Compute

Shifting gears, let’s take a look at compute performance on GTX 1060.

As we already had the chance to categorize the Pascal architecture’s compute performance in our GTX 1080 review, there shouldn’t be any surprises here. But it will be interesting to see whether the GTX 1060’s higher ratio of memory bandwidth per FLOP materially impacts overall compute performance.

Starting us off for our look at compute is LuxMark3.1, the latest version of the official benchmark of LuxRender. LuxRender’s GPU-accelerated rendering mode is an OpenCL based ray tracer that forms a part of the larger LuxRender suite. Ray tracing has become a stronghold for GPUs in recent years as ray tracing maps well to GPU pipelines, allowing artists to render scenes much more quickly than with CPUs alone.

Compute: LuxMark 3.1 - Hotel

While GTX 1060 could hang with GTX 980 in gaming benchmarks, we don’t start off the same way with compute benchmarks, with the last-generation flagship holding about 17% ahead. Unfortunately for NVIDIA, this is about where GTX 1060 needed to be to best RX 480; instead it ends up trailing the AMD competition. Otherwise the performance gain versus the GTX 960 stands at 65%.

For our second set of compute benchmarks we have CompuBench 1.5, the successor to CLBenchmark. CompuBench offers a wide array of different practical compute workloads, and we’ve decided to focus on face detection, optical flow modeling, and particle simulations.

Compute: CompuBench 1.5 - Face Detection

Compute: CompuBench 1.5 - Optical Flow

Compute: CompuBench 1.5 - Particle Simulation 64K

Like with GTX 1080, relative performance is all over the place. GTX 1060 wins with face detection, loses at optical flow, and wins again at particle simulation. Even the gains versus GTX 960 are a bit more uneven, though at the end of the day GTX 1060 ends up being significantly faster than its predecessor with all 3 sub-benchmarks.

Moving on, our 3rd compute benchmark is the next generation release of FAHBench, the official Folding @ Home benchmark. Folding @ Home is the popular Stanford-backed research and distributed computing initiative that has work distributed to millions of volunteer computers over the internet, each of which is responsible for a tiny slice of a protein folding simulation. FAHBench can test both single precision and double precision floating point performance, with single precision being the most useful metric for most consumer cards due to their low double precision performance. Each precision has two modes, explicit and implicit, the difference being whether water atoms are included in the simulation, which adds quite a bit of work and overhead. This is another OpenCL test, utilizing the OpenCL path for FAHCore 21.

Compute: Folding @ Home Single Precision

Compute: Folding @ Home Double Precision

Finally, in Folding@Home, we see the usual split between single precision and double precision performance. GTX 1060 is solidly in the lead when using FP32, but NVIDIA’s poor FP64 rate means that if double precision is needed, RX 480 will pull ahead.

Hitman Synthetics
Comments Locked

189 Comments

View All Comments

  • Greeba77 - Wednesday, August 10, 2016 - link

    I don't subscribe to some of the bitchier comments on here and I do read Anandtech because of the depth of the reviews. I do feel a trick or two has been missed on this one, because since the 1080 and 1070 review, the results came out elsewhere to show the low level API performance on the Pascal cards seemed to be a problem (the Vulkan Doom benchmarks were awful compared to the RX480). I was hoping this review would include a deeper dive in to this as it's a potential question mark on the 1060's future proofing vs the RX480.
  • Thermalzeal - Monday, August 15, 2016 - link

    I will please ask that you use "other" review sites for your insatiable impatient needs.

    No way AnandTech will be left out because it's the only site that Industry professionals actually read these days. And if for some reason this catastrophe happened...I'm pretty sure one call to Anand, and he'll walk right into Jen Hsun's office and take the card right out of his computer. Especially now that he's just hanging out at Apple. ,
  • Colin1497 - Friday, August 5, 2016 - link

    Clearly the last couple months have been unfortunate, as whatever dynamic has happened has corresponded with the most exciting time in the GPU world in years. The question is where he can get back on track. I doubt that Ryan is unaware of all of the issues surrounding this. We will see how he responds. :)
  • just4U - Saturday, August 6, 2016 - link

    100% certain he's aware of the criticisms.. as he does comment on some of them. Also a likely reason why Daniel Williams was brought on board to help with overflow.. Although I'd really like to see Ian take a stab at the video card reviews.. (just for sh... and giggles) as processors are not so interesting these days with incremental bumps... but they all have their niche areas to work with so whatever.

    The only real criticism I have would be the waiting game.. If Ryan says something will be ready on a certain date (or in a few days..) and then it takes weeks.. well shoot that's disappointing and shouldn't happen. But hey.. I was one of those happy with the Preview the Review was just icing on the cake (even if I disagree on the 480 vs 1060 assessment)

    Ryan's good at what he does late or not, many of us look forward to Anandtech's takehome on new Video Cards.
  • ridic987 - Sunday, August 7, 2016 - link

    what about his 480 vs 1060 assessment do you disagree with?
  • just4U - Monday, August 8, 2016 - link

    He puts it at and around 10% faster than the RX480. The feel I am getting from most who have hands on experience with both cards suggest the 480 might be a little better.

    It reminds me of the 380/960 comparisons online initially.. both excellent products overall but after a lot of time with both I kind of felt ripped off by the 960 which I paid a premium for and it certainly wasn't as good as the 380.
  • Cygni - Friday, August 5, 2016 - link

    Then go start your own review website, nerd. This one ain't yours, and nobody cares about how you think it should be run.
  • Devo2007 - Friday, August 5, 2016 - link

    Actually, very few (if any) sites were sampled Titan X Pascal cards - has nothing to do with Anandtech's review schedule.
  • Nagorak - Saturday, August 6, 2016 - link

    They didn't sample anyone on Titan X it seems like.
  • Tabalan - Friday, August 5, 2016 - link

    What about RX470? It would be nice to see some decent AIB cards compared. Also, I really would like you to add 2 charts in summary - overall performance and perf/$. This would help comparing your review to other reviews without going though all benchmarks and game tests.

Log in

Don't have an account? Sign up now