| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Locked Total posts in this thread: 7
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
OK, this is something weird:
faah0837_ bdb379_ mx1sdu_ dry_ 01 Valid 10/26/2006 09:44:55 10/26/2006 19:55:52 6.64 51 / 233 aah0837_ bdb379_ mx1sdu_ dry_ 01 Valid 10/19/2006 00:01:25 10/24/2006 17:25:21 3.44 247 / 233 faah0837_ bdb379_ mx1sdu_ dry_ 01 Valid 10/18/2006 23:53:53 10/20/2006 20:12:43 9.09 233 / 233 I have never seen such a difference between claimed and granted points. Anyone got an explanation? |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Well, Toly, you have run into a ticklish situation. There does not seem to be a way to accurately and authoritatively score the points awarded a work unit unless: A) everybody uses the same compiler and OS (and type of CPU) or B) an enormous amount of floating point tabulation code is placed inside each project application (with lots of possibility for bugs). So BOINC, responding to criticism / pleas has added a facility for adjusting scores requested for a particular machine. Some people use it, most don't. Two of the people on your quorum seem to be using the same modified adjustment ratio. In case you have not been following the occasional flame wars on this subject, the best advice is to avoid Lawrence |
||
|
|
olympic
Senior Cruncher Joined: Jun 12, 2005 Post Count: 156 Status: Offline |
I've had that happen to me once or twice. The other 2 results were crunched by computers using an optimized BOINC client. These clients do have a purpose on other DC projects, but here on WCG, they just serve to claim many times the points that the stock BOINC client would claim. 99.9% of the time there is only 1 optimized client in a quorum, so the high result gets rejected. But in the other 0.1%, 2 or 3 optimized clients get sent the same work unit and the result is that all 3 machines get awarded the claim of one of the optimized clients.
----------------------------------------![]() |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I see what you guys saying now. Basically it is artificialy adjusted numbers not representative of reality. Well, Personally I couldn't care less about points since that is not the reason why I am crunching. I was just wondering about disperity as I thought there was other reason for it. But since it is just about kids bragging I am cool with it. ;)
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
See what I meant by ?SIMAP ( http://boinc.bio.wzw.tum.de/boincsimap/ ) is a good project that I push for people with limited memory. I am uncertain of the commo requirements for people on dialup, but SIMAP offers a number of versions. However: Project status The project has currently no work (see below). The similarity matrix of the SIMAP project is currently up to date. The calculation of new simap workunits containing the novel proteins from october 2006 will start around november 1st. The initial calculation of protein domains has been finished. The calculation of updates (new sequences and new domains from InterPro 13.0) will start around november 15th. October 12, 2006 BOINCSIMAP turns into periodic project: As to be seen in the project stats and batch monitor, the work for simap and hmmer is limited to few remaining batches. The hmmer calculations went much faster than expected due to the performance optimizations of this application. Thus our databases for protein similarities and domains will soon be up-to-date. We are celebrating this very important milestone and would like to thank all the BOINCSIMAP users that made this happen. Now we will focus on the integration of SIMAP with more bioinformatics methods, portals and databases - thus increasing the impact of out project on computational biology and biosciences. In the future SIMAP will need your contribution by crunching regular updates. Every month updated sequences and domains will be written into simap and hmmer workunits that will need to be calculated. We would be happy if you would stay connected with our project by crunching these periodic batches of work. http://distributedcomputing.info/projects.html is a good site to check for current Distributed Computing projects. Lawrence |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Yes, SIMAP is out of work. The reason is because they mean business and they know where to put their talent. They optimised their science apps and reduced the quorum to 2 in order to reduce waste and show some respect for the crunchers who donate their resources. The result of the optimisations and quorum reduction was that they finished their job 10 months earlier than expected. Too bad WCG isn't of the same caliber.
SIMAP will have more work in the future. But this isn't a post to promote SIMAP, this is a post to shake the cob webs and complacency out of WCG, a project I used to believe in and spent a lot of time promoting. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Just to add my little 2 bits about "points":
1. I don't compute here for "points". 2. I disliked spending time computing points for points. 3. My interest and presumed value here is in completed "results". 4. I would like to see points just ... go away ... !! 5. I fail to: see the value/understand the need/or desire for points. 6. Points are useless, nonproductive, a waste of computation time. Points please go away .... don't get larger .... don't get mad .... just ....go away. PTabolinsky |
||
|
|
|