Own an ATI X1, HD or NV 7/8 series GPU and have 5-10 minutes? Please run this.

Discussion in 'GPGPU Technology & Programming' started by PeterT, May 29, 2007.

  1. PeterT

    Regular

    Joined:
    May 14, 2002
    Messages:
    702
    Likes Received:
    14
    Location:
    Austria
    Here's a collection of most V-cycle results in this thread. Obviously wrong ones or ones that were very similar to any already present were excluded. (Next time I'll write a program for this, I just spent 2 hours copying & pasting numbers :/)

    [​IMG]

    I don't want to make too many comments, just note that:
    - Results below 255^2 are not likely to be GPU limited
    - Apperantly NV's 7xxx Vista drivers are better than the 8xxx ones
    - Most of those results probably use 2D clocks on cards that have such a thing
    - In Vista, 2900XT and 8800GTX results are close, in XP the latter is 6 times faster
    - 88 series cards are the only ones that can complete the 2047^2 test correctly
     
  2. nyt

    nyt
    Newcomer

    Joined:
    May 14, 2003
    Messages:
    80
    Likes Received:
    0
    Location:
    Mtl
    Hi,
    I can't get anything but a BSoD (infinite loop) on my X1900XTX (AcceleroX2 fan), whatever the frequency (even 2D). It happens mostly at the end (size 2047). I have a taskmanager running on the side and can see the refresh stalling by periods of 5 to 10s at times or other times slowing down a lot. Eventually it freezes and 10s later, BSoD. This is on XP32/CAT7.5. What other test do you recommend to see if hardware is really bad? I have very few crashes in game even OC'd.
     
  3. PeterT

    Regular

    Joined:
    May 14, 2002
    Messages:
    702
    Likes Received:
    14
    Location:
    Austria
    If you don't get any GPU-related crashes in games I wouldn't worry. As discussed earlier in this thread, the crashes in this application may also be unrelated to stability. Especially if yours are independent of clock speed.
     
  4. BRiT

    BRiT (╯°□°)╯
    Moderator Legend Alpha

    Joined:
    Feb 7, 2002
    Messages:
    12,824
    Likes Received:
    9,177
    Location:
    Cleveland
    Here's a run but this time on a X3220 Core 2 Quad @ 2.7Ghz, 4Gig DDR2 900Mhz 5-4-4-12, still running the Nvidia 158.45 drivers for the MSI 8800 GTX HDOC on a fresh Vista Ultimate x64 install.

    Still plenty of odd numbers for the first four timings, though VCycle seems +/- a few percentage off the previous numbers.

    Code:
    GL multigrid benchmark 0.5 by Peter Thoman 2007
    
    DevIL initialized successfully.
    OpenGL initialized successfully.
    ILUT OpenGL mode set successfully.
    GLEW version 1.3.5 initialized successfully.
    Loaded required OpenGL extensions for GLPixelShader.
    Loaded required OpenGL extensions for GLRenderTexture.
    Loaded required OpenGL extensions for GLFilterStep.
    Initialization complete.
    
    
    
    ------------------
    -- size: 3
    
    BENCHMARK
    Jacobi iteration: 3.125 micros
    Residual calculation: 1.5625 micros
    Restriction: fw: 3.125 micros
    Interpolation + add: 1.5625 micros
    VCycle: 6453.13 micros
    
    
    BENCHMARK
    Jacobi iteration: 3.125 micros
    Residual calculation: 1.5625 micros
    Restriction: fw: 3.125 micros
    Interpolation + add: 1.5625 micros
    VCycle: 6296.88 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.5625 micros
    Residual calculation: 3.125 micros
    Restriction: fw: 3.125 micros
    Interpolation + add: 1.5625 micros
    VCycle: 6375 micros
    
    
    
    ------------------
    -- size: 7
    
    BENCHMARK
    Jacobi iteration: 1.73611 micros
    Residual calculation: 1.73611 micros
    Restriction: fw: 3.47222 micros
    Interpolation + add: 1.73611 micros
    VCycle: 10677.1 micros
    
    
    BENCHMARK
    Jacobi iteration: 3.47222 micros
    Residual calculation: 1.73611 micros
    Restriction: fw: 3.47222 micros
    Interpolation + add: 1.73611 micros
    VCycle: 10642.4 micros
    
    
    BENCHMARK
    Jacobi iteration: 3.47222 micros
    Residual calculation: 1.73611 micros
    Restriction: fw: 3.47222 micros
    Interpolation + add: 1.73611 micros
    VCycle: 11684 micros
    
    
    
    ------------------
    -- size: 15
    
    BENCHMARK
    Jacobi iteration: 1.95313 micros
    Residual calculation: 1.95313 micros
    Restriction: fw: 1.95313 micros
    Interpolation + add: 3.90625 micros
    VCycle: 14824.2 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.95313 micros
    Residual calculation: 1.95313 micros
    Restriction: fw: 3.90625 micros
    Interpolation + add: 1.95313 micros
    VCycle: 14941.4 micros
    
    
    BENCHMARK
    Jacobi iteration: 3.90625 micros
    Residual calculation: 1.95313 micros
    Restriction: fw: 1.95313 micros
    Interpolation + add: 3.90625 micros
    VCycle: 14785.2 micros
    
    
    
    ------------------
    -- size: 31
    
    BENCHMARK
    Jacobi iteration: 2.23214 micros
    Residual calculation: 2.23214 micros
    Restriction: fw: 4.46429 micros
    Interpolation + add: 2.23214 micros
    VCycle: 19129.5 micros
    
    
    BENCHMARK
    Jacobi iteration: 2.23214 micros
    Residual calculation: 2.23214 micros
    Restriction: fw: 4.46429 micros
    Interpolation + add: 2.23214 micros
    VCycle: 19062.5 micros
    
    
    BENCHMARK
    Jacobi iteration: 2.23214 micros
    Residual calculation: 2.23214 micros
    Restriction: fw: 4.46429 micros
    Interpolation + add: 2.23214 micros
    VCycle: 18995.5 micros
    
    
    
    ------------------
    -- size: 63
    
    BENCHMARK
    Jacobi iteration: 2.60417 micros
    Residual calculation: 2.60417 micros
    Restriction: fw: 2.60417 micros
    Interpolation + add: 2.60417 micros
    VCycle: 23489.6 micros
    
    
    BENCHMARK
    Jacobi iteration: 2.60417 micros
    Residual calculation: 2.60417 micros
    Restriction: fw: 2.60417 micros
    Interpolation + add: 2.60417 micros
    VCycle: 23515.6 micros
    
    
    BENCHMARK
    Jacobi iteration: 2.60417 micros
    Residual calculation: 2.60417 micros
    Restriction: fw: 2.60417 micros
    Interpolation + add: 2.60417 micros
    VCycle: 23333.3 micros
    
    
    
    ------------------
    -- size: 127
    
    BENCHMARK
    Jacobi iteration: 3.125 micros
    Residual calculation: 3.125 micros
    Restriction: fw: 9.375 micros
    Interpolation + add: 12.5 micros
    VCycle: 29031.3 micros
    
    
    BENCHMARK
    Jacobi iteration: 3.125 micros
    Residual calculation: 3.125 micros
    Restriction: fw: 34.375 micros
    Interpolation + add: 3.125 micros
    VCycle: 28218.7 micros
    
    
    BENCHMARK
    Jacobi iteration: 3.125 micros
    Residual calculation: 3.125 micros
    Restriction: fw: 9.375 micros
    Interpolation + add: 9.375 micros
    VCycle: 27750 micros
    
    
    
    ------------------
    -- size: 255
    
    BENCHMARK
    Jacobi iteration: 0 micros
    Residual calculation: 23.4375 micros
    Restriction: fw: 42.9688 micros
    Interpolation + add: 42.9688 micros
    VCycle: 33476.6 micros
    
    
    BENCHMARK
    Jacobi iteration: 3.90625 micros
    Residual calculation: 19.5313 micros
    Restriction: fw: 46.875 micros
    Interpolation + add: 42.9688 micros
    VCycle: 33554.7 micros
    
    
    BENCHMARK
    Jacobi iteration: 0 micros
    Residual calculation: 23.4375 micros
    Restriction: fw: 42.9688 micros
    Interpolation + add: 42.9688 micros
    VCycle: 33593.8 micros
    
    
    
    ------------------
    -- size: 511
    
    BENCHMARK
    Jacobi iteration: 5.20833 micros
    Residual calculation: 0 micros
    Restriction: fw: 145.833 micros
    Interpolation + add: 197.917 micros
    VCycle: 42239.6 micros
    
    
    BENCHMARK
    Jacobi iteration: 5.20833 micros
    Residual calculation: 0 micros
    Restriction: fw: 145.833 micros
    Interpolation + add: 197.917 micros
    VCycle: 42187.5 micros
    
    
    BENCHMARK
    Jacobi iteration: 5.20833 micros
    Residual calculation: 0 micros
    Restriction: fw: 145.833 micros
    Interpolation + add: 197.917 micros
    VCycle: 42187.5 micros
    
    
    
    ------------------
    -- size: 1023
    
    BENCHMARK
    Jacobi iteration: 0 micros
    Residual calculation: 7.8125 micros
    Restriction: fw: 0 micros
    Interpolation + add: 585.938 micros
    VCycle: 79140.6 micros
    
    
    BENCHMARK
    Jacobi iteration: 7.8125 micros
    Residual calculation: 0 micros
    Restriction: fw: 0 micros
    Interpolation + add: 593.75 micros
    VCycle: 79296.9 micros
    
    
    BENCHMARK
    Jacobi iteration: 0 micros
    Residual calculation: 0 micros
    Restriction: fw: 7.8125 micros
    Interpolation + add: 585.938 micros
    VCycle: 79296.9 micros
    
    
    
    ------------------
    -- size: 2047
    
    BENCHMARK
    Jacobi iteration: 0 micros
    Residual calculation: 0 micros
    Restriction: fw: 0 micros
    Interpolation + add: 15.625 micros
    VCycle: 234375 micros
    
    
    BENCHMARK
    Jacobi iteration: 0 micros
    Residual calculation: 0 micros
    Restriction: fw: 0 micros
    Interpolation + add: 15.625 micros
    VCycle: 233906 micros
    
    
    BENCHMARK
    Jacobi iteration: 0 micros
    Residual calculation: 0 micros
    Restriction: fw: 0 micros
    Interpolation + add: 0 micros
    VCycle: 234063 micros
    
     
  5. Anarchist4000

    Veteran Regular

    Joined:
    May 8, 2004
    Messages:
    1,439
    Likes Received:
    359
    Ok did a run with 3D clocks and Aero disabled.

    1900XT512MB
    Vista32 w/ Cat 7.5
    3D Clocks/No Aero
    All driver settings disabled

    Code:
    GL multigrid benchmark 0.5 by Peter Thoman 2007
    
    DevIL initialized successfully.
    OpenGL initialized successfully.
    ILUT OpenGL mode set successfully.
    GLEW version 1.3.5 initialized successfully.
    Loaded required OpenGL extensions for GLPixelShader.
    Loaded required OpenGL extensions for GLRenderTexture.
    Loaded required OpenGL extensions for GLFilterStep.
    Initialization complete.
    
    
    
    ------------------
    -- size: 3
    
    BENCHMARK
    Jacobi iteration: 15.0381 micros
    Residual calculation: 15.1358 micros
    Restriction: fw: 10.8391 micros
    Interpolation + add: 15.1358 micros
    VCycle: 334.939 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.0381 micros
    Residual calculation: 14.9405 micros
    Restriction: fw: 10.7415 micros
    Interpolation + add: 15.1357 micros
    VCycle: 332.987 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.2334 micros
    Residual calculation: 15.1358 micros
    Restriction: fw: 10.9368 micros
    Interpolation + add: 15.0381 micros
    VCycle: 333.963 micros
    
    
    
    ------------------
    -- size: 7
    
    BENCHMARK
    Jacobi iteration: 15.5154 micros
    Residual calculation: 15.5156 micros
    Restriction: fw: 11.1754 micros
    Interpolation + add: 15.841 micros
    VCycle: 550.096 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.407 micros
    Residual calculation: 15.2986 micros
    Restriction: fw: 10.9584 micros
    Interpolation + add: 15.19 micros
    VCycle: 546.84 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.19 micros
    Residual calculation: 15.407 micros
    Restriction: fw: 11.067 micros
    Interpolation + add: 15.0814 micros
    VCycle: 550.096 micros
    
    
    
    ------------------
    -- size: 15
    
    BENCHMARK
    Jacobi iteration: 15.1358 micros
    Residual calculation: 15.3799 micros
    Restriction: fw: 11.1077 micros
    Interpolation + add: 15.3799 micros
    VCycle: 762.89 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.502 micros
    Residual calculation: 15.624 micros
    Restriction: fw: 10.9856 micros
    Interpolation + add: 15.3799 micros
    VCycle: 766.553 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.2578 micros
    Residual calculation: 15.2579 micros
    Restriction: fw: 11.1076 micros
    Interpolation + add: 15.7461 micros
    VCycle: 762.89 micros
    
    
    
    ------------------
    -- size: 31
    
    BENCHMARK
    Jacobi iteration: 14.9266 micros
    Residual calculation: 15.2054 micros
    Restriction: fw: 11.0206 micros
    Interpolation + add: 15.066 micros
    VCycle: 969.524 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.066 micros
    Residual calculation: 15.2056 micros
    Restriction: fw: 11.16 micros
    Interpolation + add: 14.9264 micros
    VCycle: 963.946 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.2054 micros
    Residual calculation: 14.9266 micros
    Restriction: fw: 11.0204 micros
    Interpolation + add: 15.066 micros
    VCycle: 970.92 micros
    
    
    
    ------------------
    -- size: 63
    
    BENCHMARK
    Jacobi iteration: 15.4613 micros
    Residual calculation: 15.2985 micros
    Restriction: fw: 11.067 micros
    Interpolation + add: 15.4612 micros
    VCycle: 1179.94 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.2985 micros
    Residual calculation: 15.2985 micros
    Restriction: fw: 10.9042 micros
    Interpolation + add: 15.2985 micros
    VCycle: 1179.94 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.1357 micros
    Residual calculation: 15.2985 micros
    Restriction: fw: 11.3925 micros
    Interpolation + add: 15.4613 micros
    VCycle: 1179.94 micros
    
    
    
    ------------------
    -- size: 127
    
    BENCHMARK
    Jacobi iteration: 15.038 micros
    Residual calculation: 15.2334 micros
    Restriction: fw: 11.5228 micros
    Interpolation + add: 15.4286 micros
    VCycle: 2191.27 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.2334 micros
    Residual calculation: 15.2334 micros
    Restriction: fw: 11.132 micros
    Interpolation + add: 15.2334 micros
    VCycle: 2206.89 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.038 micros
    Residual calculation: 15.2334 micros
    Restriction: fw: 10.9368 micros
    Interpolation + add: 15.2334 micros
    VCycle: 2206.89 micros
    
    
    
    ------------------
    -- size: 255
    
    BENCHMARK
    Jacobi iteration: 15.3798 micros
    Residual calculation: 15.1358 micros
    Restriction: fw: 10.9857 micros
    Interpolation + add: 15.3798 micros
    VCycle: 8825.12 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.3798 micros
    Residual calculation: 15.1358 micros
    Restriction: fw: 10.9857 micros
    Interpolation + add: 15.1358 micros
    VCycle: 9616.08 micros
    
    
    BENCHMARK
    Jacobi iteration: 15.1358 micros
    Residual calculation: 15.1358 micros
    Restriction: fw: 10.9855 micros
    Interpolation + add: 15.1358 micros
    VCycle: 9613.64 micros
    
    
    
    ------------------
    -- size: 511
    
    BENCHMARK
    Jacobi iteration: 15.2987 micros
    Residual calculation: 14.973 micros
    Restriction: fw: 11.067 micros
    Interpolation + add: 14.973 micros
    VCycle: 36488.6 micros
    
    
    BENCHMARK
    Jacobi iteration: 24.738 micros
    Residual calculation: 25.0637 micros
    Restriction: fw: 25.0633 micros
    Interpolation + add: 25.389 micros
    VCycle: 38773.6 micros
    
    
    BENCHMARK
    Jacobi iteration: 25.0637 micros
    Residual calculation: 24.738 micros
    Restriction: fw: 25.389 micros
    Interpolation + add: 25.0633 micros
    VCycle: 38776.8 micros
    
    
    
    ------------------
    -- size: 1023
    
    BENCHMARK
    Jacobi iteration: 15.136 micros
    Residual calculation: 15.1355 micros
    Restriction: fw: 11.23 micros
    Interpolation + add: 15.1355 micros
    VCycle: 144019 micros
    
    
    BENCHMARK
    Jacobi iteration: 83.491 micros
    Residual calculation: 92.7675 micros
    Restriction: fw: 83.4905 micros
    Interpolation + add: 83.491 micros
    VCycle: 154424 micros
    
    
    BENCHMARK
    Jacobi iteration: 83.0025 micros
    Residual calculation: 83.0025 micros
    Restriction: fw: 93.744 micros
    Interpolation + add: 83.0025 micros
    VCycle: 154429 micros
    
    
    
    ------------------
    -- size: 2047
    
    BENCHMARK
    Jacobi iteration: 15.624 micros
    Residual calculation: 14.647 micros
    Restriction: fw: 11.718 micros
    Interpolation + add: 14.648 micros
    VCycle: 6737.85 micros
    
    
    BENCHMARK
    Jacobi iteration: 8309.04 micros
    Residual calculation: 1270.43 micros
    Restriction: fw: 341.775 micros
    Interpolation + add: 298.809 micros
    VCycle: 91917.9 micros
    
    
    BENCHMARK
    Jacobi iteration: 8235.8 micros
    Residual calculation: 1793.83 micros
    Restriction: fw: 360.329 micros
    Interpolation + add: 280.255 micros
    VCycle: 93441.3 micros
    
    
    The residual calculation looks closer to being correct but the others still seem off. Times improved a bit with the 3D clocks and Aero disabled but that doesn't seem to be causing the problem.

    Any chance of converting things to DirectX and seeing if that makes a difference?
     
  6. PeterT

    Regular

    Joined:
    May 14, 2002
    Messages:
    702
    Likes Received:
    14
    Location:
    Austria
    Your numbers are by far the strangest I've yet encountered (these and the earlier ones). The component times are obviously completely unusable, and the V-cycle times are very bad for a 8800 GTX. In fact, you'd probably be better off running the multigrid method with an SSE-optimized solver on your monstrous CPU :p

    Then again your numbers are also the only 8800 Vista results reported, so maybe the NV Vista 8-series drivers just really suck for what I'm doing...

    Not at this point, sorry, my thesis should have been done "end of May" ... by the way, I have to get back to writing :D
     
  7. Davros

    Legend

    Joined:
    Jun 7, 2004
    Messages:
    15,050
    Likes Received:
    2,386
    Ok maybe not to someone like yourself who understands the hardware, but to an adverage joe (i.e: me)
    it appears that the 8800 Really is 6 times faster than the x2900 and the reason they are close in vista is because nvidia's vista drivers are truely awful

    but then again im just a gamer ;)
     
  8. BRiT

    BRiT (╯°□°)╯
    Moderator Legend Alpha

    Joined:
    Feb 7, 2002
    Messages:
    12,824
    Likes Received:
    9,177
    Location:
    Cleveland
    Rys also posted some 8800 GTX (or) GTS Vista 64 numbers also using 158.42. His Vcycle times come in mostly around what mine are, with his being a bit higher in the larger sets. I would hope, for my sake, his numbers were for the GTX in his system and not his GTS. Is there any way of telling which OpenGL adapter the application bound itself to? I suspect it's just a matter of Vista drivers not being as optimized for this code path as the XP ones are.
     
  9. Albuquerque

    Albuquerque Red-headed step child
    Veteran

    Joined:
    Jun 17, 2004
    Messages:
    3,845
    Likes Received:
    329
    Location:
    35.1415,-90.056
    Machine:

    Dell Inspiron e1505
    Intel 945PM chipset
    T2300 (Core Duo) 1.66Ghz
    2Gb 667Mhz DDR2 5-5-5-15
    7300Go 128mb PCI-E
    Vista Home Premium x32 w/Dell-provided drivers

    On "battery mode":
    Code:
    GL multigrid benchmark 0.5 by Peter Thoman 2007
    
    DevIL initialized successfully.
    OpenGL initialized successfully.
    ILUT OpenGL mode set successfully.
    GLEW version 1.3.5 initialized successfully.
    Loaded required OpenGL extensions for GLPixelShader.
    Loaded required OpenGL extensions for GLRenderTexture.
    Loaded required OpenGL extensions for GLFilterStep.
    Initialization complete.
    
    
    
    ------------------
    -- size: 3
    
    BENCHMARK
    Jacobi iteration: 277.68 micros
    Residual calculation: 282.36 micros
    Restriction: fw: 254.28 micros
    Interpolation + add: 287.04 micros
    VCycle: 18922.8 micros
    
    
    BENCHMARK
    Jacobi iteration: 290.16 micros
    Residual calculation: 255.84 micros
    Restriction: fw: 274.56 micros
    Interpolation + add: 257.4 micros
    VCycle: 18876 micros
    
    
    BENCHMARK
    Jacobi iteration: 255.84 micros
    Residual calculation: 257.4 micros
    Restriction: fw: 248.04 micros
    Interpolation + add: 254.28 micros
    VCycle: 18876 micros
    
    
    
    ------------------
    -- size: 7
    
    BENCHMARK
    Jacobi iteration: 256.533 micros
    Residual calculation: 251.333 micros
    Restriction: fw: 280.8 micros
    Interpolation + add: 266.933 micros
    VCycle: 30957.3 micros
    
    
    BENCHMARK
    Jacobi iteration: 280.8 micros
    Residual calculation: 287.733 micros
    Restriction: fw: 280.8 micros
    Interpolation + add: 284.267 micros
    VCycle: 30870.7 micros
    
    
    BENCHMARK
    Jacobi iteration: 282.533 micros
    Residual calculation: 284.267 micros
    Restriction: fw: 249.6 micros
    Interpolation + add: 256.533 micros
    VCycle: 30922.7 micros
    
    
    
    ------------------
    -- size: 15
    
    BENCHMARK
    Jacobi iteration: 276.9 micros
    Residual calculation: 278.85 micros
    Restriction: fw: 284.7 micros
    Interpolation + add: 267.15 micros
    VCycle: 42939 micros
    
    
    BENCHMARK
    Jacobi iteration: 278.85 micros
    Residual calculation: 284.7 micros
    Restriction: fw: 274.95 micros
    Interpolation + add: 259.35 micros
    VCycle: 42997.5 micros
    
    
    BENCHMARK
    Jacobi iteration: 269.1 micros
    Residual calculation: 282.75 micros
    Restriction: fw: 282.75 micros
    Interpolation + add: 286.65 micros
    VCycle: 42939 micros
    
    
    
    ------------------
    -- size: 31
    
    BENCHMARK
    Jacobi iteration: 274.114 micros
    Residual calculation: 271.886 micros
    Restriction: fw: 265.2 micros
    Interpolation + add: 258.514 micros
    VCycle: 54756 micros
    
    
    BENCHMARK
    Jacobi iteration: 287.486 micros
    Residual calculation: 276.343 micros
    Restriction: fw: 260.743 micros
    Interpolation + add: 254.057 micros
    VCycle: 54466.3 micros
    
    
    BENCHMARK
    Jacobi iteration: 285.257 micros
    Residual calculation: 276.343 micros
    Restriction: fw: 267.429 micros
    Interpolation + add: 287.486 micros
    VCycle: 54510.9 micros
    
    
    
    ------------------
    -- size: 63
    
    BENCHMARK
    Jacobi iteration: 280.8 micros
    Residual calculation: 291.2 micros
    Restriction: fw: 278.2 micros
    Interpolation + add: 260 micros
    VCycle: 65156 micros
    
    
    BENCHMARK
    Jacobi iteration: 286 micros
    Residual calculation: 280.8 micros
    Restriction: fw: 296.4 micros
    Interpolation + add: 273 micros
    VCycle: 65130 micros
    
    
    BENCHMARK
    Jacobi iteration: 275.6 micros
    Residual calculation: 299 micros
    Restriction: fw: 280.8 micros
    Interpolation + add: 265.2 micros
    VCycle: 65338 micros
    
    
    
    ------------------
    -- size: 127
    
    BENCHMARK
    Jacobi iteration: 636.48 micros
    Residual calculation: 733.2 micros
    Restriction: fw: 1070.16 micros
    Interpolation + add: 393.12 micros
    VCycle: 76440 micros
    
    
    BENCHMARK
    Jacobi iteration: 645.84 micros
    Residual calculation: 726.96 micros
    Restriction: fw: 1067.04 micros
    Interpolation + add: 396.24 micros
    VCycle: 76408.8 micros
    
    
    BENCHMARK
    Jacobi iteration: 639.6 micros
    Residual calculation: 726.96 micros
    Restriction: fw: 1063.92 micros
    Interpolation + add: 390 micros
    VCycle: 76252.8 micros
    
    
    
    ------------------
    -- size: 255
    Error using previous program Object.
    
    On "AC mode"
    Code:
    GL multigrid benchmark 0.5 by Peter Thoman 2007
    
    DevIL initialized successfully.
    OpenGL initialized successfully.
    ILUT OpenGL mode set successfully.
    GLEW version 1.3.5 initialized successfully.
    Loaded required OpenGL extensions for GLPixelShader.
    Loaded required OpenGL extensions for GLRenderTexture.
    Loaded required OpenGL extensions for GLFilterStep.
    Initialization complete.
    
    
    
    ------------------
    -- size: 3
    
    BENCHMARK
    Jacobi iteration: 68.64 micros
    Residual calculation: 68.64 micros
    Restriction: fw: 68.64 micros
    Interpolation + add: 68.64 micros
    VCycle: 4446 micros
    
    
    BENCHMARK
    Jacobi iteration: 68.64 micros
    Residual calculation: 68.64 micros
    Restriction: fw: 68.64 micros
    Interpolation + add: 68.64 micros
    VCycle: 4446 micros
    
    
    BENCHMARK
    Jacobi iteration: 68.64 micros
    Residual calculation: 68.64 micros
    Restriction: fw: 40.56 micros
    Interpolation + add: 59.28 micros
    VCycle: 4274.4 micros
    
    
    
    ------------------
    -- size: 7
    
    BENCHMARK
    Jacobi iteration: 69.3333 micros
    Residual calculation: 67.6 micros
    Restriction: fw: 71.0667 micros
    Interpolation + add: 67.6 micros
    VCycle: 7366.67 micros
    
    
    BENCHMARK
    Jacobi iteration: 69.3333 micros
    Residual calculation: 69.3333 micros
    Restriction: fw: 69.3333 micros
    Interpolation + add: 69.3333 micros
    VCycle: 7366.67 micros
    
    
    BENCHMARK
    Jacobi iteration: 69.3333 micros
    Residual calculation: 67.6 micros
    Restriction: fw: 69.3333 micros
    Interpolation + add: 69.3333 micros
    VCycle: 7384 micros
    
    
    
    ------------------
    -- size: 15
    
    BENCHMARK
    Jacobi iteration: 72.15 micros
    Residual calculation: 72.15 micros
    Restriction: fw: 70.2 micros
    Interpolation + add: 70.2 micros
    VCycle: 10354.5 micros
    
    
    BENCHMARK
    Jacobi iteration: 72.15 micros
    Residual calculation: 70.2 micros
    Restriction: fw: 72.15 micros
    Interpolation + add: 68.25 micros
    VCycle: 10354.5 micros
    
    
    BENCHMARK
    Jacobi iteration: 72.15 micros
    Residual calculation: 70.2 micros
    Restriction: fw: 72.15 micros
    Interpolation + add: 68.25 micros
    VCycle: 10354.5 micros
    
    
    
    ------------------
    -- size: 31
    
    BENCHMARK
    Jacobi iteration: 140.4 micros
    Residual calculation: 138.171 micros
    Restriction: fw: 100.286 micros
    Interpolation + add: 80.2286 micros
    VCycle: 13304.6 micros
    
    
    BENCHMARK
    Jacobi iteration: 138.171 micros
    Residual calculation: 140.4 micros
    Restriction: fw: 102.514 micros
    Interpolation + add: 84.6857 micros
    VCycle: 13326.9 micros
    
    
    BENCHMARK
    Jacobi iteration: 140.4 micros
    Residual calculation: 142.629 micros
    Restriction: fw: 100.286 micros
    Interpolation + add: 82.4571 micros
    VCycle: 13282.3 micros
    
    
    
    ------------------
    -- size: 63
    
    BENCHMARK
    Jacobi iteration: 512.2 micros
    Residual calculation: 543.4 micros
    Restriction: fw: 374.4 micros
    Interpolation + add: 301.6 micros
    VCycle: 17550 micros
    
    
    BENCHMARK
    Jacobi iteration: 512.2 micros
    Residual calculation: 540.8 micros
    Restriction: fw: 377 micros
    Interpolation + add: 301.6 micros
    VCycle: 17472 micros
    
    
    BENCHMARK
    Jacobi iteration: 512.2 micros
    Residual calculation: 540.8 micros
    Restriction: fw: 377 micros
    Interpolation + add: 301.6 micros
    VCycle: 17524 micros
    
    
    
    ------------------
    -- size: 127
    GLFilterStep -- Error doing makeCurrent on target.
    
     
  10. Unknown Soldier

    Veteran

    Joined:
    Jul 28, 2002
    Messages:
    2,238
    Likes Received:
    33
    Hi Peter,

    I ran this on my PC and also found that it used my CPU, funnily enough, the 2047 test utilizes all CPU's to a degree. At the beginning, at least one CPU runs at 30%(memory at 277Mb). This is pretty constant until 1023 where the CPU or two hits around 34% with memory 330Mb+-

    The last test - 2047 - utilises all CPU's to a degree with CPU usage maxxing at 34%. Memory climbed to 720Mb+

    I ran a few tests after each other.

    WinXP QX6600, 2Gig 800 DDR2 memory, Nvidia GTS - Default speeds all - and driver 158.22

    I didn't use test 3 because I was trying to capture screenshots but that didn't work. CPU pics at the end were captured at the end of the test.

    Last, my GTS 320Mb ran all tests .. including the 2047. ;)

    Test 1

    Code:
    GL multigrid benchmark 0.5 by Peter Thoman 2007
    
    DevIL initialized successfully.
    OpenGL initialized successfully.
    ILUT OpenGL mode set successfully.
    GLEW version 1.3.5 initialized successfully.
    Loaded required OpenGL extensions for GLPixelShader.
    Loaded required OpenGL extensions for GLRenderTexture.
    Loaded required OpenGL extensions for GLFilterStep.
    Initialization complete.
    
    
    
    ------------------
    -- size: 3
    
    BENCHMARK
    Jacobi iteration: 3.125 micros
    Residual calculation: 1.5625 micros
    Restriction: fw: 1.5625 micros
    Interpolation + add: 1.5625 micros
    VCycle: 500 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.5625 micros
    Residual calculation: 1.5625 micros
    Restriction: fw: 1.5625 micros
    Interpolation + add: 3.125 micros
    VCycle: 484.375 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.5625 micros
    Residual calculation: 1.5625 micros
    Restriction: fw: 3.125 micros
    Interpolation + add: 1.5625 micros
    VCycle: 484.375 micros
    
    
    
    ------------------
    -- size: 7
    
    BENCHMARK
    Jacobi iteration: 1.73611 micros
    Residual calculation: 1.73611 micros
    Restriction: fw: 1.73611 micros
    Interpolation + add: 3.47222 micros
    VCycle: 833.333 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.73611 micros
    Residual calculation: 1.73611 micros
    Restriction: fw: 3.47222 micros
    Interpolation + add: 1.73611 micros
    VCycle: 815.972 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.73611 micros
    Residual calculation: 1.73611 micros
    Restriction: fw: 3.47222 micros
    Interpolation + add: 1.73611 micros
    VCycle: 815.972 micros
    
    
    
    ------------------
    -- size: 15
    
    BENCHMARK
    Jacobi iteration: 1.95313 micros
    Residual calculation: 1.95313 micros
    Restriction: fw: 1.95313 micros
    Interpolation + add: 1.95313 micros
    VCycle: 1152.34 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.95313 micros
    Residual calculation: 1.95313 micros
    Restriction: fw: 1.95313 micros
    Interpolation + add: 1.95313 micros
    VCycle: 1152.34 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.95313 micros
    Residual calculation: 1.95313 micros
    Restriction: fw: 1.95313 micros
    Interpolation + add: 1.95313 micros
    VCycle: 1152.34 micros
    
    
    
    ------------------
    -- size: 31
    
    BENCHMARK
    Jacobi iteration: 2.23214 micros
    Residual calculation: 2.23214 micros
    Restriction: fw: 2.23214 micros
    Interpolation + add: 2.23214 micros
    VCycle: 1517.86 micros
    
    
    BENCHMARK
    Jacobi iteration: 2.23214 micros
    Residual calculation: 2.23214 micros
    Restriction: fw: 2.23214 micros
    Interpolation + add: 2.23214 micros
    VCycle: 1473.21 micros
    
    
    BENCHMARK
    Jacobi iteration: 0 micros
    Residual calculation: 2.23214 micros
    Restriction: fw: 2.23214 micros
    Interpolation + add: 2.23214 micros
    VCycle: 1473.21 micros
    
    
    
    ------------------
    -- size: 63
    
    BENCHMARK
    Jacobi iteration: 5.20833 micros
    Residual calculation: 5.20833 micros
    Restriction: fw: 2.60417 micros
    Interpolation + add: 2.60417 micros
    VCycle: 1822.92 micros
    
    
    BENCHMARK
    Jacobi iteration: 2.60417 micros
    Residual calculation: 5.20833 micros
    Restriction: fw: 2.60417 micros
    Interpolation + add: 5.20833 micros
    VCycle: 1822.92 micros
    
    
    BENCHMARK
    Jacobi iteration: 5.20833 micros
    Residual calculation: 5.20833 micros
    Restriction: fw: 2.60417 micros
    Interpolation + add: 2.60417 micros
    VCycle: 1875 micros
    
    
    
    ------------------
    -- size: 127
    
    BENCHMARK
    Jacobi iteration: 21.875 micros
    Residual calculation: 18.75 micros
    Restriction: fw: 15.625 micros
    Interpolation + add: 12.5 micros
    VCycle: 2218.75 micros
    
    
    BENCHMARK
    Jacobi iteration: 18.75 micros
    Residual calculation: 18.75 micros
    Restriction: fw: 15.625 micros
    Interpolation + add: 12.5 micros
    VCycle: 2218.75 micros
    
    
    BENCHMARK
    Jacobi iteration: 18.75 micros
    Residual calculation: 18.75 micros
    Restriction: fw: 15.625 micros
    Interpolation + add: 12.5 micros
    VCycle: 2218.75 micros
    
    
    
    ------------------
    -- size: 255
    
    BENCHMARK
    Jacobi iteration: 82.0313 micros
    Residual calculation: 78.125 micros
    Restriction: fw: 74.2188 micros
    Interpolation + add: 50.7813 micros
    VCycle: 2851.56 micros
    
    
    BENCHMARK
    Jacobi iteration: 82.0313 micros
    Residual calculation: 78.125 micros
    Restriction: fw: 74.2188 micros
    Interpolation + add: 50.7813 micros
    VCycle: 2851.56 micros
    
    
    BENCHMARK
    Jacobi iteration: 82.0313 micros
    Residual calculation: 78.125 micros
    Restriction: fw: 74.2188 micros
    Interpolation + add: 50.7813 micros
    VCycle: 2851.56 micros
    
    
    
    ------------------
    -- size: 511
    
    BENCHMARK
    Jacobi iteration: 328.125 micros
    Residual calculation: 296.875 micros
    Restriction: fw: 312.5 micros
    Interpolation + add: 192.708 micros
    VCycle: 5729.17 micros
    
    
    BENCHMARK
    Jacobi iteration: 328.125 micros
    Residual calculation: 302.083 micros
    Restriction: fw: 317.708 micros
    Interpolation + add: 187.5 micros
    VCycle: 5729.17 micros
    
    
    BENCHMARK
    Jacobi iteration: 322.917 micros
    Residual calculation: 302.083 micros
    Restriction: fw: 317.708 micros
    Interpolation + add: 187.5 micros
    VCycle: 5729.17 micros
    
    
    
    ------------------
    -- size: 1023
    
    BENCHMARK
    Jacobi iteration: 1304.69 micros
    Residual calculation: 1210.94 micros
    Restriction: fw: 1312.5 micros
    Interpolation + add: 804.688 micros
    VCycle: 24375 micros
    
    
    BENCHMARK
    Jacobi iteration: 1304.69 micros
    Residual calculation: 1210.94 micros
    Restriction: fw: 1320.31 micros
    Interpolation + add: 812.5 micros
    VCycle: 24296.9 micros
    
    
    BENCHMARK
    Jacobi iteration: 1296.88 micros
    Residual calculation: 1218.75 micros
    Restriction: fw: 1328.13 micros
    Interpolation + add: 804.688 micros
    VCycle: 24375 micros
    
    
    
    ------------------
    -- size: 2047
    
    BENCHMARK
    Jacobi iteration: 687.5 micros
    Residual calculation: 4687.5 micros
    Restriction: fw: 6656.25 micros
    Interpolation + add: 4671.88 micros
    VCycle: 138750 micros
    
    
    BENCHMARK
    Jacobi iteration: 734.375 micros
    Residual calculation: 4687.5 micros
    Restriction: fw: 6640.63 micros
    Interpolation + add: 4687.5 micros
    VCycle: 138594 micros
    
    
    BENCHMARK
    Jacobi iteration: 718.75 micros
    Residual calculation: 4656.25 micros
    Restriction: fw: 6656.25 micros
    Interpolation + add: 4718.75 micros
    VCycle: 138438 micros
    
    
    Test 2

    Code:
    GL multigrid benchmark 0.5 by Peter Thoman 2007
    
    DevIL initialized successfully.
    OpenGL initialized successfully.
    ILUT OpenGL mode set successfully.
    GLEW version 1.3.5 initialized successfully.
    Loaded required OpenGL extensions for GLPixelShader.
    Loaded required OpenGL extensions for GLRenderTexture.
    Loaded required OpenGL extensions for GLFilterStep.
    Initialization complete.
    
    
    
    ------------------
    -- size: 3
    
    BENCHMARK
    Jacobi iteration: 1.5625 micros
    Residual calculation: 3.125 micros
    Restriction: fw: 1.5625 micros
    Interpolation + add: 1.5625 micros
    VCycle: 500 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.5625 micros
    Residual calculation: 1.5625 micros
    Restriction: fw: 1.5625 micros
    Interpolation + add: 3.125 micros
    VCycle: 484.375 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.5625 micros
    Residual calculation: 3.125 micros
    Restriction: fw: 1.5625 micros
    Interpolation + add: 1.5625 micros
    VCycle: 500 micros
    
    
    
    ------------------
    -- size: 7
    
    BENCHMARK
    Jacobi iteration: 1.73611 micros
    Residual calculation: 1.73611 micros
    Restriction: fw: 3.47222 micros
    Interpolation + add: 1.73611 micros
    VCycle: 815.972 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.73611 micros
    Residual calculation: 1.73611 micros
    Restriction: fw: 1.73611 micros
    Interpolation + add: 1.73611 micros
    VCycle: 833.333 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.73611 micros
    Residual calculation: 1.73611 micros
    Restriction: fw: 1.73611 micros
    Interpolation + add: 1.73611 micros
    VCycle: 850.694 micros
    
    
    
    ------------------
    -- size: 15
    
    BENCHMARK
    Jacobi iteration: 1.95313 micros
    Residual calculation: 1.95313 micros
    Restriction: fw: 1.95313 micros
    Interpolation + add: 1.95313 micros
    VCycle: 1171.88 micros
    
    
    BENCHMARK
    Jacobi iteration: 3.90625 micros
    Residual calculation: 1.95313 micros
    Restriction: fw: 1.95313 micros
    Interpolation + add: 1.95313 micros
    VCycle: 1152.34 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.95313 micros
    Residual calculation: 1.95313 micros
    Restriction: fw: 1.95313 micros
    Interpolation + add: 1.95313 micros
    VCycle: 1171.88 micros
    
    
    
    ------------------
    -- size: 31
    
    BENCHMARK
    Jacobi iteration: 2.23214 micros
    Residual calculation: 0 micros
    Restriction: fw: 2.23214 micros
    Interpolation + add: 2.23214 micros
    VCycle: 1495.54 micros
    
    
    BENCHMARK
    Jacobi iteration: 2.23214 micros
    Residual calculation: 2.23214 micros
    Restriction: fw: 0 micros
    Interpolation + add: 2.23214 micros
    VCycle: 1495.54 micros
    
    
    BENCHMARK
    Jacobi iteration: 2.23214 micros
    Residual calculation: 2.23214 micros
    Restriction: fw: 2.23214 micros
    Interpolation + add: 2.23214 micros
    VCycle: 1495.54 micros
    
    
    
    ------------------
    -- size: 63
    
    BENCHMARK
    Jacobi iteration: 5.20833 micros
    Residual calculation: 5.20833 micros
    Restriction: fw: 2.60417 micros
    Interpolation + add: 2.60417 micros
    VCycle: 1848.96 micros
    
    
    BENCHMARK
    Jacobi iteration: 5.20833 micros
    Residual calculation: 2.60417 micros
    Restriction: fw: 5.20833 micros
    Interpolation + add: 2.60417 micros
    VCycle: 1848.96 micros
    
    
    BENCHMARK
    Jacobi iteration: 5.20833 micros
    Residual calculation: 2.60417 micros
    Restriction: fw: 5.20833 micros
    Interpolation + add: 2.60417 micros
    VCycle: 1848.96 micros
    
    
    
    ------------------
    -- size: 127
    
    BENCHMARK
    Jacobi iteration: 18.75 micros
    Residual calculation: 18.75 micros
    Restriction: fw: 15.625 micros
    Interpolation + add: 12.5 micros
    VCycle: 2218.75 micros
    
    
    BENCHMARK
    Jacobi iteration: 18.75 micros
    Residual calculation: 18.75 micros
    Restriction: fw: 15.625 micros
    Interpolation + add: 12.5 micros
    VCycle: 2250 micros
    
    
    BENCHMARK
    Jacobi iteration: 18.75 micros
    Residual calculation: 18.75 micros
    Restriction: fw: 15.625 micros
    Interpolation + add: 12.5 micros
    VCycle: 2250 micros
    
    
    
    ------------------
    -- size: 255
    
    BENCHMARK
    Jacobi iteration: 82.0313 micros
    Residual calculation: 78.125 micros
    Restriction: fw: 70.3125 micros
    Interpolation + add: 50.7813 micros
    VCycle: 2929.69 micros
    
    
    BENCHMARK
    Jacobi iteration: 82.0313 micros
    Residual calculation: 78.125 micros
    Restriction: fw: 70.3125 micros
    Interpolation + add: 50.7813 micros
    VCycle: 2890.63 micros
    
    
    BENCHMARK
    Jacobi iteration: 82.0313 micros
    Residual calculation: 78.125 micros
    Restriction: fw: 74.2188 micros
    Interpolation + add: 46.875 micros
    VCycle: 2929.69 micros
    
    
    
    ------------------
    -- size: 511
    
    BENCHMARK
    Jacobi iteration: 328.125 micros
    Residual calculation: 296.875 micros
    Restriction: fw: 317.708 micros
    Interpolation + add: 187.5 micros
    VCycle: 5781.25 micros
    
    
    BENCHMARK
    Jacobi iteration: 322.917 micros
    Residual calculation: 302.083 micros
    Restriction: fw: 317.708 micros
    Interpolation + add: 187.5 micros
    VCycle: 5781.25 micros
    
    
    BENCHMARK
    Jacobi iteration: 322.917 micros
    Residual calculation: 302.083 micros
    Restriction: fw: 317.708 micros
    Interpolation + add: 187.5 micros
    VCycle: 5781.25 micros
    
    
    
    ------------------
    -- size: 1023
    
    BENCHMARK
    Jacobi iteration: 1312.5 micros
    Residual calculation: 1203.13 micros
    Restriction: fw: 1335.94 micros
    Interpolation + add: 828.125 micros
    VCycle: 24531.3 micros
    
    
    BENCHMARK
    Jacobi iteration: 1304.69 micros
    Residual calculation: 1187.5 micros
    Restriction: fw: 1343.75 micros
    Interpolation + add: 812.5 micros
    VCycle: 24609.4 micros
    
    
    BENCHMARK
    Jacobi iteration: 1312.5 micros
    Residual calculation: 1203.13 micros
    Restriction: fw: 1320.31 micros
    Interpolation + add: 835.938 micros
    VCycle: 24531.3 micros
    
    
    
    ------------------
    -- size: 2047
    
    BENCHMARK
    Jacobi iteration: 671.875 micros
    Residual calculation: 4453.13 micros
    Restriction: fw: 6359.38 micros
    Interpolation + add: 4609.38 micros
    VCycle: 142031 micros
    
    
    BENCHMARK
    Jacobi iteration: 718.75 micros
    Residual calculation: 4453.13 micros
    Restriction: fw: 6359.38 micros
    Interpolation + add: 4578.13 micros
    VCycle: 141875 micros
    
    
    BENCHMARK
    Jacobi iteration: 687.5 micros
    Residual calculation: 4531.25 micros
    Restriction: fw: 6328.13 micros
    Interpolation + add: 4593.75 micros
    VCycle: 141563 micros
    
    Test 4

    Code:
    GL multigrid benchmark 0.5 by Peter Thoman 2007
    
    DevIL initialized successfully.
    OpenGL initialized successfully.
    ILUT OpenGL mode set successfully.
    GLEW version 1.3.5 initialized successfully.
    Loaded required OpenGL extensions for GLPixelShader.
    Loaded required OpenGL extensions for GLRenderTexture.
    Loaded required OpenGL extensions for GLFilterStep.
    Initialization complete.
    
    
    
    ------------------
    -- size: 3
    
    BENCHMARK
    Jacobi iteration: 1.5625 micros
    Residual calculation: 3.125 micros
    Restriction: fw: 1.5625 micros
    Interpolation + add: 1.5625 micros
    VCycle: 484.375 micros
    
    
    BENCHMARK
    Jacobi iteration: 3.125 micros
    Residual calculation: 1.5625 micros
    Restriction: fw: 1.5625 micros
    Interpolation + add: 1.5625 micros
    VCycle: 500 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.5625 micros
    Residual calculation: 1.5625 micros
    Restriction: fw: 1.5625 micros
    Interpolation + add: 3.125 micros
    VCycle: 484.375 micros
    
    
    
    ------------------
    -- size: 7
    
    BENCHMARK
    Jacobi iteration: 1.73611 micros
    Residual calculation: 3.47222 micros
    Restriction: fw: 1.73611 micros
    Interpolation + add: 1.73611 micros
    VCycle: 833.333 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.73611 micros
    Residual calculation: 1.73611 micros
    Restriction: fw: 1.73611 micros
    Interpolation + add: 1.73611 micros
    VCycle: 815.972 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.73611 micros
    Residual calculation: 1.73611 micros
    Restriction: fw: 3.47222 micros
    Interpolation + add: 1.73611 micros
    VCycle: 815.972 micros
    
    
    
    ------------------
    -- size: 15
    
    BENCHMARK
    Jacobi iteration: 1.95313 micros
    Residual calculation: 1.95313 micros
    Restriction: fw: 1.95313 micros
    Interpolation + add: 1.95313 micros
    VCycle: 1152.34 micros
    
    
    BENCHMARK
    Jacobi iteration: 1.95313 micros
    Residual calculation: 1.95313 micros
    Restriction: fw: 1.95313 micros
    Interpolation + add: 1.95313 micros
    VCycle: 1152.34 micros
    
    
    BENCHMARK
    Jacobi iteration: 3.90625 micros
    Residual calculation: 1.95313 micros
    Restriction: fw: 1.95313 micros
    Interpolation + add: 1.95313 micros
    VCycle: 1132.81 micros
    
    
    
    ------------------
    -- size: 31
    
    BENCHMARK
    Jacobi iteration: 2.23214 micros
    Residual calculation: 2.23214 micros
    Restriction: fw: 2.23214 micros
    Interpolation + add: 2.23214 micros
    VCycle: 1473.21 micros
    
    
    BENCHMARK
    Jacobi iteration: 2.23214 micros
    Residual calculation: 2.23214 micros
    Restriction: fw: 2.23214 micros
    Interpolation + add: 2.23214 micros
    VCycle: 1473.21 micros
    
    
    BENCHMARK
    Jacobi iteration: 0 micros
    Residual calculation: 2.23214 micros
    Restriction: fw: 2.23214 micros
    Interpolation + add: 2.23214 micros
    VCycle: 1473.21 micros
    
    
    
    ------------------
    -- size: 63
    
    BENCHMARK
    Jacobi iteration: 5.20833 micros
    Residual calculation: 5.20833 micros
    Restriction: fw: 2.60417 micros
    Interpolation + add: 2.60417 micros
    VCycle: 1822.92 micros
    
    
    BENCHMARK
    Jacobi iteration: 5.20833 micros
    Residual calculation: 5.20833 micros
    Restriction: fw: 2.60417 micros
    Interpolation + add: 2.60417 micros
    VCycle: 1822.92 micros
    
    
    BENCHMARK
    Jacobi iteration: 5.20833 micros
    Residual calculation: 5.20833 micros
    Restriction: fw: 2.60417 micros
    Interpolation + add: 2.60417 micros
    VCycle: 1848.96 micros
    
    
    
    ------------------
    -- size: 127
    
    BENCHMARK
    Jacobi iteration: 18.75 micros
    Residual calculation: 18.75 micros
    Restriction: fw: 15.625 micros
    Interpolation + add: 12.5 micros
    VCycle: 2218.75 micros
    
    
    BENCHMARK
    Jacobi iteration: 21.875 micros
    Residual calculation: 18.75 micros
    Restriction: fw: 15.625 micros
    Interpolation + add: 12.5 micros
    VCycle: 2187.5 micros
    
    
    BENCHMARK
    Jacobi iteration: 18.75 micros
    Residual calculation: 18.75 micros
    Restriction: fw: 15.625 micros
    Interpolation + add: 12.5 micros
    VCycle: 2218.75 micros
    
    
    
    ------------------
    -- size: 255
    
    BENCHMARK
    Jacobi iteration: 82.0313 micros
    Residual calculation: 78.125 micros
    Restriction: fw: 74.2188 micros
    Interpolation + add: 46.875 micros
    VCycle: 2890.63 micros
    
    
    BENCHMARK
    Jacobi iteration: 82.0313 micros
    Residual calculation: 78.125 micros
    Restriction: fw: 70.3125 micros
    Interpolation + add: 50.7813 micros
    VCycle: 2890.63 micros
    
    
    BENCHMARK
    Jacobi iteration: 82.0313 micros
    Residual calculation: 78.125 micros
    Restriction: fw: 70.3125 micros
    Interpolation + add: 50.7813 micros
    VCycle: 2890.63 micros
    
    
    
    ------------------
    -- size: 511
    
    BENCHMARK
    Jacobi iteration: 322.917 micros
    Residual calculation: 302.083 micros
    Restriction: fw: 317.708 micros
    Interpolation + add: 187.5 micros
    VCycle: 5729.17 micros
    
    
    BENCHMARK
    Jacobi iteration: 328.125 micros
    Residual calculation: 302.083 micros
    Restriction: fw: 312.5 micros
    Interpolation + add: 187.5 micros
    VCycle: 5781.25 micros
    
    
    BENCHMARK
    Jacobi iteration: 328.125 micros
    Residual calculation: 302.083 micros
    Restriction: fw: 312.5 micros
    Interpolation + add: 187.5 micros
    VCycle: 5781.25 micros
    
    
    
    ------------------
    -- size: 1023
    
    BENCHMARK
    Jacobi iteration: 1304.69 micros
    Residual calculation: 1210.94 micros
    Restriction: fw: 1320.31 micros
    Interpolation + add: 804.688 micros
    VCycle: 24453.1 micros
    
    
    BENCHMARK
    Jacobi iteration: 1312.5 micros
    Residual calculation: 1203.13 micros
    Restriction: fw: 1312.5 micros
    Interpolation + add: 812.5 micros
    VCycle: 24375 micros
    
    
    BENCHMARK
    Jacobi iteration: 1304.69 micros
    Residual calculation: 1203.13 micros
    Restriction: fw: 1328.13 micros
    Interpolation + add: 804.688 micros
    VCycle: 24453.1 micros
    
    
    
    ------------------
    -- size: 2047
    
    BENCHMARK
    Jacobi iteration: 734.375 micros
    Residual calculation: 4656.25 micros
    Restriction: fw: 6718.75 micros
    Interpolation + add: 4640.63 micros
    VCycle: 138594 micros
    
    
    BENCHMARK
    Jacobi iteration: 718.75 micros
    Residual calculation: 4703.13 micros
    Restriction: fw: 6656.25 micros
    Interpolation + add: 4671.88 micros
    VCycle: 138438 micros
    
    
    BENCHMARK
    Jacobi iteration: 734.375 micros
    Residual calculation: 4671.88 micros
    Restriction: fw: 6687.5 micros
    Interpolation + add: 4640.63 micros
    VCycle: 138438 micros
    
    [​IMG]

    [​IMG]

    [​IMG]

    US
     
    #110 Unknown Soldier, Jun 5, 2007
    Last edited by a moderator: Jun 5, 2007
Loading...

Share This Page

  • About Us

    Beyond3D has been around for over a decade and prides itself on being the best place on the web for in-depth, technically-driven discussion and analysis of 3D graphics hardware. If you love pixels and transistors, you've come to the right place!

    Beyond3D is proudly published by GPU Tools Ltd.
Loading...