Search results

Loading Google Results...
  1. mitch_de

    NootRX

    Yep, no one will reach the theoretical PCIE bandwidth speeds (doesnt matter 16 GB/s or 32 GB/s Limit). There is only an prob if you get xy GB/s below the next slower PCIe Version - for example you habe PCIe4 System and getting much below 16 GB/S. PCie4 Systems will get 19-26 GB/s. Another Speed...
  2. mitch_de

    NootRX

    Yep, WEG is needed for non newer NAVI gpus - i think. NootRX wasnt made for older AMDs.
  3. mitch_de

    NootRX

    Some update of NootRX - for 14.4 . https://github.com/ChefKissInc/NootRX/commit/fa929990d1c3e0d72bc5e9f97fc02f735785462c Great that the dev is working hard to save the future usage!
  4. mitch_de

    NootRX

    PCIE Transferspeed is a bit low for an fullspeed PCIe4 x16 Lane system. Can you run the app again if same low 12 GB/s? OS X system info shows also basic PCie Info abour the gpu card: AMD Radeon RX 5600 XT: Chipsatz-Modell: AMD Radeon RX 5600 XT Typ: GPU Bus: PCIe PCIe-Lane-Breite...
  5. mitch_de

    NootRX

    Updated GPU Bench Nbody to V2.5: - added function to run nbody Gflops again (Button) - Nbody now runs longer - 30 Sec and also Result will be the average GFlops - added Info about autor of orig. Nbody Metal , James Price - click on the the link to shown his github website viewn within the app...
  6. mitch_de

    NootRX

    Hi, thanks for your comments. 1. The OpenCL API is only used to measure PCIe & VRAM Speed! CPU is there for sure involved because ists in the que of the transferway CPU-PCIe-VRAM. CPU speed for that two values doesn't matter. 2. Metal API is used for measure GFLOPS. Its GPU only .. but as always...
  7. mitch_de

    NootRX

    It was updated in post: https://www.tonymacx86.com/threads/nootrx.327414/post-2395311
  8. mitch_de

    NootRX

    What i know is, that multiple displays reduce the always the available VRAM , but should normally not reduce also compute speed - only if VRAM would get to low - but Nbody doesnt use very much VRAM.. less than 100MB!
  9. mitch_de

    NootRX

    Now Main gpu nbody Metal Bench updated to V 2.4 (main changes: nbody can bench also newer, fast GPUs) EDIT: new DL, codesigned again.
  10. mitch_de

    NootRX

    Yep, great. And you see my guess of around 13000 GFLOPS - was Ok. I will include that new GFLOPS (no more limit for fast , newer gpus) in the next days into the main GPU Bench Metal app. be patient .... but for tests of GFLOPS only you can use latest Test Nbody also ... I fear the RX 6950XT...
  11. mitch_de

    NootRX

    No real problem. GFLOPS in your case seems to be above 9999 GFLOPs .. GREAT!! ... will fix that Your GPU reaches 38 FPS much higher than my 16FPS 5600Xt and also higher than 24 FPS (8750 GFLOPS) with 6700XT values with new version ... both have enough gap to the 60 FPS limit now. Because FPS &...
  12. mitch_de

    NootRX

    Yep, test is + should running endless. Simple look for the highest GFLOPS + FPS you see. You must Quit the test to stop with Q. After I know that faster GPUS will not run close to 60 FPS (= reaching FPS limit) I will put that code in the orig. app - then it will find highest measured GFLOPs...
  13. mitch_de

    NootRX

    Hi, all. I am the dev the GPU Bench Nbody Metal. The Nbody GFLOPS seems to run in an 60 FPS limit ... your newer GPUs ware tooo fast ;) As I made the app, 5600XT was high end. I made a new build of the Nbody GFLOPS part of the App. Please try it and tell the new shown GFLOPS (+ shown FPS, near...
  14. mitch_de

    NootRX

    Yep, 6950XT fps VVAI are a bit low compared with the 6800XT. But maybe the PCIe Speed in his older Mainboard is not PCIe4? Also older highend CPU/DDR4 vs new highend CPU /DDR5 6000! has an effect why the 6950XT cant show her real power? You can try to check PCIe Speed (beside VRAM Speed +...
  15. mitch_de

    NootRX

    An good first step for searching reason of the reset / fail with Topaz can be first to have an eye on the TEMP of the AMD 6950 XT GPU. Theoretical the gpu should trottle down clk before overheats but .... The is an Radeon Gadget , Radeon Sensor which shows the temp while running such heavy load...
  16. mitch_de

    NootRX

    Thanks for your RX 6700 XT & 6800 Results! Helps me to decide buying the "right" used RX 6xxx card as an RX 5600XT upgrade. Yep, Topaz Video AI Benchmark takes some time ;) But its also good test for stability of GPU power/temp stability.
  17. mitch_de

    NootRX

    Hi, i follow this tread because i will upgrade my RX 5600 XT to an RX 6650 XT or RX 6700 XT (DRM not needed). I often use video tools which use the GPU for some tasks beside dec(/enc. One Bench is the Neatvideo Bench (the company produces plugins)...
  18. mitch_de

    AMD Radeon Performance Enhanced SSDT

    Thanks! You are right to wait - if i have time - for an upgrade. I think i will look for an used RX 5600 XT and perhaps get this card end of 2022 for an normal / near normal price.
  19. mitch_de

    AMD Radeon Performance Enhanced SSDT

    I am on RX 570 ... looking for an new. bit faster GPU. I think 5500 XT isnt the rigth for me. Performance little bit above an RX 580-590 would be enough boost for me. Anyone have RX 5600 (XT) or RX 6500/6600 bench results? Geekbench is Version 5.4!!!, not comparable with GB 4.x bench values.
  20. mitch_de

    Big Sur Graphics performance

    This benchmark is really old - years old and only OpenGL - better use new BasemarkGPU 1.2 (Metal!!, also M1 native)!!! https://www.basemark.com/benchmarks/basemark-gpu/ (Even DL name is with ..BIG SUR, it also runs with Catalina ;) ) Some results from their result database (OS X): 1.2 MEDIUM...
Back
Top