Since this is the highest tier of consumer graphics hardware, we also include some professional tests. Nvidia supports more applications, and particularly in ray tracing and path tracing software, it maintains an easy lead. But differences in their approach to drivers and driver enablements mean there are certain professional tasks where AMD's consumer GPUs are better than Nvidia's GeForce cards.
SPECviewperf 2020 consists of eight different benchmarks. We've also included an "overall" chart that uses the geometric mean of the eight results to generate an aggregate score. Note that this is not an official score, but it gives equal weight to the individual tests and provides a high-level overview of performance. However, few professionals use all of these programs, so it's more useful to look at the results for the applications you actually run.
Interestingly, unless we include the Titan RTX (which has a few extra driver enablements compared to GeForce), AMD's RX 6950 XT came in first overall with our aggregate score. Several of the tests favor AMD hardware, sometimes by a wide margin, while others favor Nvidia GPUs. energy-03, medical-03, and especially snx-04 go to AMD's RX 6950 XT over the RTX 3090 Ti. Again, if the applications in SPECviewperf match up with something you need to run, that could be the final deciding factor. However, professionals will often benefit even more from using an Nvidia RTX A-series card or AMD's Radeon Pro W6000-series.
Blender is a popular rendering application that has been used to make full-length films. We're using the latest Blender Benchmark, which supports Blender 3.10 and has three tests. Blender 3.10 is important because it can use the ray tracing hardware on both AMD and Nvidia GPUs.
Much like the DXR gaming tests, AMD fell far behind Nvidia's offerings in Blender 3.1. We didn't run these tests on the RTX 3070 or RX 6700 cards yet, but the pattern is pretty clear. Blender 3.10 was over twice as fast on the RTX 3080 compared to the RX 6950 XT, and even the Titan RTX (roughly equal to RTX 3070 performance) was substantially faster.
[Note: We're still looking for a good AI/machine learning benchmark, "good" meaning it's easy to run, preferably on Windows systems, and that the results are relevant. We don't want something that only works on Nvidia GPUs or AMD GPUs, or that requires tensor cores. Ideally, it will use tensor cores if available (Nvidia RTX and Intel Arc) or GPU cores if not (GTX GPUs and AMD's current consumer lineup). If you have any suggestions, please contact me — DM me in the forums, or send me an email. Thanks!]
- MORE: Best Graphics Cards
- MORE: GPU Benchmarks and Hierarchy
- MORE: All Graphics Content