| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Chat Room
Thread: example of member who should not be in the top 5000 in results |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 4
|
|
| Author |
|
|
RicktheBrick
Senior Cruncher Joined: Sep 23, 2005 Post Count: 206 Status: Offline Project Badges: 
|
Here is the stats of a member in the top 5000 in results returned but is in last place in both cpu time and points.
Total Run Time (y:d:h:m:s) (Rank) 0:000:01:05:48 (#322,527) Points Generated (Rank) 0 (#323,353) Results Returned (Rank) 3,935 (#3,350) Avg. Run Time Per Calendar Day (y:d:h:m:s) 0:000:00:00:09 Avg. Run Time Per Result (y:d:h:m:s) 0:000:00:00:01 Avg. Points Per Hour of Run Time 0.00 Avg.Points Per Calendar Day 0.00 Avg. Points Per Result 0.00 Avg. Results Per Calendar Day 9.13 Last Result Returned (UTC) 02/06/2007 10:42:15 [5,583+ hour(s) ago] Device Installations 2 Registered Member Since 07/24/2006 3935 results with only 1 hour of cpu time and no points is an example of ud software fault. I am disappointed that someone who has put no effort into this project is ranked above me in results returned. I would think that anyone with zero points per result could be safely removed from the top 5000. There is probably about 500 members like this and is why my ranking in results returned is 500 more than cpu time and 1000 more than points. |
||
|
|
JmBoullier
Former Community Advisor Normandy - France Joined: Jan 26, 2007 Post Count: 3716 Status: Offline Project Badges: 
|
Rick,
----------------------------------------I think your assumption that there are so many cases of members returning null results in the top 5000 is wrong by much. First because members like that usually give up faster than the one you mention. But more because there is a perfectly valid reason to explain the relative discrepancy between your rankings. This reason was the "Genome Comparison" project where WUs were very short when compared to all other projects. People like you who probably haven't run many GC WUs have a "by results" ranking higher than their "by points" ranking. People like me who mostly ran GC WUs at this time have a "by results" ranking much better than the two others. Although we have about the same average daily production in the stats your average points per result is twice mine but my average results per day is twice yours. That's all. Note that with the multiplication of very fast machines a similar discrepancy is starting to develop between points and run time. People with the fastest machines tend to have a much worse "by time" ranking than their "by points" one. And after all, never forget that all this is only for fun, the important point is the help that we might bring to researchers. Cheers. Jean. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
It has been my observation that points are the best indication of work done and run time is the best indication of effort put in. Work units don't really indicate anything.
----------------------------------------(although, the run time/effort connection falls apart with multi-core machines) For example, a guy in my group has the same number of points as me, but twice as many work units and 6x the run time. [Edit 1 times, last edit by Former Member at Sep 27, 2007 6:49:55 AM] |
||
|
|
cargod01
Veteran Cruncher USA Joined: Apr 25, 2007 Post Count: 508 Status: Offline Project Badges: 
|
It has been my observation that points are the best indication of work done and run time is the best indication of effort put in. Work units don't really indicate anything. (although, the run time/effort connection falls apart with multi-core machines) For example, a guy in my group has the same number of points as me, but twice as many work units and 6x the run time. I have a fleet of P4-1.7Ghz machines that crunch 24/7. They return results slower than many newer/faster machines, but that is why they have separate catagories for run time, results returned and points. Everyone can make claims about what they do being 'more important' or more helpful that what someobne else does. The point is that you are contributing, not that you are a certain number in the rankings. Being competetative is OK, but no need to think it is the main reason we are here. There are those with higher rank than me, but I don't mind. The life they save may be mine...or yours! Avg. Run Time Per Calendar Day (y:d:h:m:s) 0:039:22:10:35 Avg. Run Time Per Result (y:d:h:m:s) 0:000:12:05:06 Avg. Points Per Hour of Run Time 30.57 Avg. Points Per Calendar Day 29,292.81 Avg. Points Per Result 369.46 Avg. Results Per Calendar Day 79.29     |
||
|
|
|