@aeva Currently the "dogshit wall" is #i486SX because #Linux dropped support for #i386 with Versions 3.4.99 & 3.6.9 respectably, tho most people would say anything below #amd64 w/ #SSE2 (or #i686) is out of #mainstream support already, with #RHEL6 being EoL'd & #Debian being the last major distros offering #32bit support.

Making all the difference in the world. #Valve #HalfLife #VR #compilers #x87 #SSE2 https://mastodon.gamedev.place/@TomF/115589904233959578

Much discussion in my feeds around the rustc problem/bug generating bad code to do with SSE2;

https://lists.debian.org/debian-release/2024/11/msg00459.html

Can't say I understand what is the root case there, but rust is not making friends in the vintage/retro computing world.

Nor the folks maintaining old kit for customers.

#i386
#rustc #SSE2 #bug
#Debian 13
#Devuan 6

@cleantext well, it depends on what your productivity tool is.

• 99% of my productivity is in #SSH so any #i486 "#shitbox" with #SSE2 acceleration is gonna be sufficiently fast (I could propably work from a #Pi0W with screen & keyboard attached.

It's also an exercise in #FrugalComputing and asking "What do I actually need to be able to work?" and the answer is pretty simple: An SSH-Terminal!

• My mere customizations would be to do like a "status bar"…

"A library that abstracts over SIMD instruction sets, including ones with differing widths. SIMDeez is designed to allow you to write a function one time and produce SSE2, SSE41, and AVX2 versions of the function."

https://github.com/arduano/simdeez

#asm #rust #programming #opt #optimize #SSE2 #SSE41 #AVX #AVX2 #dev #C0DE #opcodes #amd64 #code #programminLanguage #code

#PaleMoon will be requiring #AVX support in CPUs for its official binaries starting from version 33.3.0, which can be expected to release on the second Tuesday of this August.

For those still stuck with #SSE2 you will either have to upgrade your processor to something more recent (AVX support has been commonly available for 10 years now) or switch to a community build that still builds support for SSE2.

https://forum.palemoon.org/viewtopic.php?f=1&t=31399

EDIT: Oh and I forgot to mention too that this is only for 64-bit. 32-bit Windows binaries will still support SSE2 (there are not 32-bit only processors that supports AVX as far as I know anyway, and even if there are it would most likely result in worse performance due to how 32-bit x86 and AVX work)

#nibrans is a #publicdomain nibble based rANS coder.

nibrans is an #singleheader #SSE2 accelerated encoder and decoder for the rANS arithmetic coding format. rANS offers near-optimal data compression while having an adaptable frequency structure, unlike Huffman coding which uses a static structure. nibrans encode and decode data at a rate of up to 75-95 megabytes per second.

Website 🔗️: https://github.com/BareRose/nibrans

#free #opensource #foss #fossmendations #programming #library #compression

#julialang #mingw64 #gcc #dSFMT #OpenMP #SSE2

二大怪獣最終決戦？😁

10億回のループでは短い感じなので、40億回にしてみました。

適当にあいだをあけながら二三十回試して一番短い方の時間をピックアップしました。

結果

mingw64 gcc with dSFMT, SSE2, and OpenMP → 5.3秒

julia @paralle → 4.3秒

Julia @parallel 強いです。

なかなか倒せない。数字がばらつくのですが、安定してJuliaの方が速い感じです。

統計処理をしてみようかと思ったのですが、色々面倒なのでやめました。

CPUの温度が計算にかかる時間のゆらぎに関係しているっぽい雰囲気。 https://mathtod.online/media/ga6fhfgzumvcQKkyZb0