Elektrine lite

← Feed

Dr. Moritz Lehmann

ProjectPhysX@mast.hpc.social

<p>Summa cum laude Physics <a href="https://mast.hpc.social/tags/PhD" class="mention hashtag" rel="tag">#<span>PhD</span></a> 🖖🧐🎓 | Graduate at EliteNet Bavaria 🧬 &amp; DLR 🛰 | Developer of <a href="https://mast.hpc.social/tags/FluidX3D" class="mention hashtag" rel="tag">#<span>FluidX3D</span></a> <a href="https://mast.hpc.social/tags/CFD" class="mention hashtag" rel="tag">#<span>CFD</span></a> 🌊 | Khronos <a href="https://mast.hpc.social/tags/OpenCL" class="mention hashtag" rel="tag">#<span>OpenCL</span></a> Advisor 💻 | <a href="https://mast.hpc.social/tags/GPU" class="mention hashtag" rel="tag">#<span>GPU</span></a> Wizard at <a href="https://mast.hpc.social/tags/Intel" class="mention hashtag" rel="tag">#<span>Intel</span></a> 🟦</p><p><a href="https://github.com/ProjectPhysX/FluidX3D" ta

Posts

  • View post

    The sun looked very very strange today ☀️🌑🔭🖖🤔 Some photos from my phone and with the phone through binoculars - unedited 🔭 #solareclipse #eclipse #spain

  • View post

    hw-smi v1.6 brings support for data logging! You have requested an option to log the telemetry data (#GPU​/VRAM usage, #VRAM bandwidth, temperature, power, fan speed, #PCIe bandwidth etc.) from hw-smi to a file. Today I have implemented exactly that. Have fun monitoring your hardware and applications, be it #Intel Arc (Pro) or any other #Nvidia​/​#AMD GPUs on Windows and #Linux! 🖖 https://github.com/ProjectPhysX/hw-smi/releases/tag/v1.6

  • View post

    *chuckles* I&amp;#39;m in danger Dual #IntelArc&amp;#39;s (an A750 and a B580) on an ancient Z87 mainboard from the dumpster with an i7-4771, good 600W PSU for 2 GPUs, Frankenstein 8+4+8+4 GB RAM config, HDD drive, fire-hazard Molex-to-6pin adaptor, the do-not-use-me red LAN cable plugged in, no monitor, no case, sitting on the edge of my desk. Everything about this build is perfect! 🖖🤪

  • View post

    #FluidX3D v3.8 is out, adding local memory optimization to the reamining rendering kernels. The previous update made the kernel for Q-criterion isosurface rendering use too many (152 32-bit) registers, which caused the kernel to fail execution on old Nvidia Maxwell/Pascal #GPU​s. Through register occupancy optimization I managed to squeeze it into 122 32-bit registers, just below the limit of 128. Full rendering performance even on the old hardware, no excuses. 🖖😎 https://github.com/ProjectPhy...

  • View post

    This was the 2nd demo I had showcased at #SC25 in St. Louis! Forgot to upload, here we go 🖖🧐 This is the NASA X-57 multi-rotor airplane simulated in #FluidX3D #CFD, on 4x #IntelArc Pro B60 GPUs with 24GB each, at a massive 1.8 Billion grid cells resolution. A good showcase of the mesh center-of-mass utility function for rotation of the 3-/5-bladed rotors. Although grid resolution struggles a bit without refinement (yet ;) https://www.youtube.com/watch?v=ehKuuLVAAtU

  • View post

    Omg omg omg it works!! Dual #IntelArc - B580 and A750 - on an ancient #Intel Haswell/Z87 mainboard from the dumpster! 🖖😎🚮 Kudos to roschank for enabling ReBAR with multi-GPU through BIOS modification - https://github.com/roschank/reBART/issues/2

  • View post

    Let&amp;#39;s try something: You can plug an animated GIF into the #GitHub social preview image. Good eye-catcher in the GitHub topics listing and when sharing repos. Also: Already 2 updates for hw-smi! Better #Intel Arc Alchemist &amp;amp; iGPU support, and more robust compiling across #Linux distros. Users testing on lots of systems and sending me feedback is win-win! https://github.com/ProjectPhysX/hw-smi

  • View post

    #FluidX3D #CFD has reached ⭐ 5000 Stargazers on #GitHub! 🖖🥳 Grid refinement update is still in development, I haven&amp;#39;t forgotten... ⬜◻️◽▫️ https://github.com/ProjectPhysX/FluidX3D

  • View post

    Newest #IntelArc #GPU family member is here, the Panther Lake Arc B390... and it... purrs? 🖖 🥺 🐈‍⬛ My OpenCL-Benchmark on the B390 measures ~7.4 TFlops FP32 and ~120GB/s memory bandwidth. hw-smi also works with the B390. #FluidX3D benchmarks here: https://github.com/ProjectPhysX/FluidX3D#single-gpucpu-benchmarks And the #OpenCL infos: - Arc B390: https://opencl.gpuinfo.org/displayreport.php?id=6718 - Core Ultra X7 358H: https://opencl.gpuinfo.org/displayreport.php?id=6717

  • View post

    #FluidX3D #CFD v3.7 brings faster Q-criterion isosurface rendering with #OpenCL local memory optimization! 🖖🤠 https://github.com/ProjectPhysX/FluidX3D/releases/tag/v3.7 Instead of 32 velocities for each #GPU thread, now an 8x8x8 workgroup loads &amp;amp; reuses 11x11x11 velocities in L1$, a 12x VRAM BW reduction. Fascinating insight: Which thread loads which cell from VRAM to L1$, and which thread renders which grid cell within the workgroup, can be very different! https://github.com/ProjectPh...

  • View post

    #OpenCL 3.1 is here!! 🖖🥳🎉 https://www.khronos.org/blog/opencl-3.1-is-here

  • View post

    Heh, I found (and reported) a bug in one of #Nvidia​&amp;#39;s oldest and probably most thoroughly tested libraries: NVML memory clocks are wrongly reported in (MT/s / 2) instead of MHz, which means they are wrong for all #GPU​s with GDDR5-7 memory types. Really no code can be trusted. 🖖🫠 https://github.com/ProjectPhysX/hw-smi/blob/master/src/main.cpp#L416