Invalid Results on hxzkv38... |
Message boards : Number crunching : Invalid Results on hxzkv38...
Author | Message |
---|---|
Matthias Lehmkuhl Send message Joined: 18 Sep 07 Posts: 6 Credit: 1,270,364 RAC: 0 |
Have some invalid results over the last days on different machines Linux and Windows. userid=485 On the same computers are also valid results for hxzkv38... in the period where the results are shown. Matthias |
harpo_the_whale Send message Joined: 31 Jul 08 Posts: 1 Credit: 297,283 RAC: 0 |
I've also had a pile... same work units. I've resorted to turning off new work units for Enigma for the next few weeks until this is sorted out. There's no sense wasting my processing power. Harpo |
Matthias Lehmkuhl Send message Joined: 18 Sep 07 Posts: 6 Credit: 1,270,364 RAC: 0 |
Found a Message from TJM, http://www.enigmaathome.net/forum_thread.php?id=2&nowrap=true#1807 For me it looks good, the new validator is working. Had no invalid results in my list since then. Matthias |
TJM Project administrator Project developer Project scientist Send message Joined: 25 Aug 07 Posts: 843 Credit: 267,994,998 RAC: 0 |
The new validator is not finished yet, I have to use default credit system (called "new credit" on BOINC wiki) which some of you may not like. The old system simply doesn't work, it didn't work at all few days ago and after a patch it tries to read values from the database, but it fails and returns 0. There's a new patch already which I'll be trying today, but I'll be experimenting with the new credit system anyway, it reduces amount of work needed to add new workunit types by a lot. M4 Project homepage M4 Project wiki |
Message boards :
Number crunching :
Invalid Results on hxzkv38...