Latest posts tagged with #SIMD on Bluesky
Vectorized, HW accelerated #FFT algorithm for my SH4ZAM fast math/linear algebra library targeting the Sega #Dreamcast's SH4 CPU. Screenshot is a single frame capture from my DC visualizing the frequency domain output of a song in real-time!
#simd #cplusplus #retrogaming #signalprocessing
github.com/NoNaeAbC/std_simd/blob/m...
...this is nice bait, I clicked.
Honestly I like boomer loops. Boomer loops are simple, easy to understand, you just have to avoid sticking twenty-seven of those inside one gigantic function like our current codebase.
#cpp #simd
Informative:
“How Many Registers Does An x86-64 CPU Have?” [2020], William Woodruff (blog.yossarian.net/2020/11/30/H...).
Via HN: news.ycombinator.com/item?id=4701...
#Intel #AMD #x86 #Hardware #CPU #ISA #SIMD #Architecture
Our March issue is now live at www.thelancet.com/gastrohep!
Content includes: #alcohol, #IBD, #HBV, #MASLD, #SIMD (#SIBO) + more!
#GastroSky #LiverSky #MedSky
📢 New ORC 0.4.42 release of the Optimised inner loop Runtime Compiler!
This release contains both bug fixes and new features.
discourse.gstreamer.org/t/orc-0-4-42...
#gstreamer #orc #assembly #release #opensource #simd
[Blog] How could we end 2025 without a terrible dad joke about...
#SIMD #vectors #performance #csharp #dotnet
blog.scooletz.com/2025/12/31/l...
simd-prng, a small webassembly library
devlog.ikumene.com/simd-prng/
#prng #webassembly #emscripten #simd
This is right up my alley. A discussion about the merits of the various #simd implementations. There is even a mention of #linaro and #arm in the episode. #FLOSS Weekly: Episode 857: SOCification b27a1a7a-dbb9-4922-ac3e-601e2d6ca1e1.libsyn.com/episode-857-socification #podcast
@riffraff Functional approach as more strict opens the compiler new doors for optimisation of the code going up to use of #SIMD.
SIMD coding is hard: platforms, inconsistencies, lane constraints... but xsimd abstracts a lot away. With my first contributions, I improved byte shuffling, now available in the latest 14.0 release.
#C++ #SIMD #xsimd #openSource
It's time for #AdventOfCode! This year I'm using #C #neovim #makefile. I'm going to focus on performance, using #SIMD instructions where possible on a #RaspberryPi
Porting DirectX Math to pure C. Taking some inspiration from cglm as well.
github.com/simstim-star...
#gamedev #simd #math #programming #directx #cprogramming #c_programming
COMING UP IN ONE WEEK!
Have you registered yet for the SIMD & @fda.gov Webinar:
Regulatory Use of Biomarkers on December 3 at 2pm ET
Register here- tinyurl.com/SIMDFDADec2025
#FDA
#inbornerrorsofmetabolism
#inheritedmetabolicdisorders
#IEMs
#SIMD
Итоги встречи ISO C++ на Гавайях: начинаем полировку стандарта С++26 Привет! На связи Антон Полухин из Техплатфо...
#contracts #reflection #c++26 #с++26 #с++ #программирование #simd #compile-time #compile #time #compiler-generated
Origin | Interest | Match
Итоги встречи ISO C++ на Гавайях: начинаем полировку стандарта С++26 Привет! На связи Антон Полухин из Техплатфо...
#contracts #reflection #c++26 #с++26 #с++ #программирование #simd #compile-time #compile #time #compiler-generated
Origin | Interest | Match
Итоги встречи ISO C++ на Гавайях: начинаем полировку стандарта С++26 Привет! На связи Антон Полухин из Техплатфо...
#contracts #reflection #c++26 #с++26 #с++ #программирование #simd #compile-time #compile #time #compiler-generated
Origin | Interest | Match
LibMpix Logo
Libmpix on Zephyr OS is an open source pipeline for image processing. #simd and #micropython support are in progress.
https://libmpix.org/
#libmpix #zephyr #microcontroller #mcu #imaging #pipeline #rtos
Via:
https://wiki.pythonlinks.info/libmpix
Folkert has become a co-maintainer of rust-lang/stdarch, which defines vendor-specific APIs that are used by the Rust standard library and Rust users writing explicit SIMD code.
In this blog, Folkert explains how they try to reduce the maintenance burden: buff.ly/X0F6yei
#rustlang #simd
Day 2 at IWOMP had a wonderful keynote by Robert Lucas of Ansys on The Evolutionary Flexibility of LS-DYNA.
IWOMP schedule www.iwomp.org/iwomp-2025/
#OpenMP #IWOMP #DYNA3D #memory #SIMD #HPC
Vectorizing Molecular Docking: Boosting CPU Performance and Portability
A study finds auto‑vectorization with -O3 nearly matches hand‑written SIMD for molecular docking; x86 CPUs have raw performance over ARM, while ARM offers similar energy efficiency Read more: getnews.me/vectorizing-molecular-do... #cpu #simd
Modern CPU performance isn’t just about clockspeed—it’s about how you write your code.
SIMD lets a single instruction operate on multiple data points.
Caden Parker breaks it down in this clear, practical guide.
🔗 vectrx.substack.com/p/simd-a-pra...
#SIMD #Vectorization #SystemsProgramming
Hacker News debated a faster Zig substring search using SIMD. Discussion covered performance, potential optimizations, and the broader impact of SIMD across architectures & languages. Key topics included worst-case scenarios and instruction set trade-offs. #SIMD 1/6
C++ std::simd (ok, std::datapar) is cool. Portable, order-of-magnitude speedups (w/ modern CPUs and the right, flop intense algorithms) 🚀 #HPC #SIMD
github.com/BLAST-Impact...
Achieving peak performance often requires SIMD optimizations. Handwritten assembly or intrinsics frequently outperform compilers because they allow for specific algorithmic and data structure redesigns that auto-vectorization can't achieve. This is where the big wins are. #SIMD 3/6
