| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Help Cure Muscular Dystrophy - Phase 2 Forum
Thread: If pairing doesn't really work... |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 12
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The discussed "let-it-run" option would be a reasonable idea. More than reasonable. The 6-12 hour limit is, for many/most machines, completely unnecessary. Machines that are on most of the time don't care whether they have done 1 or 2 or 4 or 10 WUs/day. More importantly, with whatever was changed in Jul/Aug, around 90% of pre-P4 Intel results in this science are marked invalid anyhow if they don't complete. That's months of CPU time being thrown away for no reason, when it would all be useful if they were simply allowed to complete. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Btw. I have some computers without internet access. So I can fetch work on another computer, put the boinc data dir on USB stick and let it run on that computer. Every ten days I have to exchange the data and transfer the results to WCG or other projects.
So optimization is nice but there will always be some losses! E.g. those standalone computers are unknown to any boinc project and therefore cannot be classified at all. But even with an inadequate speed class they get more work done as if they would run idle... |
||
|
|
|