No work unit errors since installing 5.2.6

Author	Message
Nova_nz_w Send message Joined: 10 Oct 05 Posts: 3 Credit: 237,533 RAC: 0	Message 2097 - Posted: 2 Nov 2005, 20:48:48 UTC Previously on a Pentium M Laptop (Dothan core) I would always get an "incorrect function" error on the "random" work units (the test units completed fine). I tried Boinc 4.19, 4.45, 5.2.5 and still had errors. After installing Boinc 5.2.6 I have completed around 40 random units without fault. Hurrah! ID: 2097 · Rating: 0 · rate: / Reply Quote

David E K Volunteer moderator Project administrator Project developer Project scientist Send message Joined: 1 Jul 05 Posts: 1018 Credit: 4,334,829 RAC: 0	Message 2098 - Posted: 2 Nov 2005, 20:50:50 UTC Thats great! ID: 2098 · Rating: 0 · rate: / Reply Quote

Andrew Send message Joined: 19 Sep 05 Posts: 162 Credit: 105,512 RAC: 0	Message 2149 - Posted: 3 Nov 2005, 13:37:12 UTC Last modified: 3 Nov 2005, 13:38:15 UTC That's weird, because according to Rom, 5.2.6 release is just localization updates, and one change for *nix style systems using /dev/random for the random number generator. And since Nova's computers are all Window machines, there shouldn't be any difference between 5.2.5 and 5.2.6. (The localization update only affects the boinc manager if I'm remember correctly.) Maybe it would be worth while doing a diff of 5.2.6 code changes to 5.2.5 code and see what actually changed in the boinc client. That might show what change actually fixed the problem... ID: 2149 · Rating: 0 · rate: / Reply Quote

Nova_nz_w Send message Joined: 10 Oct 05 Posts: 3 Credit: 237,533 RAC: 0	Message 2185 - Posted: 3 Nov 2005, 20:10:06 UTC Oooh that is weird! To throw another spanner in the works, my boxes at home (AMD Athlon, and Intel Celeron) have both had no problems with any of the random units. They are running TMR's optimized boinc clients. ID: 2185 · Rating: 0 · rate: / Reply Quote