Compute Performance

As always we'll start with our DirectCompute game example, Civilization V, which uses DirectCompute to decompress textures on the fly. Civ V includes a sub-benchmark that exclusively tests the speed of their texture decompression algorithm by repeatedly decompressing the textures required for one of the game’s leader scenes.  While DirectCompute is used in many games, this is one of the only games with a benchmark that can isolate the use of DirectCompute and its resulting performance.

Compute: Civilization V

Our next benchmark is LuxMark2.0, the official benchmark of SmallLuxGPU 2.0. SmallLuxGPU is an OpenCL accelerated ray tracer that is 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 2.0

Our 3rd benchmark set comes from CLBenchmark 1.1. CLBenchmark contains a number of subtests; we’re focusing on the most practical of them, the computer vision test and the fluid simulation test. The former being a useful proxy for computer imaging tasks where systems are required to parse images and identify features (e.g. humans), while fluid simulations are common in professional graphics work and games alike.

Compute: CLBenchmark 1.1 Computer Vision

Compute: CLBenchmark 1.1 Fluid Simulation

Moving on, our 4th compute benchmark is 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, as Folding @ Home is moving exclusively OpenCL this year with FAHCore 17.

Compute: Folding @ Home: Explicit, Single Precision

Compute: Folding @ Home: Implicit, Single Precision

Our 5th compute benchmark is Sony Vegas Pro 12, an OpenGL and OpenCL video editing and authoring package. Vegas can use GPUs in a few different ways, the primary uses being to accelerate the video effects and compositing process itself, and in the video encoding step. With video encoding being increasingly offloaded to dedicated DSPs these days we’re focusing on the editing and compositing process, rendering to a low CPU overhead format (XDCAM EX). This specific test comes from Sony, and measures how long it takes to render a video.

Compute: Sony Vegas Pro 12 Video Render

Wrapping things up, our final compute benchmark is an in-house project developed by our very own Dr. Ian Cutress. SystemCompute is our first C++ AMP benchmark, utilizing Microsoft’s simple C++ extensions to allow the easy use of GPU computing in C++ programs. SystemCompute in turn is a collection of benchmarks for several different fundamental compute algorithms, as described in this previous article, with the final score represented in points. DirectCompute is the compute backend for C++ AMP on Windows, so this forms our other DirectCompute test.

Compute: SystemCompute v0.5.7.2 C++ AMP Benchmark

Civilization V Synthetics
POST A COMMENT

107 Comments

View All Comments

  • Death666Angel - Friday, March 22, 2013 - link

    "pulling 7W more than the 7770, a hair more than the 5W difference in AMD’s TBP"
    That 5W is not at the wall though. Factoring in rounding PSU efficiencies, it's very possible that the cards are only drawing 5W more. :)
    "The Sapphire card, despite being overclocked, draws 6W less than our reference 7790."
    Seeing how the Sapphire runs cooler in Furmark, that might explain a Watt or two in reduced power draw, coupled with the efficiency of the PSU, it might explain three or four even. :)
    Reply
  • pandemonium - Saturday, March 23, 2013 - link

    "NVIDIA has for a long time set the bar on efficiency, but with the 7790 it looks like AMD will finally edge out NVIDIA."

    What is your definition of a long time? As far as efficiency standards, I consider AMD to be better for the end result when looking at the full definition and application of the word. See the spreadsheet I created here about 16 months ago to understand what I mean: http://forums.anandtech.com/showthread.php?t=21507...
    Reply
  • silverblue - Saturday, March 23, 2013 - link

    You just called Ryan a "dummy", did you, without even checking the statement further down which reads:

    "For anyone looking to pick up a 7790 today, this is being launched ahead of actual product availability (likely to coincide with GDC 2013 next week). Cards will start showing up in the market on April 2nd, which is about a week and a half from now."

    If YOU had read the article, blah blah dumb idiot blah blah. As you've not replied to anybody in particular, your mistargeted rants could be construed as being directed toward the staff themselves, so keep it up and you won't HAVE to worry about what AT is reviewing in future.

    Bottom line - it's faster than the 650 Ti, it's looking to be more efficient than the 650 Ti, and oh look, both have 1GB of GDDR5 on a 128-bit memory interface, which you seem to have forgotten when you leapt down AMD's throat about the 7790, and when you went on your childish tirade about the 5770's 128-bit memory interface earlier.

    As far as I recall, Ryan didn't mention anything about when Titan was available to buy, only that it had launched. Pretty much blows your theory of Ryan hating NVIDIA out of the water, doesn't it?

    I'm not sure if I've said this before, and apologies to everybody else if I have, but I'm done with you, full stop. I can only hope everybody else here decides that not feeding the ignorance you perpetuate on every single AMD article would save them time they could be devoting to something far less boring instead.

    To the staff - is there anything you can do to introduce an Ignore List? Thanks in advance for your response.
    Reply
  • silverblue - Saturday, March 23, 2013 - link

    A note about threading - doesn't look to be stepping in consistently, so sometimes it's a little difficult to see whom replied to whom. Reply
  • CeriseCogburn - Sunday, March 24, 2013 - link

    You got eveything wrong again, and you failed to read the article not I, and you failed to read my reply addressing half your idiotic non points, so you're the non reader, fool.
    Now I have to correct you multiple times. And you're a waste.
    650TI overclocks and it's only faster in a few amd favor games which are here, of course.
    Strike one for tardboy.
    650Ti runs fine OC'd too, which it does well: " We pushed Gigabyte's GeForce GTX 650 as far as it'd go and achieved a maximum core overclock of 1125 MHz, with the GDDR5 memory operating at 1600. All it took was a 1.15 V GPU voltage. "
    http://www.tomshardware.com/reviews/geforce-gtx-65...
    The 128 bit bus - REPAYMENT for you FOOLS SQUEALING prior, what's so hard to understand ?
    Did you forget all your WHINING ?
    Did you forget your backing up the FAILED theorists with the nVidia dual speed memory ?
    ROFL
    You're up to strike 4 already.
    " Ryan didn't mention anything about when Titan was available to buy, only that it had launched. Pretty much blows your theory of Ryan hating NVIDIA out of the water, doesn't it?"
    NO, so why would it be mentioned if he didn't want anyone to buy it ? Why mention it, that would key in to save for release date, right ?
    Instead we get this gem first off in BOLD to start the article: " Who’s Titan For, Anyhow? "

    Guess that just crushed your idiot backwards bullhockey forever.
    For all you know Ryan mentioned release date anyway.

    You're not "done with me", you get everything WRONG, so you'll be opening your big fat piehole forever, that's how people like you do it. Idiot amd fanboys, all the same.

    Also a beggar child for extra "control", since you "can't be an adult and control yourself" - please give me an ignore button ! I'm a crybaby who can't handle it !
    ROFL
    Reply
  • philipma1957 - Sunday, March 24, 2013 - link

    One question does your 650Ti pays for itself? this amd will pay for itself via bitcoin. even with the asics. especially if you heat your home with electrical heat.

    nuff said
    Reply
  • Rajan7667 - Sunday, March 24, 2013 - link

    @form @LinusTech This is new New app for intel lovers. http://www.intel.com/content/www/us/en/gamers/vip-... Reply
  • colonelclaw - Sunday, March 24, 2013 - link

    To Ryan and staff
    As a long-time admirer of AnandTech, I always enjoy reading pretty much every article you post, and have immense respect for all your writers.
    However, I am now utterly fed up with the direction the comment discussions have taken. The general pattern is they start out as debates and end up as pretty nasty personal attacks that have nothing to do with the articles. You may say 'don't read the comments', to which I reply that they used to be an extension of the articles themselves, and were always a source of valuable information.
    It pains me to say this, but if you don't start removing the trolls I will no longer come to this site at all, and I would guess I am not alone in having this opinion.
    Reply
  • haze4peace - Sunday, March 24, 2013 - link

    I agree 100% and actually sent off a few emails to the staff earlier in the day. I urge others to do so as well so we can put this problem behind us. Reply
  • KnightRAF - Monday, March 25, 2013 - link

    I agree the trolls are out of control and need some pruning back. They have massively lessened my enjoyment of the site the last couple of times I've visited. Reply

Log in

Don't have an account? Sign up now