[DPC]Stephan202 2003-04-24 13:24:42 | Version 4.31b saves the results of a quarantined simulation separately, so for example:3.100000, 3.200000 (200 Mpts) [v4.31b] {F242E1AE} <SolenoidsOnly> But, when the client is shut down and then restarted again, and when it then finishes the next run, all previous yields are deleted, while the nuber of runs is increased by one. So, let's say I run a simulation three times, and then, half way the 4th, I stop the client. When I restart it again, it continues. But when it finishes the 4th run, there's only one yield and a few Mpts is in queue.txt, while it says #runs=004;. Seems like a bug to me. --- Dutch Power Cow. MOOH! |
[DPC]DemerZel 2003-04-24 14:10:53 | Will this make the results for v4.31b invalid as it makes the results less acurate ? |
[DPC]WAU-Spons 2003-04-25 04:16:44 | i've had something similar with the 4.3 unix client, my que was emptied, while the counter was increased. this lead to a 4.3 instead of a 7.x result, so probably there are some false low results. |
Bewolf 2003-04-26 01:54:13 | I just had the exact same thing happen. 4 of 5 runs were completed and the queue file looked normal. I terminated the client, and restarted it. It resumed the 5th run from an autosave point. After completing, results.txt showed #runs=005; but only counted 121.9 Mpts (for a 5.59% result, should have been around 620Mpts for 5 runs). |
[AMD Users] Michal Hajicek 2003-04-26 12:20:20 | This is what I was talking about in the thread Autosaved-parameter-sets-rechecking (2nd bug). |
[DPC]inquestos 2003-05-04 08:49:10 | version 4.3 had the same bug... (I found that out this afternoon)... |