@tagesschau@DLR@dlr_next@stim3on 2 hours later, May 11 00:39-00:54, round 2. The #aurora coveres the entire sky over Bavaria, Germany. It's even visible to the naked eye when looking south! This is insanity!! 🖖😳
First image even looks like the camera Sensor of my phone got hit with radiation particles. #polarlights#northernlights#Bavaria#Germany
#FluidX3D#CFD update alert! v2.15 speeds up framerate in interactive graphics by 20-70%. 🖖🥳💻
How? Turns out iterating over 2 million pixels with a single CPU core is... really slow. I did that 3 times more than necessary for every frame rendered on screen! 🖖😆
I've now eliminated a memory copy of the frame (in favor of pointer swap), and a clear frame/zbuffer operation on CPU since that's already done on #GPU.
One of my #PhD papers got selected for the 2022 Best Paper Award of MDPI Computation! 🖖🥳📃🏆
That was a very bold publication for multiple reasons:
I solo-authored it
I wrote that paper in only 2 weeks
the title contains "Esoteric" twice
I submitted it on April 1st
It's serious science though: I discovered a simple algorithm to cut memory demand of the #LBM in half, allowing huge simulations on cheap #GPUs. This is one of the key innovations in #FluidX3D#CFD.
How realistic can a #CFD simulation be? Here is a 1 billion cell #FluidX3D simulation of an impacting raindrop, fully raytraced in 8K. FluidX3D contains state-of-the-art volume-of-fluid and surface tension models for highly accurate free surface simulations. Combined with my own #OpenCL#raytracing engine, results are rendered on-the-fly at resolution as large as remaining #GPU VRAM can hold. 🖖😋💧📺 https://youtu.be/MmLNQIW_Sic
FluidX3D is on #GitHub: https://github.com/ProjectPhysX/FluidX3D
@badlogic for 2 GPUs better go with a mainboard that supports PCIe bifurcation to two x8 slots to the CPU. Some Z690/Z790 mainboards support this, like the Taichi ones. And make sure the 4090s are only 3-slot, as all the 4-slot models will block the second PCIe slot.
For AI stuff, 3090 (non-Ti) will perform about the same but are cheaper, and 2x 100W less power.
So the RTX 2080 Ti in my current office workstation is starting to get a little cramped for me. If I have a large KiCAD design and a ngscopeclient session with a lot of waveforms and filters open simultaneously, I often run out of VRAM.
More compute would be nice as long as it doesn't come with a power budget much higher than my existing 2080 Ti (250W TDP). I plan to stick with NVIDIA since I'm very familiar with their shader debug tools etc.
The only option with more VRAM in the consumer space is the RTX 4090 which I'd like to avoid due to the ludicrous 450W TDP and incompatible power connector.
@azonenberg Nvidia Ada has crippled VRAM bandwidth on all but the very expensive high end. Cheapest good option a 2nd hand 3090 (non-Ti) and undervolting it to reduce TDP.
This is wild: #FluidX3D can "SLI" together 🔵 #Intel Arc A770 + 🟢 #Nvidia Titan Xp, pooling 12GB+12GB of their VRAM for one large 450M cell #CFD simulation. Top half on A770, bottom half on Titan Xp. They seamlessly communicate over PCIe. Performance is ~1.7x of what either #GPU could do on its own. 🖖😋🖥🔥 #OpenCL shows its true power here - one implementation works on literally all GPUs at full performance, even at the same time. Happy #SimulationFriday! https://youtu.be/PscbxGVs52o
The circle to search feature on the S24 series works so unreasonably well. Took this random 50x zoom photo, did a quick circle and right away got an answer.
This feature in particular uses Google for the actual searching, but even the completely self-developed Galaxy AI features worked surprisingly well. More thoughts in the Friday Checkout
@Techaltar such online AI features are a nightmare for data privacy. Send all your photos straight to Google, hallelujah. And of what use is this phone when I can't even plug in headphones? 🚫🎧
@Techaltar@champingsajt do you really think it performs the web search only upon request?
It looks more like it does it in the background for any new photo you take, sends all the data to Google, and caches the search URL so that when the request comes it instantly has the search result.
The final part of my #PhD thesis has now been accepted and published in #Microplastics and #Nanoplastics! 🖖🥳📃🎓
I'm proud to have coauthored this study by Lisa Marie Oehlschlägel. We looked at water-air transfer of microplastics during #bubble bursting in lab experiments, with surprising results: https://doi.org/10.1186/s43591-023-00079-x 🌊🫧💥
OK so I'm ready for today's #GPGPU lesson with the new laptop. My only gripe for the lesson will be that #Rusticl in #Mesa 23.2 doesn't support #profiling information. Apparently the feature was merged at a later commit https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24101
and I even tried upgrading to my distro's experimental 23.3-rc1 packages, but trying to use rusticl on those packages segfaults. So either I've messed up something with this mixed upgrade, or I've hit an actual bug.
This thread https://mk.absturztau.be/notes/9lain2utf5untfm4
by @niconiconi is both fascinating and frustrating. #NVIDIA has a bad habit of doing market segmentation in software (there have been some infamous cases of NVIDIA releasing driver updates that just uncrippled their desktop GPU performance on new #AMD releases, bringing their performance on par with the equivalent workstation and server GPUs). I wouldn't be surprised if this were the case for the hardware @niconiconi is experimenting on.
@giuseppebilotta@niconiconi Nvidia have a long history of artificially crippling their GeForce lineup to not eat into the more profitable Quadro market. Either with drivers, or by motivating certain companies to make their "professional" software run like shit if the GPU name is not on the Quadro list. https://youtu.be/uwCu-b7htV8
I suspect similar marketing reasoning with software-crippling the mining cards. Software locks are a major root cause for e-waste unfortunately.
#FluidX3D has passed 2000 Stars! It is the most popular #CFD software on #GitHub now! 🖖😊⭐️ https://github.com/ProjectPhysX/FluidX3D
Feeling blessed that my work is useful to so many people across the globe, with users in 75 countries already! 🌍
42% EU, 30% Americas, 25% Asia, 3% Oceania+Africa
The red lightning bolt continues: #FluidX3D has passed 3000 Stargazers on #GitHub - from 82 countries! 🖖🥳⭐
Releasing this software for free really has turned out win-win: I've received so much valuable feedback, and answered with as many bug fixes and updates, with many more to come. I am enabling cutting-edge #CFD simulations for everyone, with very little hardware resources, on literally every computer that has a #GPU, regardless of vendor.
👉 https://github.com/ProjectPhysX/FluidX3D #SimulationFriday#OpenCL
