Message boards : Number crunching : New Machine - Strange errors!
Author | Message |
---|---|
Bob Browett Send message Joined: 14 Dec 05 Posts: 11 Credit: 2,275,743 RAC: 0 |
Hi I have built myself a new Quad Core machine with 4GB of memory, huge HD space and a Radeon 4870. It runs XP (SP3). On my old P4 boxes Rosetta was chugging away well, with no errors. On my new machine I have had quite a few Computational error glitches (see Task ID 196046667 as an example). Can someone tell me if I have built my machine wrongly of have I just stumbled over a Rosetta problem I have not seen before Best Regards Bob |
Greg_BE Send message Joined: 30 May 06 Posts: 5691 Credit: 5,859,226 RAC: 0 |
your machine with 81 tasks and 4 computation errors is running ok. these are just problems that crop up from time to time. Be sure to post your errors in the correct program threads. the team needs to see the errors. Just post a link and summarise the error, like the access error XXXXX code line. I saw you have 2 or more good results. so I would say your machine is ok. Hi |
Evan Send message Joined: 23 Dec 05 Posts: 268 Credit: 402,585 RAC: 0 |
If you have a quad does it run Rosetta faster if you run only one work unit rather than 4 simultaneously. By faster I mean more results per work unit. |
Mod.Sense Volunteer moderator Send message Joined: 22 Aug 06 Posts: 4018 Credit: 0 RAC: 0 |
If you have a quad does it run Rosetta faster if you run only one work unit rather than 4 simultaneously. By faster I mean more results per work unit. No. The case where that might occur would be with a hyperthreaded vs. not hyperthreaded CPU, because both threads must use the same math coprocessor. Rosetta Moderator: Mod.Sense |
Bob Browett Send message Joined: 14 Dec 05 Posts: 11 Credit: 2,275,743 RAC: 0 |
Thanks for the info Much releived! |
Message boards :
Number crunching :
New Machine - Strange errors!
©2024 University of Washington
https://www.bakerlab.org