Posts by Sailor

21) Message boards : Number crunching : I've had enough also (Message 5375) Posted 8 Oct 2008 by Sailor Post: Someone else said that this is a volunteer effort. Yes, perhaps Dave and Travis should learn this on their own, but if they are to learn it on their own, then the people who are more knowledgable should stop poking at them with the "we're smarter than you" type of comments. I agree here, what I (and I think many others) would like to see, is a constructive enviroment where the project devs and the guys that have the knowledge and time to optimize the code work together and get this problem over. About the cheating/overclocking/high credit thing: I disagree. There are people out running the WUs 10x..50x times faster then a normal CPU from the same type. You think you can gain a 50x speedup with overclocking? not even close to 10x. And it definetly spoils the fun for a lot of guys who like to check their credits and positions in the stats. This should be stopped as soon as possible, no matter where we are going with a new application. Personally, It doesnt really bother me, but I can understand why there are ppl out that that are offended by that, and I think they have a good reason to be. Also, up to here, we havent any confirmation if those homebrew apps return valid results, if not, it would ofc disqualify them straight away. Point is: I dont see where the problem right now is? There is a lot of talks going on from both sides for a long time now, but where does it hang? Are those guys with their fast applications even interested in making their apps public (and by that loosing their credithorses)? Is the project interested in using this code? I think ive read in some older thread, that the focus is not on speeding up the current app, cuz there is a new on the way anyway? (but dont quote me on that, might remember wrong). Maybe if both sides could just simply state what they want and would be willing to do (and without stupid "A 2 year old child could do this and that"), we might have something to start, or at least know where we stand and what to expect.
22) Message boards : Number crunching : I've had enough also (Message 5360) Posted 8 Oct 2008 by Sailor Post: The server and network infrastructure need to be able to handle the additional load. This is not me trying to make excuses, it is just reality. If the load causes the server to crash and/or the work generator can't keep up and/or people use up all of their daily quota, there'd just be more bellyaching. It needs to be done right, but not necessarily "right now"... What would be so bad if ppl reach daily quota? Its not like MW is the only project out there. About the server load, youve got a point there, still its not an argument to not make the application more effective. When we only had the very short WUs, things were messy, agree. If a better code would run 10x faster (or even more), we would be in server trouble again, yes. But keep in mind, that we would do science also 10 times faster ! There are workarounds, like dont create so much work (might not be the best solutiuon for us crunchers) or simply lower the credits given, so MW isnt on top of the BOINC projects anymore, that would drive some ppl off that are only here for the credits. Before someone brings up the point, that by this MW would decrease their computing power - this aint true when u keep in mind, that a new application would be 10x faster (i keep using 10x faster, I dont know how much it could be tho, just a guess when looking at some of the guys profiles running self coded apps).
23) Message boards : Number crunching : I've had enough also (Message 5345) Posted 7 Oct 2008 by Sailor Post: This is madness... If those xx times faster apps are producing fine results, what shouldnt be so hard to check (project staff anyone??) then get this code and make it the default application - fast! We are waisting time & science every second. And no, I dont want to make 1 million credits a day, if that may sound so, if the new app takes only 10% of the actual time and i get 10% of the actual credit - perfect. So come on do something about it or ill leave be the weekend ( not that it matters anyway with people like Milksop at try throwing out the WUs like waterdrops down the Niagarafalls :P )
24) Message boards : Number crunching : Checkpointing? (Message 5076) Posted 28 Aug 2008 by Sailor Post: Allright, thx for the info :)
25) Message boards : Number crunching : Checkpointing? (Message 5070) Posted 27 Aug 2008 by Sailor Post: Up to here I thought, MW units dont checkpoint. Today I got forced to reboot, and it looked like, the WU continued from the point where I suspended it before the reboot. Confirmation?
26) Message boards : Number crunching : WU Credits (Message 4871) Posted 21 Aug 2008 by Sailor Post: Uff, I really think all this talk about credits etc should come to an end here now, so that the project admins can return to the more important things around MW, and not having to state the same facts over and over again :/ The optimizations Crunch3r did are really for a specific processor, it's not like we can just take his code and it'll work anywhere. So for the time being, either everyone gets their credit reduced so that the project isn't outputting way too much credit, or we have a speed limit. If everyone was using an optimized app, the amount of credit per WU would be reduced anyways, to keep it in line with what other projects are awarding. This should say it all, if the optimized application would work for everyone this well, it should have been made public (if it returns usable data for the project) and ofc the credits would have shriked then by a LOT. Some guys really expect to get 260 credits for some minutes of work it looks like :S Long story short: I think this issue was handled very well, thumbs up Travis!
27) Message boards : Number crunching : Increased WU Credit (Message 2814) Posted 25 Mar 2008 by Sailor Post: Way too much- Sad that the credit sharks have taken over the voice here, look on the sign up dates from the biggest complainers and wonder: they all came after the Application improvement made in early march. After ive forced myself reading through that first, now locked credit discussion thread (If I wouldnt had to run 2 long mysql backups right now, i probably would have never going through all that stupidity in that thread. ) - Im shocked on what kind of people the high credits have attracted now, some even use abusive language to get what they want. /signing off
28) Message boards : Number crunching : Poll: How much credit do you think is fair? (Message 2693) Posted 24 Mar 2008 by Sailor Post: 3.25
29) Message boards : Number crunching : 20 workunit limit (Message 2349) Posted 18 Mar 2008 by Sailor Post: If these WUs are time sensitive, then why do they have a 5 day deadline? The server has been out of work for nearly 24 hours now, and there are still over 5000 WUs out in the wind. Seems to me that if new work generation depends on the old results, the deadline should be shortened. Maybe try 12 hours and see how that goes, at least if the WU ends up on a duffer, it would time out and possibly get sent to a faster, more reliable host. Fish I couldnt agree more - nothing that bothers me more then when I have to read that im computing for the trash. If the project needs a shoter deadline then thats the way it has to be, I think most will agree here.
30) Message boards : Number crunching : credit issues (Message 1970) Posted 6 Mar 2008 by Sailor Post: Was just running a new 1.19 wu and after 6 mins it hardly had 10% done. So it would approx needed ~60 mins to complete. I suspended my CPDN model running on the other core and the Milkyway WU finished the last 90% within 2 mins. CPU usage was normal @ 100% with 50/50 on both Apps. CPU is AMD x2 4200+ @ stock Boinc version number 5.10.20 Win XP SP 2 nvm this post, its a problm in the progress bar not showing the real progress. Must have hit the suspend button on the CPDN model around the moment when it starts to catch up on % :) The bar goes within 2 mins then from 10% to 100% in big steps
31) Message boards : Number crunching : credit issues (Message 1965) Posted 5 Mar 2008 by Sailor Post: Was just running a new 1.19 wu and after 6 mins it hardly had 10% done. So it would approx needed ~60 mins to complete. I suspended my CPDN model running on the other core and the Milkyway WU finished the last 90% within 2 mins. CPU usage was normal @ 100% with 50/50 on both Apps. CPU is AMD x2 4200+ @ stock Boinc version number 5.10.20 Win XP SP 2

Previous 20