Professional Testing: SPECviewperf, LuxMark, Cinebench
If the Radeon Vega Frontier Edition suffers a bit on the gaming side of things, maybe it does better in the professional application environment. To test this, we put the Vega FE up against the current top card from NVIDIA's Titan line, the Titan Xp, and ran them through SPECviewperf, LuxMark and Cinebench. Quadro and Radeon Pro Duo results are included for reference as well.
Keep in mind that NVIDIA segments the market to differentiate between GeForce and Quadro products, disabling some features in the consumer driver, leaving them exclusively on the more expensive products. AMD knows this and is hoping to take advantage of it by NOT disabling all or some of those similar capabilities and features in its Vega FE driver.
SPECviewperf 12.1.1
SPECviewperf 12 is a benchmarking application centered around workstation graphics performance. Both OpenGL and DirectX performance is measured by the various workloads or "viewsets." It is important to note that the AMD Radeon Pro Duo seemed to be running in single GPU mode for these tests. SPECviewperf does not provide any controls to target additional GPUs.
3DS Max Viewset (3dsmax-05)
The 3dsmax-05 viewset was created from traces of the graphics workload generated by 3ds Max 2016 using the default Nitrous DX11 driver.
The models for this viewset came from the SPECapc for 3ds Max 2015 benchmark and other sources. In order to best approximate real-world use cases, several tests incorporate multiple viewsets on screen, each using a different rendering method. The styles of rendering in the viewset reflect those most commonly used in major markets, including realistic, shaded and wireframe. Some lesser-used but interesting rendering modes such as facets, graphite and clay are also incorporated. The animations in the viewset are a combination of model spin and camera fly-through, depending on the model.
The Vega FE has a good showing on this viewset, producing a score 92% higher than the Radeon Pro Duo (single GPU) and coming within 20% of the Titan Xp.
CATIA Viewset (Catia-04)
The catia-04 viewset was created from traces of the graphics workload generated by the CATIA V6 R2012 application from Dassault Systemes. Model sizes range from 5.1 to 21 million vertices.
The viewset includes numerous rendering modes supported by the application, including wireframe, anti-aliasing, shaded, shaded with edges, depth of field, and ambient occlusion
The two Radeon cards scale well with Catia, with the Vega Frontier Edition getting a win over the Titan Xp by 25% or so. But notice that the Quadro P5000, which is essentially identical hardware to the GTX 1080, is noticeably faster than the Titan Xp, indicate work and improvement from the Quadro driver stack.
Creo viewset (Creo-01)
The creo-01 viewset was created from traces of the graphics workload generated by the Creo 2™ application from PTC. Model sizes range from 20 to 48 million vertices.
The viewset includes numerous rendering modes supported by the application, including wireframe, anti-aliasing, shaded, shaded with edges, and shaded reflection modes.
Creo uses a super high poly count model for a portion of its workload and the Radeon Vega Frontier Edition does exceedingly well, besting the Titan Xp by 48%.
Energy Viewset (Energy-01)
The energy-01 viewset is representative of a typical volume rendering application in the seismic and oil and gas fields. Similar to medical imaging such as MRI or CT, geophysical surveys generate image slices through the subsurface that are built into a 3D grid. Volume rendering provides a 2D projection of this 3D volumetric grid for further analysis and interpretation.
At every frame, depending on the viewer position, a series of coplanar slices aligned with the viewing angle are computed on the CPU and then sent to the graphics hardware for texturing and further calculations such as transfer function lookup, lighting and clipping to reveal internal structures. Finally, the slices are blended together before the image is displayed.
The Energy viewset gives the Radeon Vega FE another win over the Titan Xp as well as the Quadro P5000.
Maya viewset (maya-04)
The maya-04 viewset was created from traces of the graphics workload generated by the Maya 2013 application from Autodesk. Model size is 727,500 vertices.
The viewset includes numerous rendering modes supported by the application, including shaded mode, ambient occlusion, multi-sample anti aliasing, and transparency.
The Maya workload shows good scaling from the single GPU Radeon Pro duo test to the Vega FE (75%) though the Titan Xp still has a significant advantage over the rest of the field.
Medical Viewset (Medical-01)
The medical-01 viewset is representative of a typical volume rendering application that renders a 2D projection of a 3D volumetric grid. A typical 3D grid in this viewset is a group of 3D slices acquired by a scanner (such as CT or MRI).
At every frame, depending on the viewer position, a series of coplanar slices aligned with the viewing angle are computed on the CPU and then sent to the graphics hardware for texturing and further calculations, such as transfer function lookup, lighting and clipping to reveal internal structures. Finally, the slices are blended together before the image is displayed.
Using the Medical viewset swaps the win back in favor of AMD, with the Radeon Vega FE card providing a 40% advantage over the Titan Xp.
Showcase Viewset (showcase-01)
The showcase-01 viewset was created from traces of Autodesk’s Showcase 2013 application. The model used in the viewset consists of 8 million vertices.
The viewset is the first viewset in SPECviewperf to feature DX rendering. Rendering modes included in the viewset include shading, projected shadows, and self-shadows.
The Showcase viewset utilizes DX11 for rendering and the advantages that the GeForce products offer here stand out. The Titan Xp is 47% faster than the Vega FE.
Siemens NX (snx-02)
The snx-02 viewset was created from traces of the graphics workload generated by the NX 8.0 application from Siemens PLM. Model sizes range from 7.15 to 8.45 million vertices.
The scores for the Radeon Pro Duo and the Titan Xp, in contrast with the scores from the Quadro and Vega cards, indicate there is a significant software difference between these driver configurations. The Vega Frontier Edition does impressively well, coming within 30% of the Quadro P5000.
Solidworks viewset (sw-03)
The sw-03 viewset was created from traces of Dassault Systemes’ SolidWorks 2013 SP1 application. Models used in the viewset range in size from 2.1 to 21 million vertices.
The viewset includes numerous rendering modes supported by the application, including shaded mode, shaded with edges, ambient occlusion, shaders, and environment maps.
Finally, the Solidworks viewset has the Radeon Vega FE well ahead of the Titan Xp (73%) but falling shore of all three of the Quadro family of products.
LuxMark 3.1
GPGPU compute performance is a big part of any modern GPU design, especially in the workstation environment. LuxMark is a long-standing OpenCL benchmark, based on the LuxRender engine and provides a good look at how different GPU architectures compare in typical OpenCL workloads. Today we are testing our field of graphics cards in the most compute intensive scene, Hotel.
With a score of 4690, the Radeon Vega Frontier Edition performs 41% faster than the Quadro P5000 (GTX 1080 equivalent) and than the Radeon Pro Duo running on a single GPU (essentially a Fury X). That’s a big shift from the gaming results we just went through on the preceding pages. NVIDIA’s Titan Xp though was able to bring a score of 5800, giving it a 23% advantage over AMD’s middle-level pro-sumer graphics offering.
Cinebench R15 OpenGL
The performance depends on various factors, such as the GPU processor on your hardware, on the drivers used. The graphics card has to display a huge amount of geometry (nearly 1 million polygons) and textures, as well as a variety of effects, such as environments, bump maps, transparency, lighting and more to evaluate the performance across different disciplines and give a good average overview of the capabilities of your graphics hardware. The result is measured in frames per second (fps). The higher the number, the faster your graphics card is.
We quickly tossed in CineBench R15 as an OpenGL rending test and the Radeon Vega Frontier Edition does very well, scoring 151.85 FPS compared to the Titan Xp at 144.19 FPS.
We do plan to run a more extensive set of professional application tests as time permits. For this review, we focused most of our available time on the gaming angle of this architecture.
Adding this stupid, manually
Adding this stupid, manually switchable gaming mode was a terrible, TERRIBLE marketing idea. This will take a big fat chunk out of the RX sales.
…oh yes, and great review
…oh yes, and great review as usual…ahem. 🙁
Terrible? It costs much more
Terrible? It costs much more than RX. So it wouldn’t be able to affect RX sales.
Well, good thing that you
Well, good thing that you didn’t make it your final review, however..
Vega performance rather is fishy more than anything.
It barely performs better than a Fury X with drastic increases in clocks as the Fury X would outright beat that with such clocks and is an older generation with older computing even if it was to just be a die shrink, those NCU’s are not even showing their prowess, we heard many times that those are probably Fiji drivers and that TBR is disabled therefor VERY bad to judge from as of yet.
I would take all of this as a grain of salt, Polaris to Hawaii shown it’s improvements by having less SU’s and half less the ROP’s with slightly higher clock speeds as we still have much less rendering yet better performance.
The power figures are actually pretty good considering how highly clocked it is compared to the old Fury X and still consume nearly the same if not slightly more than it.
It’s pretty much like they are sand bagging at this point, or it would look like they are because people are reviewing an incomplete piece of technology and ruining it’s marketing for it, there’s a reason why AMD didn’t send samples to anyone and despite not doing it, because reviewers are doing it like this, it might have killed some people interest into buying it later because the majority are pretty ignorant in the first place.
Final thoughts on the matter though, I’m sure they shipped Vega as it is like this because optimizing productive software is much less of a hassle compared games thus, they started to want making profits earlier, from the results of certain software we can see that Vega performs extremely well in some and lose in others, to be quite fair against it’s competitor, it looks like a very compelling card and we’ve just got to have benchmarks in a very early stage that we shouldn’t have.
Please, just don’t think Vega is a flop yet with all those factors in place.
Good write up Ryan
well done
Good write up Ryan
well done on putting out some clarification points to the obvious questions that always arise with these new products. This is a beast of a GPU with all sorts of new hardware goodness in there from the materials used right down to the finished product it oozes quality and quantity. I can see this card retailing for around $1500 in Australia.
Nice any chance you can run
Nice any chance you can run the DeepBench that AMD showed off before.
Why does the chart wrongly
Why does the chart wrongly show the 1080Ti as not having the same G5X interface as Titan Xp and 1080 vanilla? Mistake.
To me, the most important
To me, the most important question is, is this card going to support 10bit output for openGL, for work in Photoshop. Because if I have a nice 10bit monitor (like BenQ SW320) I would like to use 10bit color in Photoshop. But Photoshop usually requires you to have Quadro or FuryPro to use 10 bit color. What about this card?
Thanks!
Very disappointing.
Expensive
Very disappointing.
Expensive and slow?
I have just registered to
I have just registered to tell you how much I liked your review.
I simply can’t believe Vega is just a die-shrink of Fiji. Which would be completely odd, because they could have at least used Polaris as a basis. My bet is also on a lot of features not turned on yet. I have a suggestion to make, to see if that assumption is true: could you test Deus Ex: Mankind Divided on both Vega and a Fiji card, possibly even at the same clocks? We have seen a demonstration of the HBCC in that game and if the performance level would be the same on both cards are similar clocks, it would appear to me a driver issue.
Would you be able to test
Would you be able to test Prey @ 4k? Raja said 2 RX Vega were running above 60 FPS. Curious to see how well FEs run compared to RX Vega.
jul 2016
AMD puts massive
jul 2016
AMD puts massive SSDs on GPUs and calls it SSG
https://semiaccurate.com/2016/07/25/amd-puts-massive-ssds-gpus-calls-ssg/
Ryan – Can you ask if these
Ryan – Can you ask if these will work for VDI on Server 2016? (or for that matter any of AMD’s S series cards) and where to get drivers for Server 2016?
AMD doesn’t seem to offer drivers for Server 2016, but they advertise their VDI cards as working with Windows Server. I would like to know if it works for Remote FX or Descrete Device Assignment for GPU’s and what kind of performance one can expect from that configuration.
Thanks.
Many thanks PC Perspective
Many thanks PC Perspective for putting this awesome bench marking piece together!
One thing I’m really wondering about is the cards performance in Folding@home. As a heavy folding@home donor I am in need of upgrading my build and am seriously considering Vega as my upgrade path. If you do a follow-up piece any chance you could throw in some Folding@home benchmarking?
If anyone can comment on
If anyone can comment on Folding@home performance that would be appreciated.
Depending on the workload, as
Depending on the workload, as an eGPU, which shouldn’t make much difference since F@H is such low bandwidth, I am getting 65k to 71k PPD. Am I disappointed? Yes. Is it way better than my W7000 I was using before, for sure. Luckily I don’t use this card for a F@H server farm. If I were pulling 100k PPD or close to it, I would be more incentivize to leave my computer on churning out folds 24/7.
Can’t believe you’re
Can’t believe you’re defending AMD for calling it “not a gaming card”. That’s a load of BS. And the Titans are not gaming cards! They are for deep learning. How much is AMD paying you to write this propaganda?
make a test undet DX 12….
make a test undet DX 12…. and not DX 11. newbies.
and you will see how nvidia sucks against VEGA