 4080 Super. Even with these improvements, the RTX 5080 might not outperform the current-gen top-tier RTX 4090. In the past, NVIDIA's 80-class GPUs have beaten the 90-class GPUs from the previous generation, but these early numbers suggest this trend might not continue for the RTX 5080.The RTX 5080 uses NVIDIA's latest Blackwell architecture, with 10,752 CUDA cores spread across 84 Streaming Multiprocessors (SMs) versus the 9,728 cores in the RTX 4080. It has 16 GB of GDDR7 memory on a 256-bit bus. NVIDIA says it can deliver 1,801 TOPS in AI performance through Tensor Cores and 171 TeraFLOPS of ray tracing performance using its RT Cores.
That said, it's important to note that these benchmark results have not been fully verified so we should wait for the review embargo to lift before concluding.
Sources:
DigitalTrends, TruthPhoenixV
The benchmark results show that the RTX 5080 scored 261,836 points in Vulkan and 256,138 points in OpenCL tests. Compared to the RTX 4080, its previous version, the RTX 5080 has a 22% boost in Vulkan performance and a small 6.7% gain in OpenCL. Reddit user TruthPhoenixV found that on the Blender Open Data platform, the GPU got a median score of 9,063.77. This score is 9.4% higher than the RTX 4080 and 8.2% better than the RTX 4080 Super. Even with these improvements, the RTX 5080 might not outperform the current-gen top-tier RTX 4090. In the past, NVIDIA's 80-class GPUs have beaten the 90-class GPUs from the previous generation, but these early numbers suggest this trend might not continue for the RTX 5080.The RTX 5080 uses NVIDIA's latest Blackwell architecture, with 10,752 CUDA cores spread across 84 Streaming Multiprocessors (SMs) versus the 9,728 cores in the RTX 4080. It has 16 GB of GDDR7 memory on a 256-bit bus. NVIDIA says it can deliver 1,801 TOPS in AI performance through Tensor Cores and 171 TeraFLOPS of ray tracing performance using its RT Cores.
That said, it's important to note that these benchmark results have not been fully verified so we should wait for the review embargo to lift before concluding.
144 Comments on New Leak Reveals NVIDIA RTX 5080 Is Slower Than RTX 4090
The whole lineup will see minimal improvements, and price v. performance won’t move. Not really much purchasing incentive if you ask me, very few games need more than 4070s class performance for the large majority of consumers.
Its effectively the same generation, which is why they are leaning so heavily on DLSS4 / framegen to pretend they did anything at all.....
Nvidia said +10% from 4080, it's certainly less than 4090.
Even the 5090 with a 600w TDP is only 35% faster than a 4090, and the 5080 is half a 5090.
Tbh I'll be surprised if it even comes close to a 4090.
I suspect, however, that those chips will get sold as some 5090D variant for China at astronomical prices before they're sold to gamers.
until/unless AMD competes in the high end, this is what to expect. Higher prices, lesser performance uplifts (vs the predecessor).
Nvidia swears they can't add more raster because it's too hard, but not adding any raster improvement per watt is pretty silly because it lays bare the whole con. They're updating the AI hardware, not the graphics engine, and they're desperate to convince everyone that they had to do it. They couldn't be expected to improve raster meaningfully. It was just impossible.
They could. They just didn't and they didn't because they're an AI company now and not a graphics company.
While the 5080 is going to be better than the 5070Ti, I don't see it getting anywhere near outperforming it by 30+ % which is how much extra they are charging for it.
I think at best the 5080 beats the 5070 Ti by 15%...