Posts by 6dj72cn8

21) Message boards : Number crunching : App v0.10 (Message 8645) Posted 19 Jan 2009 by 6dj72cn8 Post: OS X still takes about twice as long on Intel Mac as it did before. Then it must depend which Intel Mac you have. 0.7, 0.8, 0.9 and 0.10 have all been pretty much the same speed for me.
22) Message boards : Number crunching : Vote for the MilkyWay favicon! (Message 8338) Posted 15 Jan 2009 by 6dj72cn8 Post: No. 7
23) Message boards : Number crunching : All downloads failing on test work (Message 6765) Posted 26 Nov 2008 by 6dj72cn8 Post: Failed downloads isn't only on the new test work. I've just had a failed download on a 633 task.
24) Message boards : Number crunching : more wus generated: nm_test1 (Message 6404) Posted 22 Nov 2008 by 6dj72cn8 Post: An apparently successful run of my first nm_test2 on Mac Intel. (Core 2 duo, Mac OS 10.5.5, BOINC 6.2.18) The only cloud on the horizon is this, which might or might not have some significance: <stderr_txt> APP: error reading checkpoint (opening file) APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 APP: error writing checkpoint (flushing checkpoint file) 0 called boinc_finish </stderr_txt>
25) Message boards : Number crunching : New App status (Message 6039) Posted 8 Nov 2008 by 6dj72cn8 Post: But the new improved workunits need to have run times of 1 hour to 2 hours so the 20 work unit cache per CPU means something. You seem to be forgetting that only a thousand or so people are running Milksop's app. For the rest of us (i.e. Mac users and people who never read the boards) the current tasks still run in the four to nine hour range. Do you think we currently want tasks made twenty times longer? I am sure the crunch time per task will be increased when the new faster app eventually becomes the stock app and is available to all.
26) Message boards : Number crunching : And if we DONT run a "FASTER" App.? (Message 5908) Posted 4 Nov 2008 by 6dj72cn8 Post: Also, regarding the Mac version, we're raising the credits back up because we think it's unfair to those that don't use the optimized app and either lower the per hour limit or keeping it the same (that's travis's deal). Without buying in to the rights and wrongs of the current situation, I merely want to say thanks to Dave and Travis for putting the credit back up so that those of us without access to the optimised app still have the option of continuing on as before. Even if it does take us hours to do what others can do in minutes. That said, I do hope a new stock app, equally speedy for all platforms, is released soon.
27) Message boards : Number crunching : And if we DONT run a "FASTER" App.? (Message 5883) Posted 4 Nov 2008 by 6dj72cn8 Post: The optimized app is freely available. It's not fair to the general boinc community if the optimized app is so much faster that those users will get so much credit. If you feel this strongly, why not just get the optimized app? Because there isn't a Mac version.
28) Message boards : Number crunching : Credit limits apparently exist ... (Message 5836) Posted 3 Nov 2008 by 6dj72cn8 Post: Windows/Linux: 215 credits per hour (max). Mac (Core 2): 26 credits per hour. Hey Travis/Dave, where's that new app? The playing field isn't very level here! Who needs to call in the Kredit Kops for interproject parity when there's this much disparity within a project? Mutter, grumble, #$%& Milkslop, mutter . . .
29) Message boards : Number crunching : Faster application (links inside) (Message 5737) Posted 1 Nov 2008 by 6dj72cn8 Post: It doesn't really change the credits too much if that is why you are conserened. See post 5686 in this thread. 216 credits per hour. I call that a change even if you don't. But more importantly, why would I waste four hours worth of electricity for a calculation that others can do in three to twelve minutes? It was different when only one or two people had the speedy app but when it is generally distributed then those without it might as well not bother. We'll see what next week brings.
30) Message boards : Number crunching : Faster application (links inside) (Message 5729) Posted 1 Nov 2008 by 6dj72cn8 Post: Well I'll say that the new app is fast. Down from 11 hours for the long units to a little under 9 minutes for the same units. Massive speed up. A little under nine minutes is it? But still four hours on my Mac for which no new app is available. Not much point in my continuing to crunch here is there? Was pushing some participants away from this project part of the aim of releasing the new app?
31) Message boards : Number crunching : Short deadlines-High priority-suspend Milkyway (Message 5212) Posted 12 Sep 2008 by 6dj72cn8 Post: It finished in 6.5 hours in the end, it was a 372... Might be more hopeful though there's also several 373s Time will tell. The initial estimate did through it off by a lot though. On your computer, if a 372 takes 6.5 hours then I estimate a 373 will take 3.5 hours and a 371 will take just under 7 minutes.
32) Message boards : Number crunching : New OSX App (Message 3033) Posted 5 Apr 2008 by 6dj72cn8 Post: Runs fine on a G5. Crunch time reduced to 71% of previous app. Thanks Dave. Whatever you did, keep doing it!
33) Message boards : Number crunching : Macintel and OS X performance issue (Message 2991) Posted 4 Apr 2008 by 6dj72cn8 Post: in the last version of the application we did a lot of optimizations to the non-intel macs, You did? On my G5 PPC, app 1.21 is taking exactly the same time per task as app 1.13 did.
34) Message boards : Number crunching : updated granted credit (Message 2634) Posted 24 Mar 2008 by 6dj72cn8 Post: I yelped a bit a couple of weeks ago when it was mooted that credits should be reduced. That is because I am running a PPC Mac and it didn't get any optimisation in the last round of app updates. However, I subsequently looked at the figures again and I concede that 6.5 per result was too high. Trouble is, 2.17 is way too low. Somewhere around 4 would be reasonable I would think.
35) Message boards : Number crunching : credit issues (Message 2015) Posted 7 Mar 2008 by 6dj72cn8 Post: At this point if the optimizations are done it looks like somewhere between 3-4 credits would be about right. But you are basing that on crunch times having been reduced. What about those of us on platforms for which crunch time is unchanged from app 1.13?
36) Message boards : Number crunching : application v1.21/v1.22 errors/memory leaks/crashes here (Message 2006) Posted 7 Mar 2008 by 6dj72cn8 Post: Running OK on Mac PPC G5 OSX. The crunch times are unchanged. The first 55% (approx) of the task is pretty slow but it races through last 45%. Given what I have been reading about how much faster the Windows versions now are, is there any way you can optimize the Mac app any further, particularly the first part? Thanks.
37) Message boards : Number crunching : Please post app 1.17/1.18/1.19 memory leaks/errors (Message 1875) Posted 5 Mar 2008 by 6dj72cn8 Post: 1.18 ppc-darwin still as broken as 1.14 and 1.17.
38) Message boards : Number crunching : App 1.17 (Message 1846) Posted 5 Mar 2008 by 6dj72cn8 Post: Still fails on PPC Mac. \|Task gs_282_1204736846_14850_0 exited with zero status but no 'finished' file "Process creation (../../projects/milkyway.cs.rpi.edu_milkyway/astronomy_1.17_powerpc-apple-darwin) failed: errno=8"
39) Message boards : Number crunching : App 1.14 (Message 1812) Posted 4 Mar 2008 by 6dj72cn8 Post: App 1.14 just installed itself on Mac PPC G5, OSX 10.4.11 and is immediately broken. Pity. 1.13 was running well. Result "Task gs_280_1204617434_197_0 exited with zero status but no 'finished' file" Edit: Just seen the news on the front page. Thanks.
40) Message boards : Number crunching : How long does it take for you? (Message 1403) Posted 9 Jan 2008 by 6dj72cn8 Post: BTW, did you know that you get slightly better credit for the odd number WUs (the long ones) with a ratio of approx. 175 Sec per CS compared to approx. 187 Sec per CS for the short ones Not here. My three-year-old Mac will do a long one in just over 18 mins and a short one in just over 4 mins. Expressed by your method, that's 167 sec/credit for long and 126 sec/credit for short. So you take the long ones, Crystallize, and I'll take the short ones. We'll both come out in front.

Previous 20 · Next 20