Actually v7 had the same Windows shutdown problem, at least in a number of versions. It just handled some of the error conditions when a core failed with a WU better.
Only 60% of WUs are Credited
Moderators: Site Moderators, FAHC Science Team
-
muziqaz
- Posts: 1410
- Joined: Sun Dec 16, 2007 6:22 pm
- Hardware configuration: 9950x, 7950x3D, 5950x, 5800x3D
7900xtx, RX9070, Radeon 7, 5700xt, 6900xt, RX 550 640SP - Location: London
- Contact:
Re: Only 60% of WUs are Credited
That's the first time I'm hearing this. I never had issues with v7 losing WUs upon windows restart. Especially when v7 had this pause state bug, where you had to wait a solid minute for your pause to register and be saved. So many times I would pause v7 and reboot and see WU restart after reboot, leaving me thinking: Wait, didn't I pause it before system restart?
FAH Omega tester
-
Joe_H
- Site Admin
- Posts: 8074
- Joined: Tue Apr 21, 2009 4:41 pm
- Hardware configuration: Mac Studio M1 Max 32 GB smp6
Mac Hack i7-7700K 48 GB smp4 - Location: W. MA
Re: Only 60% of WUs are Credited
The v7 Pause state was written to the config.xml file. That was done well after the stop signal was sent to the folding cores and they had time to exit. But the client may not have written out the config.xml file before Windows shutdown caused the client process to exit.
I would have to dig through the v7 change logs to see where Joseph made changes to the client around shutdown processing and such. Another thing that changed for the CPU folding cores was that older versions just exited when given the signal. I think beginning with the A8 core it would attempt to write out a checkpoint that would be newer than the default one done every 15 minutes. It might have been earlier with the A7 core which Joseph was tasked with writing to be an update from the A4 core.
I would have to dig through the v7 change logs to see where Joseph made changes to the client around shutdown processing and such. Another thing that changed for the CPU folding cores was that older versions just exited when given the signal. I think beginning with the A8 core it would attempt to write out a checkpoint that would be newer than the default one done every 15 minutes. It might have been earlier with the A7 core which Joseph was tasked with writing to be an update from the A4 core.