↩ Accueil

Vue lecture

Trying Out The AMD Developer Cloud For Evaluating Instinct + ROCm

Last week alongside announcing the AMD Instinct MI350X/MI355X and the ROCm 7.0 software preview, AMD also introduced the AMD Developer Cloud as a new means for developers to easy try out Instinct accelerators with their own software and with the ROCm compute stack already setup. Having tried out prior AMD cloud compute environments, as soon as my email invite for the AMD Developer Cloud arrived I decided to give it a try.
  •  

Rust Surveying Developers To Find Biggest Compiler Performance Issues

Rust developers acknowledge lengthy compiler times can be a significant issue that limits the productivity of developers working with this programming language. For helping in determining different combinations of issues around compiler performance, the Rust team has started a survey to collect more information on the issues...
  •  

FreeRDP 3.16 Released With Better SDL3 Client Support

FreeRDP 3.16 is out today as the newest update to this open-source Remote Desktop Protocol (RDP) library and client implementation. This Apache-licensed project continues to be one of the leading implementations of the Microsoft RDP protocol for use outside the confines of Windows...
  •  

Linux 6.16-rc2 Released With An Initial Batch Of Fixes

Following the release of Linux 6.16-rc1 last Sunday that capped off the Linux 6.16 merge window, Linux 6.16-rc2 is now available with an initial week's worth of bug/regression fixes. Linux 6.16 development continues in aiming toward a stable release around the end of July...
  •  

Linux 6.17 Looks Like It Could Go Ahead And Make SMP Support Unconditional

Back in May a big patch series was published for reworking the Linux kernel to make the SMP support unconditional. Right now those that happen to be running Linux in a uniprocessor (1 CPU core) configuration can build with "CONFIG_SMP" disabled but the proposed patches would make symmetric multi-processing support always present. Those patches took a step forward this week and could be merged for the Linux 6.17 cycle later in the summer...
  •  

Linux 6.16 Lands Proper Power Management Fix For Code That Caused Power Regression

Linux 6.15 mistakenly shipped with a nasty power regression for some systems, such as those relying on the "nosmt" option to disable Simultaneous Multi-Threading / Hyper Threading. That idle power regression was fixed for Linux 6.15.2 and Linux 6.16 Git by reverting the troubled patch that introduced the regression. Now merged ahead of Linux 6.16-rc2 is a proper fix for that problematic patch so it could be re-merged without the power fallout...
  •  

Intel Begins Preparing Linux For Next-Gen DSA 3.0 Accelerators

In addition to Intel recently upstreaming Linux support for new QAT "Gen 6" hardware as their next-generation QuickAssist Technology IP, Intel today began posting Linux kernel driver patches for a new version of their Data Streaming Accelerator (DSA). It looks like upcoming Xeon processors will be rolling out a lot of new accelerator IP...
  •  

LibreOffice 25.8 Beta Released For Testing

LibreOffice 25.8 beta is now available for this popular open-source office suite. LibreOffice 25.8 has been baking many improvements for this popular Microsoft Office alternative and leading office suite option for the Linux desktop...
  •  

Google Cloud C4D Performance Benchmarks At The Top-End Show 39% Generational Improvement With EPYC Turin

Back in April at Google Cloud Next was the introduction of the new C4D family of VMs powered by AMD EPYC 9005 "Turin" processors. Back on launch day I looked at the C3D vs. C4D performance at some of the smaller, more common VM sizes. In today's article is a look at the top-end performance of the C4D family with 384 vCPUs. For those wondering about the compute potential of the c4d-standard-384, here are some benchmarks of this 192-core / 384-thread EPYC Turin configuration compared to the prior C3D AMD EPYC Genoa based instance that topped out at 360 vCPUs.
  •  

OpenMoonRay Introduces NUMA Support

Two years ago DreamWorks Animation open-sourced their MoonRay renderer that is an award-winning, state-of-the-art production MCRT renderer used for a number of feature films. Since then they have continued advancing this open-source code as OpenMoonRay and adding more features. The newest feature release of OpenMoonRay is now available with yet more capabilities for this impressive renderer...
  •  

Measuring The Performance Cost To AMD Memory Guard With Ryzen AI PRO CPUs

While the AMD Ryzen AI Max+ PRO 395 and Ryzen AI Max+ 395 are very similar processors just as the Ryzen PRO processors are to other non-PRO parts, one of the differences with the AMD PRO Technologies come down to AMD Memory Guard providing full system memory encryption. From the HP BIOS with the ZBook Ultra G1a there is a convenient toggle for this full memory encryption support and thus I decided to carry out some benchmarks to measure the performance cost to this memory encryption feature on this AMD Strix Halo SoC.
  •  

HP ZBook Ultra G1a: An Incredible, Powerful Mobile Workstation Powered By AMD's Ryzen AI Max

Over the past month I have been testing out the HP ZBook Ultra G1a powered by AMD Ryzen AI Max+ PRO Strix Halo. Simply put: WOW! I don't remember the last time I have been so fascinated by a laptop SoC from its incredible performance generationally and even compared to existing AMD SoCs within the Ryzen AI 300 series and outright dominating against the Intel Lunar Lake for its Xe2 integrated graphics. The HP ZBook Ultra G1a thanks to AMD Strix Halo offers an incredibly potent integrated GPU and allowing up to 16 cores / 32 threads offers immense CPU performance too. HP packages Strix Halo up into a very well built, mobile workstation oriented laptop design to create an amazing laptop. It's a reliable laptop with captivating performance but does carry a high price tag but with good Linux support too except for one caveat.
  •