OpenCL Benchmark
Your rating: Now say why...

0

Terminal app for benchmarking your system.   Free
Add to my Watch List
Email me when discounted
  • Download
    13.5 MB
  • Developer
    Andreas Michalak
OpenCL Benchmark, for Snow Leopard only
  • NVIDIA: 8xxx(M), 9xxx(M), GT1xx, GT2xx, GTX2xx
  • ATI: 4850, 4870 - maybe 10.6.2 update needed to run withput failures OPENCL - background information :
    • OpenCL is an API for universal GPU(CPU) computing , Mac OS X 10.6 supports that
    • Main difference to CUDA / ATI STEAM is: both APIs ate vendor specific. A CUDA (NV) app like badaboom(h264 on GPU) can't work on an ATI gpu and vice versa
    • OpenCL is universal means:
    • Xcode compiles an code which includes the source (C like) for the gpu That source is , different to
  • What's New
    Version V2.0:
    • NEW: added 3 OpenCl benches/demos from Nvidia (may run on ATI latzer with 10.6.2)
    • 1. Teapot - very much difference between CPU and GPU
    • 2. Memory Bandwithtest : shows CPU<>GPU bottleneck for memory transfers over PCIe
    • 3. Volume Renderer
    Requirements
    • Mac OS X 10.6 only
    • Intel CPUs : Core2Duo and Xeon Series, not Core Duo
    • OpenCL GPUs : Mac 2008+ with NVIDIA GPUs (Macbook /MBP/iMac GT120/MacPro)... Macs 2008+ with ATI GPUs - Mac OS X 10.6.2+ recommended (soon available, ATI works hard on OpenCL fixes)



    MacUpdate - OpenCL Benchmark



    OpenCL Benchmark User Discussion (Write a Review)
    ver. V 2.x:
    Your rating: Now say why...
    Overall:

    sort: smiles | time
    burypromote
    +1

    -1
    L008com commented on 03 Nov 2010
    Download link sends me to some crappy rapidshare page where I have to wait 15 minutes before I can download this app. Hows about real download links only on MacUpdate? Sounds like a good plan.
    [Version V 2.0]


    burypromote

    +85
    Mitch_de commented on 16 Apr 2010
    Better use smallluxGPU for OpenCL benches now !
    Its also on macupdate.com does much better OpenCL benching.
    [Version V 2.0]


    burypromote
    Robert Baruch commented on 02 Nov 2009
    A rapidshare download? Really?

    Could this be released on, say, SourceForge? And under, say, an open source license?

    I would love to be able to see the source code so that I could learn from it. I haven't yet been able to download the zip because "there aren't available free user slots" on rapidshare :(
    [Version V 2.0]


    burypromote
    N00b32 commented on 09 Oct 2009
    Hi,

    I added a table for better comparison of the OpenCL benchmarks:

    http://wiki.osx86project.org/wiki/index.php/OpenCL

    What would be the best benchmark for evaluation of OpenCL performance? Galaxies?

    @mitch_de:

    could you provide in this standard benchmark a build number visible while benchmarking (better comparison)?

    Thanks

    Jason
    [Version V 2.0]


    burypromote

    +85
    Mitch_de commented on 05 Oct 2009
    from dev:
    Ibuild an new Galaxis V2 32K. Apple updated OpenCL API usage , perhaps helps ATI to run ? But i think ATI will need really 10.6.2 update.
    Others (NVIDIA) will have a bit higher GPU GigaFlops with that version!
    Be patient in CPU modes - CPU has much work to compute the 32000 stars. GPU runs much faster here but also has less that 20 FPS (GTX285) 9 FPS 8800GT.

    Link (will be official if some report : work)
    http://rapidshare.com/files/286234291/Galaxies_32K_V2.zip
    Users with Mobile Nvidia should better use the also new V2 8K version.
    GigaFlops for that mobile GPUs same as with 32K version, but more FPS especially in CPU mode
    on lowend C2D mobile

    http://rapidshare.com/files/286235157/Galaxies_8K_V2.zip
    [Version V 2.0]


    burypromote

    +11
    MikeMyDay commented on 10 Sep 2009
    Still not working on my MacPro with HD4870.

    Console outputs:
    Build Log for Device[0]:
    then after 15 seconds or so:
    setupComputeDevices() failed: -7
    and then I'll have to force quit the app.

    Hope the X.6.1 update will fix the ati drivers.


    PS.
    Schöne Grüße aus Bonn ;)
    [Version V 1.0]

    3 Replies

    burypromote

    +85
    mitch_de replied on 11 Sep 2009
    OpenCL Framework got noz updated with 10.6.1. ATI users must wait for an 10.6.2 or an extra Apple / ATI OpenCL Update.
    Not really an problem for waiting, because no OpenCL application is on the market and i think the first will be developed end of 2009, where ati users sure got their OpenCL Framework fixed by Apple/ATI.
    burypromote

    +52
    RiseOfChaos replied on 12 Sep 2009
    I think Pixelmator 1.5 support OpenCL :-)
    burypromote

    +85
    Mitch_de replied on 20 Sep 2009
    No, Pixelmator, aner very good Photoapplication "only" supports CoreImage. CoreImage also uses GPU for computing, but only for doing pictures. CoreImage cant do any universal computing (by programming in OpenCL) . OpenCL focus is different. Also GPU requirements are much higher than CoreImage!
    burypromote
    +1

    +1
    Ben Papworth commented on 10 Sep 2009
    This is fantastic.

    I'm lucky enough to own a 17" MacBook Pro - and it's very interesting that I can use both GPUs at once (9400M as renderer and 9600M as compute engine), as I didn't think this would be possible.

    9600M = 46 GFLOPS
    9400M = 21 GFLOPS
    Hybrid = 25 GFLOPS

    Given the speed of calculations, it looks like hybrid CPU+GPU must be using 9400M for GPU compute (or hybrid is not nearly as efficient). It would be a great addition if the benchmark gave two hybrid options (one for each GPU), or identified which GPU it was using.

    On another note, do you think it's possible to make Galaxies use multiple GPUs for compute? I realise that this is a big ask - as you've already indicated that the 9400M GPU can only process 4K elements...

    Keep up the good work!
    [Version V 1.0]

    1 Reply

    burypromote

    +85
    mitch_de replied on 10 Sep 2009
    Thanks !
    But main work is made by Apple for the Galaxies ;)
    Running parallel on more than one GPU is more complicated , but i will take a look in this direction.
    burypromote

    +85
    mitch_de commented on 08 Sep 2009
    by uploader:
    Old OpenCL "Benchmark" was removed and now its replaced by Galaxies OpenCL Demo-Benchmark V1.0
    - Galaxies is an much more real world OpenCL Demo and Benchmark
    - no more an terminal application - OpenGL fullscreen support
    - computes an universum star szene on CPU, GPU and Hybrid (CPu+GPUs) - switch with key s between that modes
    - does measure computed GigaFlops and sure you also will see the speed in the moving speed of the stars - on an MacPro 1,1 an NVDIA GTX 285 is about 5 times faster - soon up to 8 times than CPUs
    - do not use or report OpenCL V0.25 results anymore, the new GigaFlop results are much, much more real for OpenCL speed
    - results, questions & answers you can find on the home forum:
    http://www.insanelymac.com/forum/index.php?showtopic=182874
    - happy benching !
    [Version V 1.0]


    burypromote

    -9
    sommop87 commented on 06 Sep 2009
    Very good speed.
    [Version V025]


    burypromote
    Minorgroove commented on 06 Sep 2009
    Nice speedup on my early 2008 MacBook Pro 2.6 Hz machine with GeForce 8800M GT.

    ...........................................................
    .................. OpenCL Bench V 0.25 by mitch ...........
    ...... C2D 3GHz = 12 sec vs Nvidia 9600GT = 0,93 sec ......
    ... time results are not comparable to older version! .....
    ...........................................................

    Number of OpenCL devices found: 2
    OpenCL Device # 0 = GeForce 8600M GT
    Device 0 is an: GPU with max. 940 MHz and 32 units/cores
    Now computing - please be patient....
    time used: 2.958 seconds

    OpenCL Device # 1 = Intel(R) Core(TM)2 Duo CPU T9500 @ 2.60GHz
    Device 1 is an: CPU with max. 2600 MHz and 2 units/cores
    Now computing - please be patient....
    time used: 14.791 seconds

    Now checking if results are valid - please be patient....
    :) Validate test passed - GPU results=CPU results :)
    [Version V025]


    burypromote
    김 승균 had trouble on 08 Sep 2009
    ...........................................................
    .................. OpenCL Bench V 0.25 by mitch ...........
    ...... C2D 3GHz = 12 sec vs Nvidia 9600GT = 0,93 sec ......
    ... time results are not comparable to older version! .....
    ...........................................................

    Number of OpenCL devices found: 5
    OpenCL Device # 0 = GeForce GT 120
    Device 0 is an: GPU with max. 1400 MHz and 32 units/cores
    Now computing - please be patient....
    time used: 1.629 seconds

    OpenCL Device # 1 = GeForce GT 120
    Device 1 is an: GPU with max. 1400 MHz and 32 units/cores
    Now computing - please be patient....
    time used: 3.984 seconds

    OpenCL Device # 2 = GeForce GT 120
    Device 2 is an: GPU with max. 1400 MHz and 32 units/cores
    Now computing - please be patient....
    time used: 1.743 seconds

    OpenCL Device # 3 = GeForce GT 120
    Device 3 is an: GPU with max. 1400 MHz and 32 units/cores
    Now computing - please be patient....
    time used: 1.636 seconds


    Error: clGetDeviceInfo CL_DEVICE_NAME
    ERROR NUMBER = -33
    BUFFER =GeForce GT 120

    When testing CPU, i got this error message.
    Program is tested on MAC PRO (Quad cores)x2 with GeForece GT 120x4
    [Version V025]


    burypromote
    stevenlpeterson had trouble on 31 Aug 2009
    yea i have the ATI Radeon 4870 and its reporting terrible speeds...very similar to your results. wtf

    Number of OpenCL devices found: 2
    OpenCL Device # 0 = Radeon HD 4870
    Device 0 is an: GPU with max. 750 MHz and 4 units/cores
    Now computing - please be patient....
    time used: 4.226 seconds

    OpenCL Device # 1 = Intel(R) Xeon(R) CPU W3540 @ 2.93GHz
    Device 1 is an: CPU with max. 2925 MHz and 8 units/cores
    Now computing - please be patient....
    Segmentation fault
    logout

    [Process completed]

    lame...
    [Version V025]


    burypromote

    +3
    armadillu had trouble on 27 Aug 2009
    Says: Bad CPU type in executable

    MacBookPro1,1
    Intel Core Duo 2,16 Ghz
    Chipset Model: ATY,RadeonX1600
    [Version V025]

    2 Replies

    burypromote

    +85
    mitch_de replied on 27 Aug 2009
    Bench compiled with on speed optimisattion for C2D+ CPU´s.
    Some features your CD didnt have.
    Also your GPU cant to only OpenCL commands - only much newer gpu cant run OpenCL.
    Easy to remember: An GPU which cant do CoreImage, cant do OpenCL too.
    So you missed nothing real, because you wont get any GPU OpenCL time values, even i would recompile the bench for CD cpus also.

    I will change the CPU needs from Intel CPU to Intel C2D, Intel XEON Series and an GPU which allows OpenCL. (near all Macs build 2008 and later).
    burypromote

    +3
    armadillu commented on 27 Aug 2009
    Actually I can run coreImage filters and shaders, but just just found that only a few of the newer cards can do openCL:

    http://www.macrumors.com/2009/06/10/snow-leopard-h-264-hardware-acceleration-and-opencl-requirements/
    burypromote
    +1

    +1
    Ive had trouble on 25 Aug 2009
    Locks up my MacPro with an ATI Radeon HD 4870 using 8GB of RAM and running SL 10.6. Passes the CPU test but locks up the entire system when testing the GPU. Can't force quit or anything. Must hold down power button to power off.
    [Version V015]

    1 Reply

    burypromote

    +85
    mitch_de commented on 26 Aug 2009
    Freeze problem running that Bench on ATI is known and fixed (i hope) with V025
    I have no ATI GPU, so such "things" are hard to fix. NVIDIAs worked all with old versions.
    But i made an new version which also should run on ATI GPUs !!!
    Try V025 and report result.
    time results will change (less time), read the readme.
    burypromote

    +48
    mac adam had trouble on 25 Aug 2009
    I run on 10.6 10A432
    The CPU test pass nice
    The Ati card do not give result after more 5min…
    mouse move, but cannot move windows, cannot kill, no dock, all os freeze without mouse…
    i was must force to reboot
    [Version V015]


    burypromote

    +48
    mac adam had trouble on 25 Aug 2009
    This app freeze my macPro with ATI HD4870 :(
    [Version V015]

    1 Reply

    burypromote

    +85
    mitch_de replied on 25 Aug 2009
    Do you use Snow Leopard ?
    What excat do you mean with freeze ?
    - tool crash , - no response of terminal/tool, - OS X no response / freeze
    If you use 10.6 , you can contact me my email ( iceguy at freenet.de), perhaps
    we find an solution. Until now you are first user which uses ATI - all others used Nvidia (MacBook, MacbookPro, Imac GT120,...)
    Thanks for submitting the prob.
    There are currently no ratings. Write a comment or review now.

    Downloads:50,383
    Version Downloads:35,876
    Type:Utilities : System
    License:Free
    Date:20 Sep 2009
    Platform:Intel 32 / OS X
    Price:Free0.00
    Overall (Version V 2.x):
    Features:
    Ease of Use:
    Value:
    Stability:
    Displaying 1-10 of 31
    1 2 3 4 >
    Displaying 1-6 of 6
    -
    -
    -
    Please login or create a new
    MacUpdate Member account
    to use this feature
    Watch Lists are available to
    MacUpdate Desktop Members
    Upgrade Now
    Install with MacUpdate Desktop.
    Save time moving files & cleaning
    up space wasting archives.
    OpenCL Benchmark, for Snow Leopard only
  • NVIDIA: 8xxx(M), 9xxx(M), GT1xx, GT2xx, GTX2xx
  • ATI: 4850, 4870 - maybe 10.6.2 update needed to run withput failures OPENCL - background information :
    • OpenCL is an API for universal GPU(CPU) computing , Mac OS X 10.6 supports that
    • Main difference to CUDA / ATI STEAM is: both APIs ate vendor specific. A CUDA (NV) app like badaboom(h264 on GPU) can't work on an ATI gpu and vice versa
    • OpenCL is universal means:
    • Xcode compiles an code which includes the source (C like) for the gpu That source is , different to CUDA/ATI STEAM, compiled later by OpenCL at runtime ! So same App can run on complete different gpus and also , without/less code change on CPU if no OPENCL gpu (newer ones) is available. GPU advantadge may vary - from 2 up to 40 faster, depends GPU type and CPU´s type OpenCL speedup could be even faster, if the memory transfer/access between GPU and CPU would be faster. PCIe Memory Transferlimit outperforms such GPUs like GTX285. I believe in 2011+ some new PCIe Versions will be developed for higher bandwidth over PCIe


  • - -