lseurttyuu/GPU-T v0.1.5

GPU-T v0.1.5

on GitHub

🚀 The Road to GPU-T v0.2.0

This release serves as a "bridge" update and a stability dry run. Behind the scenes, a lot of structural changes have been made to the codebase to prepare for official Nvidia support. Before that goes live, v0.1.5 ensures that the core application remains perfectly stable for current AMD users.

Mark your calendars: GPU-T v0.2.0 with full Nvidia support drops exactly one week from today on May 9th at 4:00 PM CEST! I will be doing a full deep dive into the new Nvidia architecture, the sidecar binary, and how it all works under the hood in those release notes.

For now, here is what is new in v0.1.5:

⚙️ Under-the-Hood

• Database Decoupling: The monolithic JSON GPU database has been decoupled into four separate files: amd_gpu_db.json, nvidia_gpu_db.json, intel_gpu_db.json, and gpu_vendors_db.json. This simplifies the internal structure and optimizes memory usage by only loading the specific vendor databases that match the hardware actually installed in your PC.
• AppImage Delta Updates Fixed: The AppImage versioning pipeline has been fixed to pull directly from the .csproj file, fully resolving the X-AppImage-Version=UNKNOWN bug. Furthermore, .zsync files are now officially generated and published alongside the .AppImage, this should resolve the false update loops in AM (AppManager) and enable proper, bandwidth-saving delta updates moving forward.

🗄️ Database Updates

• Added hardware specs for: AMD Radeon HD 8610G.

Installation: Download the .AppImage, make it executable (chmod +x), and run! For detailed instructions, check the README.