Elektrine
Log in Register
Paige Chat Timeline Gallery Friends Email Drive DNS Private DNS Domains VPN Kairo Nerve
Remote

Dr. Moritz Lehmann

@ProjectPhysX@mast.hpc.social
mastodon 4.7.3
  • Open on mast.hpc.social

Summa cum laude Physics #PhD 🖖🧐🎓 | Graduate at EliteNet Bavaria 🧬 & DLR 🛰 | Developer of #FluidX3D #CFD 🌊 | Khronos #OpenCL Advisor 💻 | #GPU Wizard at #Intel 🟦

https://github.com/ProjectPhysX/FluidX3D

614 Followers
506 Following
12 Posts
Joined December 10, 2022
GitHub:
https://github.com/ProjectPhysX
YouTube:
https://youtube.com/@ProjectPhysX
Open post
Dr. Moritz Lehmann @ProjectPhysX@mast.hpc.social
· 4mo ago

#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 & 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/ProjectPhysX/FluidX3D/blob/master/src/kernel.cpp#L2827-L2956

PS: plugged X-wing Gif in #GitHub preview 🖖😜

mast.hpc.social

HPC.social Mastodon

11
0
4
0
Open post
Dr. Moritz Lehmann @ProjectPhysX@mast.hpc.social
· 5mo ago

Heh, I found (and reported) a bug in one of #Nvidia​'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

mast.hpc.social
14
0
6
0
Open post
Dr. Moritz Lehmann @ProjectPhysX@mast.hpc.social
· 2mo ago
Replying to
@niconiconi@mk.absturztau.be they sure gonna hoard and mod some cards before they make the memory unlock live on the repo. Gotta properly test the exploit, for purely scientific purposes of course^^
2
5
0
0
Open post
Dr. Moritz Lehmann @ProjectPhysX@mast.hpc.social
· 5mo ago

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

mast.hpc.social
6
0
5
0
Open post
Dr. Moritz Lehmann @ProjectPhysX@mast.hpc.social
· 5mo ago

Let'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 & 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

GitHub

GitHub - ProjectPhysX/hw-smi: A minimal, cross-compatible CPU/GPU telemetry monitor with accurate data directly from vendor APIs and beautiful ASCII visualization.

A minimal, cross-compatible CPU/GPU telemetry monitor with accurate data directly from vendor APIs and beautiful ASCII visualization. - ProjectPhysX/hw-smi

7
0
0
0
Open post
Dr. Moritz Lehmann @ProjectPhysX@mast.hpc.social
· 5mo ago

#FluidX3D #CFD has reached ⭐ 5000 Stargazers on #GitHub! 🖖🥳
Grid refinement update is still in development, I haven't forgotten... ⬜◻️◽▫️
https://github.com/ProjectPhysX/FluidX3D

mast.hpc.social

HPC.social Mastodon

6
0
1
0
Open post
Dr. Moritz Lehmann @ProjectPhysX@mast.hpc.social
· 5mo ago

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

mast.hpc.social

HPC.social Mastodon

3
0
0
0
Open post
Dr. Moritz Lehmann @ProjectPhysX@mast.hpc.social
· 5mo ago
Replying to
@sjvn@mastodon.social that's right! It goes in the square hole.
1
0
1
0
Open post
Dr. Moritz Lehmann @ProjectPhysX@mast.hpc.social
· 5mo ago
Replying to
@giuseppebilotta@fediscience.org that effective throughput depends on the algorithm. Usually something like particles/s or grid cells/s. For LBM, the latter is called MLUPs/s, Mega Lattice UPdates/second.
1
0
0
0
Open post
Dr. Moritz Lehmann @ProjectPhysX@mast.hpc.social
· 2mo ago
Replying to
@azonenberg@ioc.exchange @Methylzero@mast.hpc.social @niconiconi@mk.absturztau.be for #HPC workloads the low arithmetic intensity is just perfect. High memory bandwidth (1.5 TB/s) and capacity are the star of the show. Even with half the die size and compute, you would still get the same performance in lattice Boltzmann for example, so memory bound these codes are. Similar story for some AI inference workloads, and that's what drives the market price of pretty much all GPUs nowadays.
0
2
0
0
Open post
Dr. Moritz Lehmann @ProjectPhysX@mast.hpc.social
· 2mo ago
Replying to
@niconiconi@mk.absturztau.be holy shit they unlocked the full 80GB VRAM too. Makes those cards quite a bit more valuable and useful. Nice!
0
9
0
0
Open post
Dr. Moritz Lehmann @ProjectPhysX@mast.hpc.social
· 2mo ago
Replying to
@niconiconi@mk.absturztau.be 🖖🏻😋🍿🍿🍿
0
7
0
0
Back
313k7r1n3
Elektrine

Tor hidden service

elekhj7afj4qnrr4yd3bkzslsyo5jgfxw3orgjkhlcxifueodybyiiad.onion

I2P eepsite

j6b6cyk6gjmepjih7jjadxgxvvf3lzzujljuu2v4biemzpg3naya.b32.i2p

Platform

  • Email
  • Chat
  • Timeline
  • VPN
  • DNS

Company

  • About
  • Contact
  • FAQ
  • Lite (no JS)

Legal

  • Terms of Service
  • Privacy Policy
  • Transparency Report
  • Report Abuse
  • Warrant Canary
  • VPN Policy

Support

  • support@elektrine.com
  • Report Security Issue
Mail client setup IMAP mail.elektrine.com:993 POP3 mail.elektrine.com:995 SMTP mail.elektrine.com:465
© 2026 Elektrine. All rights reserved. Server: 21:19:46 UTC