Message boards : Number crunching : Minirosetta credits
Author | Message |
---|---|
Path7 Send message Joined: 11 Feb 08 Posts: 56 Credit: 4,974 RAC: 0 |
Hello all, Just a newbie at Ralph@home, I have done 8 Wu's so far. These minirosetta's ran fine, but generated relatively few credits. It's not I am concerned about my own credits that much (after all this is alpha). But people at Rosetta@home might be disappointed. Last 8 minirosetta 1.07 average: 5.37 points / hour. Last 8 Rosetta beta 5.93 average: 9.39 points / hour. (From rosetta@home) Since I saw the same lower points while crunching minirosetta at rosetta@home / longer runtime I am expecting the points are not lower due to the lower preference runtime at Ralph. I'm running Windows XP – home on a one-core AMD sempron 3000+ 1.8 G Hz, 1.4 GB RAM. Somebody else having the same credit experience? Have a nice day, Path7. |
Luuklag Send message Joined: 5 Jan 08 Posts: 15 Credit: 80 RAC: 0 |
Hello all, well i dont know if the creditsystem of ralphy is the same as that of rosy. but in rosy you get the credits for the average of claimed credit so far. so everyone should get the same credit, and if it really takes so long to crunch it, the claimed credit will rise when running on rosy |
feet1st Send message Joined: 7 Mar 06 Posts: 313 Credit: 116,623 RAC: 0 |
...everyone should get the same credit... I wanted to clarify that statement a little. Credit is issued per model completed, based on the average of work reported back to date when you report yours back. So, the average varies very little, and a given type of task should grant the same credit per model. But different systems will take different amounts of time to complete each model. And so credit/hour varies depending on how well suited your processor is to the capabilities required to complete the models of the task. Path7's credit looks like it suddenly hit a point where they are granted about half of what would be normal. kkupsch reports just 10% of normal. So something is definately wrong there. I see similar credit issue on Rosetta now as well, but previously it was normal. I've EMailed Ingemar & DK to see if they can locate the cause. |
Message boards :
Number crunching :
Minirosetta credits
©2024 University of Washington
http://www.bakerlab.org