New #ZLUDA 5 Preview Released For #CUDA On Non-NVIDIA #GPU
For now this ability to run unmodified CUDA apps on non-#NVIDIA GPUs is focused on #AMD GPUs of the #Radeon RX 5000 series and newer, which is AMD Radeon GPUs with #ROCm. Besides CUDA code samples, GeekBench has been one of the early targets for testing.
https://www.phoronix.com/news/ZLUDA-5-preview.43

#AMD warns of new #Meltdown, #Spectre-like bugs affecting #CPU
Four bugs do not appear too venomous – two have medium-severity ratings other two are rated "low." However, low-level nature of #exploit's impact has nonetheless led Trend Micro and CrowdStrike to assess them as "critical."
Rasons for low severity scores are high degree of complexity involved in successful attack – AMD said it could only be carried out by attacker able to run arbitrary code on a target machine
https://www.theregister.com/2025/07/09/amd_tsa_side_channel/

#ZLUDA Making Progress In 2025 On Bringing #CUDA To Non-NVIDIA #GPU
ZLUDA #opensource effort that started half-decade ago as drop-in CUDA implementation for #Intel GPUs and then for several years was funded by ##AMD as a CUDA implementation for #Radeon GPUs atop #ROCm and then open-sourced but then reverted has been continuing to push along a new path since last year. Current take on ZLUDA is a multi-vendor CUDA implementation for non-NVIDIA GPUs for #AI workloads & more.
https://www.phoronix.com/news/ZLUDA-Q2-2025-Update

Science submits CE mark application for PRIMA retinal implant – A critical step towards making it available to patients in Europe https://science.xyz/news/prima-ce-mark-application-submitted/ #AMD #blindness

#Nvidia, #AMD, #Microsoft, and others court niche Japanese supplier to ease #AI bottlenecks
#NittoBoseki, currently the only supplier in the world of the highest-end glass cloth.
Formally known as low-coefficient-of-thermal-expansion (#CTE) glass and called "#T-glass" by #Nittobo, glass cloth is valued for its dimensional stability, rigidity and ability to facilitate high-speed data transmission, which is critical in AI computing.
https://archive.ph/gl43y#selection-2657.0-2661.85

Say about Linux what you will, but getting this much detail about what your graphics card is doing is pretty cool.

`amdgpu_top --gui`

#Intel #Xeon 6300 vs. #AMD #EPYC 4005 #SMT#/HT Performance Review
At the same core counts, #Zen5 based #EPYC4005 series was showing great benefit out of SMT than the flagship #Xeon6300
Also fascinating to see was that AMD EPYC 4345P even with SMT disabled was still faster than the Xeon 6369P with its full load-out thanks to the EPYC Grado #CPU supporting AVX-512 and other advantages over the Xeon 6300 series that in turn is largely rehashed from the Xeon #E2400 series.
https://www.phoronix.com/review/intel-xeon-6300-amd-epyc-4005-smt

¿#HDMI o #DisplayPort en #Linux? Sin lugar a dudas, DisplayPort, ya que HDMI Forum ha vetado la implementación de la versión 2.1 de su especificación en #AMDGPU.

DisplayPort, por su parte, es una interfaz abierta desarrollada por #VESA, por lo que su implementación es totalmente libre, como en USB.

La pena es que apenas existen capturadoras que se apoyen en DisplayPort en lugar de HDMI. https://www.phoronix.com/news/HDMI-2.1-OSS-Rejected

#AMD #EPYC #4345P 8-Core #CPU Performance Review
The EPYC 4345P was delivering 1.31x the performance overall of the #Xeon #6369P at the same core/thread count.
The EPYC 4345P can be purchased online right now for $329 USD compared to the Xeon 6369P at $606, or 2.4x the performance-per-dollar.
https://www.phoronix.com/review/amd-epyc-4345p

Aviation weather for Sardar Vallabhbhai Patel International airport in Ahmedabad area (India) is “VAAH 140300Z 24004KT 4000 HZ SCT020 31/24 Q1001 NOSIG” : See what it means on https://www.bigorre.org/aero/meteo/vaah/en #ahmedabad #india #sardarvallabhbhaipatelinternationalairport #vaah #amd #metar #aviation #aviationweather #avgeek #airport vl

#AMD's Instinct #MI355X accelerator will consume 1,400 watts. #CDNA4 challenges #BlackwellUltra.
AMD's #MI350X-series #GPU are based on #CDNA 4 architecture that introduces support for FP4 and FP6 precision formats alongside FP8 and FP16.
Both SKUs will come with 288GB HBM3E memory that will offer up to 8 TB/s of bandwidth, but MI350X will offer maximum FP4/FP6 performance of 18.45 PFLOPS, whereas the MI355X is said to push the maximum FP4/FP6 performance to 20.1 PFLOPS.
https://www.tomshardware.com/pc-components/gpus/amds-instinct-mi355x-accelerator-will-consume-1-400-watts

AMD gibt proprietäre 3D-Grafiktreiber für Linux auf

Auch im professionellen Umfeld setzt AMD zur 3D- und Video-Beschleunigung auf quelloffene Treiber, die Linux-Distributionen seit langem standardmäßig nutzen.

https://www.heise.de/news/AMD-gibt-proprietaere-3D-Grafiktreiber-fuer-Linux-auf-10438121.html?wt_mc=sm.red.ho.mastodon.mastodon.md_beitraege.md_beitraege&utm_source=mastodon

I've been lightly annoyed for months on my all-AMD desktop that most Electron apps start, but never show a window (under Wayland, with niri). Killing the GPU sub-process for each app magically makes the window appear. It's not a problem on my also-AMD Framework 13.

Finally got around to investigating, and it appears the issue is from having two GPUs, the integrated graphics of the CPU and my 7800 XT.

Disabling the iGPU in the BIOS fixed it.

RyzenAdj v0.17.0 Adds Support for AMD Strix Halo CPUs: https://github.com/FlyGoat/RyzenAdj/releases/tag/v0.17.0
#linux #linuxgaming #update #release #foss #hardware #amd #strixhalo #cpu #ryzenadj

AMD is dropping their proprietary driver and officially supporting the open source one. I have been saying for a long time that you should ditch your #NVIDIA card and go #AMD if you are on #Linux, this is one more reason about why. https://www.phoronix.com/news/Radeon-Software-Drop-Prop-GL-VK

SteamOS 3.7.8 Brings Official Support to Non-Steam Deck Devices: https://boilingsteam.com/steam-os-3-7-brings-official-support-to-non-steam-deck-devices/
#linux #linuxgaming #update #release #gaming #steamdeck #steam #steamos #lenovo #legiongos #rogally #compatibility #pcgaming #handheld #amd

#AMD Reveals #Ryzen #Threadripper 9000 Family: Big #Zen5 Chips for HEDT and Workstations with 96 cores and 192 threads.
The chips have the same core counts, base clocks, and cache capacities (up to 384MB) as the prior-gen models across the range of the product stack, but the peak boost clocks have been bumped up to 5.4 GHz for all models, an increase ranging from 100 to 300 MHz. The TDP ratings also remain the same 350W for all models.
https://www.servethehome.com/amd-reveals-ryzen-threadripper-9000-family-big-zen-5-chips-for-hedt-and-workstations/
#Threadripper9000