Hi everyone,
I’m reaching out because I’m at my wits' end with my Mac Pro 6,1 (Late 2013) running OCLP. I love this machine, but the experience has become a "karma" of instability. Here are my specs:
- Model: Mac Pro 6,1 (Trashcan)
- CPU: Intel Xeon (Standard)
- RAM: 48 GB
- GPU: Dual AMD FirePro D500
- OS: macOS Sonoma (currently using OCLP 2.4.1)
The Issues:
- Visual Glitches: Chrome and other Electron-based apps show "red blocks" or artifacts.
- General Lag: The UI feels heavy, windows stutter, and there is a noticeable delay when opening apps.
- Stability: The system becomes unresponsive during hardware-intensive tasks (like running local LLMs via Ollama).
What I’ve Tried So Far:
- Updated to the latest OCLP (2.4.1) and re-applied Root Patches.
- Attempted to disable one of the GPUs using
-wegnoegpuin boot-args, but I ran into "Error setting variable not permitted" (SIP issue) and UI limitations within the OCLP app (the NVRAM tab or Advanced settings aren't showing as expected). - I work remotely on this machine, so physical intervention (like removing a GPU) is a high-effort/high-risk last resort.
My Questions for the Experts:
- Is there a specific stable OCLP version or configuration known to work better with the dual D500s?
- How can I properly disable one of the GPUs in OCLP 2.4.1 specifically for this model to test if one card is failing?
- Are the "Red Chrome" glitches a known driver issue with Metal translation on Sonoma for these specific AMD cards?
- Should I give up on Sonoma/Sequoia and revert to Monterey (Native) to maintain stability for local AI (Ollama) work?
I really want to keep this Mac Pro alive, but I need it to be a reliable workstation. Any advice from fellow 6,1 owners or OCLP devs would be greatly appreciated.
Thanks in advance!
[link] [comments]
Post a Comment