Project: 5506 and ver 1.19 EUE

[shdbcamping]

*I have had instances of 1.19 version core failing for nans ending in EUE client shutdowns, also at the beginning of 5506 Project WU's. This happens after Starting GUI and before completing 1%. Originally, I would got to the GPU folders... delete the usual files, folders etc and restart the client until it picked up a working Project.

* The other morning 2 clients went down overnight while I was at work (both 5506). Instead of deleting the stuff in the GPU folders and retrying for work, I simply tried copying the older 1.13 version of core_11 over the 1.19 that the client was using. That was the only change to my system. I started Both clients. Both clients picked the work up and started crunching without a hitch, finished and sent also without a problem or error. Is this action to dealing with the EUE's going to cause problems for the integrity of the work that is sent?

* I do not know what this means because I don't have the core coding. I just wanted to let you know of what transpired. It is odd that only the change in core version would make a failed project work and send error free.

*If there are any questions or more info needed by PG on the issue, please EMail me as I'm just trying to point out what I've seen as an anomally. This 1.19 nans EUE happens with a number of 5506 WU's.

[Iannis]

Hi

Go back to 1.15

[shdbcamping]

Hi

Go back to 1.15

Thanks, but it'll DL the 1.19 core every it tries to assign me the 5748 WU. I'm not sure if that's the only 5700 WU that does that... Just all I notice.

Also, I want to make sure that "backing up" the Core version won't impact the integrity of the Projects. That's really the only reason I started Folding in the first place. I can deal with Hotter and slower and EUE's as annoying as they are to me. I just want to make sure that what I'm sending is both useable and accurate for science.

[7im]

By the number of new topics with "Project: 5506" in the title, I'd say there are an above average number of EUEs with this project. That happens from time to time. Doesn't mean it's a bad project, just pusing the limits of science.

Vijay is aware of the problem: http://foldingforum.org/viewtopic.php?f ... 325#p69325

[shdbcamping]

By the number of new topics with "Project: 5506" in the title, I'd say there are an above average number of EUEs with this project. That happens from time to time. Doesn't mean it's a bad project, just pusing the limits of science.

Vijay is aware of the problem: http://foldingforum.org/viewtopic.php?f ... 325#p69325

Thanks for the update. I'm more worried that what I'm doing may corrupt the results of WU's that I'm doing this with. I just want to make sure I'm not doing something that may hurt data and that Bad results don't get through any cracks.