Recent content by DmitryKo

Interesting. Their previous roadmap was to refactor the open-source DirectX Compiler as a HLSL frontend and DXIL backend in the main branch of Clang/LLVM, instead of making yet another proprietary fork. Though SPIR-V backend has been available for a while, the language of the announcement blog...

We've just reached #175 with 18 billion points! :runaway: We could be ~4 months away from #150, but it would still take more than two years for us to enter the Top 100 at our current rate of ~48M points per day: Rank Project Team Name Points 24hr Avg Points Total Points Diff Gain Daily Date...

Nope, still 30 to 35 °C (86 to 95 °F) holding steady since July..

Well, HIP is not limited to AMD hardware, the runtime can also run on NVidia hardware conforming to CUDA 6.0 or later - so if you program CUDA directly, you can convert your code to HIP and it will still run on NVidia hardware (though HIP is a subset of CUDA, so you will probably lose some...

Isn't that a reference to Nvidia's current 'hardware subscription' model, where Nvidia is installing their own DGX-series supercomputers at major data centers to charge customers a $4500 yearly usage fee per each GPU? https://www.theregister.com/2022/03/25/nvidia_trillion_dollar_revenue/...

BTW here are the slides from a presentation by Victor Peng, the head of the Xilinx division, published on the AMD Financial Analyst Day 2022 (animated GIF): https://ir.amd.com/news-events/financial-analyst-day Victor Peng. Adaptive and Embedded Computing Leadership (FAD 2022_Victor_Peng_Final...

I don't think it's correct. OpenCL specs are open, but most driver implementations are proprietary closed source; and ROCm is an exact opposite of that. OpenCL 1.x C-language spec was originally developed by Apple then submitted to the Khronos Group, so it was not 'crowdsourced'. C++ language...

Microsoft released the first public build of the Dilithium semester in the Canary channel, but there is still no SDK/WDK and no plans to release it "for the time being" - so it seems APIs / headers for new features like shader model 6_9 with Mesh Shader Nodes (Work Graphs tier 1_1) won't...

It has been a very hot summer and the heat coming from my rig has become unbearable, so I had to stop folding for now; I may be able to resume by end of August or in early September. In the meantime, I'm in the process of upgrading my PC; I've already installed a water-cooled Ryzen 7 7700...

New HIP SDK for Windows has been released, version 6.1.2 (download page); still no support for AI libraries (MIOpen, MIGraphX) or AI frameworks (PyTorch, TensorFlow, etc.), but at least this should finally enable OpenMM 8.1.1 kernels in Folding@home client, which have some sizeable performance...

So, AMD designed this 'core parking' control driver for dual-CCD 7000-series X3D processors - and now enabled even for regular 9000-series non-X3D models, BUT... unlike Intel Speed Shift implemented with on-chip hardware, it uses a proprietary kernel-mode driver that can't be uninstalled or...

Exactly, I don't know why there has to be so much drama. 9000-series processors are up to $50 cheaper at launch than the 7000-series, and they seem to offer an average 11-12% performance uplift on similar clocks, along with lower core temperatures; this can be further tweaked with Precision...

Why don't you clone the disk first, then use the MBR2GPT tool on the new NVMe SSD? You can install the NVMe disk in the target PC then run the tool offline from the command prompt on the Windows Recovery USB flash drive (or a custom Windows ISO image prepared with uupdump / rufus). You can also...

BTW part of the reason why WaveMMA is postponed could be that Microsoft DXC HLSL compiler is finally moving from a 10-year-old proprietary fork of Clang and LLVM 3.7 bytecode, which became impossible to update to current C++11/14/17 etc. standards, to an official 'HLSL' front-end of Clang and a...

WaveMMA has never been intended for rendering applications, both NVidia and AMD advertised it for custom ML/AI and compute workloads, as opposed to a predefined set of 'fixed-function' Direct3D/DirectML metacommands implemented by the graphics card driver...