Posts by MB Atlanos

1) Message boards : Number crunching : N-Body long processing time (Message 62557) Posted 13 Oct 2014 by MB Atlanos Post: It wasn't running all the time in the beginning to keep the Mac cool, I was fooled by the wrong estimated time. Only the last few days it got full time with raised fanspeed and is catching slowly up. Between, this is a Core2Duo dualcore with 2.66 Ghz, the real CPU time ist now at 253 hours - Oh, this will certainly be the last nbody for me. ;)
2) Message boards : Number crunching : N-Body long processing time (Message 62549) Posted 12 Oct 2014 by MB Atlanos Post: Could someone from the staff please extent the deadline for this WU?: ps_nbody_08_05_orphan_sim_3_1411504411_51255 http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=624878750 Or mark this WU for no resends for the next 2 days, i dont want this monster to be serveraborted. It is at 77% after 130.5 hours of computation and needs additionally 40-50 hours.
3) Message boards : News : New Website Coming Soon (Message 40737) Posted 29 Jun 2010 by MB Atlanos Post: Where are the the link to the list of applications? A Link to the old website would be usefull.
4) Message boards : News : new osx applications (Message 37964) Posted 2 Apr 2010 by MB Atlanos Post: Any updates fÃ¼r the PPC application? Its pretty outdated and there are no highoptimized versions available.
5) Message boards : Number crunching : Optimized OS X Applications (Message 14240) Posted 7 Mar 2009 by MB Atlanos Post: jup, opti apps for PPC G4 and G3 too would be very nice, now with the double workload WUs around. Thanks for your work.
6) Message boards : Number crunching : v0.18/v0.19 issues here (Message 11272) Posted 17 Feb 2009 by MB Atlanos Post: The x87 and SSE versions are created using different compilers. So it is perfectly possible that the x87 version is faster than the SSE one, as SSE cannot be used for the time consuming stuff here at MW requiring double precision. Also be aware there are WUs with different length floating around. One should compare only the times of similar ones. Ah that would explain it, thanks. I am not 100 percent sure, but it was the same mix of searches s86 and s79 before and after the switch yesterday. No s20/21 with longer runtimes, the are downloaded today.
7) Message boards : Number crunching : v0.18/v0.19 issues here (Message 11237) Posted 17 Feb 2009 by MB Atlanos Post: On a P3 1233 Mhz with Win2000 the new stock app v0.19 with 55-57 min is faster than the previous optimized win32 app v0.16SSE with 1:15h. Nice work Travis - Thanks ;) As stated before, there are no improvements in the codebase of the relabeled v0.19 opti app beside the additional logging. So SSE optimized app v0.16 and v0.19 for windows should give the same runtimes. Does this mean the stockapp is better for old PCs with only SSE?
8) Message boards : Number crunching : v0.18/v0.19 issues here (Message 11032) Posted 16 Feb 2009 by MB Atlanos Post: The new v0.18 stock app is also significant faster at PPC Macs to: Mac mini G4 1,5 GHz are now at 1:15h from 2:15h in the past. iMac G3 350 MHz now 4:20h vs. 8:15h before - very nice ;)
9) Message boards : Number crunching : app v12 (Message 9082) Posted 25 Jan 2009 by MB Atlanos Post: The first line of Intel Macs from 2006 used the Core Solo/Duo with: All models support: MMX, SSE, SSE2, SSE3, EIST, XD bit Only the following Macs with Core 2 Solo/Duo has SSSE3 and Intel 64. Thanks to the stupid labeling by intel, its easy to confuse.
10) Message boards : Number crunching : post milkyway_powerpc-apple-darwin problems here (Message 8892) Posted 23 Jan 2009 by MB Atlanos Post: Not really a problem, but an observation. On my G3 app v0.10 was the fastest version (10 minutes less), v0.12 and v0.07 gives the roughly same times. All versions are granted with the same credits of 39.xx. On my G4 v0.12 (8150 sec) are a little slower than v0.09 or v0.10 (ca 7900), but are allways granted with 41.xx gredits. v0.07 also gets 39.xx on this Mac.
11) Message boards : Number crunching : app v0.9 (Message 8555) Posted 18 Jan 2009 by MB Atlanos Post: no difference in runtime between v0.7 and v0.9 for PPC on a G4 Mac mini: http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=1905
12) Message boards : Number crunching : post milkyway_powerpc-apple-darwin problems here (Message 6568) Posted 24 Nov 2008 by MB Atlanos Post: Hmm, i changed only the settings to allow test-WUs and let Boinc request for work. First download was a test-WU. Try to reset the project in the BOINC client, but this will delete all work you have received. Maybe there are no test-WU ready to send at this time, i also get only work for 1.22 @Travis Second test WU with app 0.01 does not surive a restart of the BOINC-Client - it quit a few seconds after restart with the "Outputfile absent" error. Oddly no automatic download of app 0.02 happend, only the test5 WU was downloaded. Looks like this might be a checkpointing problem. Did you have a WU for the new app running when you did the restart? Yes it was an test5 WU.
13) Message boards : Number crunching : post milkyway_powerpc-apple-darwin problems here (Message 6465) Posted 23 Nov 2008 by MB Atlanos Post: Hmm, i changed only the settings to allow test-WUs and let Boinc request for work. First download was a test-WU. Try to reset the project in the BOINC client, but this will delete all work you have received. Maybe there are no test-WU ready to send at this time, i also get only work for 1.22 @Travis Second test WU with app 0.01 does not surive a restart of the BOINC-Client - it quit a few seconds after restart with the "Outputfile absent" error. Oddly no automatic download of app 0.02 happend, only the test5 WU was downloaded.
14) Message boards : Number crunching : post milkyway_powerpc-apple-darwin problems here (Message 6456) Posted 22 Nov 2008 by MB Atlanos Post: Dont think the error is PPC related, but it happens on a G3-350 MHz iMac running Mac OS 10.3.9. App was Version 0.01. BOINC was limitet to use only 128 MB RAM at this time, i upped it to 192 MB. Whats the limit of RAM-usage for this type of WUs? #### Sat 22 Nov 12:09:08 2008\|Milkyway@home\|Starting task nm_test3_5_1227379417_0 using milkyway version 1 Sat 22 Nov 18:08:56 2008\|Milkyway@home\|Aborting task nm_test3_5_1227379417_0: exceeded memory limit 138.19MB > 128.03MB Sat 22 Nov 18:08:57 2008\|Milkyway@home\|Computation for task nm_test3_5_1227379417_0 finished Sat 22 Nov 18:08:57 2008\|Milkyway@home\|Output file nm_test3_5_1227379417_0_0 for task nm_test3_5_1227379417_0 absent
15) Message boards : Application Code Discussion : The New Application Optimizations (Message 6299) Posted 19 Nov 2008 by MB Atlanos Post: Thanks for the link John Clark. great news. So hopefully my old iMac G3-350 can meet the deadlines again. ;) Last time it takes over 66 hours for one WU.
16) Message boards : Application Code Discussion : The New Application Optimizations (Message 6281) Posted 19 Nov 2008 by MB Atlanos Post: Don't forget the Mac's out there with no access to the vastly faster milksoup app. Even if the new official app are only 2-3 times faster than the old official it would be an significant improvment, esspecially for older G4 or G3 PPC Macs.
17) Message boards : Number crunching : 20 workunit limit (Message 2281) Posted 16 Mar 2008 by MB Atlanos Post: It would be fine to have 5-10 wu's at a time when the time is extended by say 5-10 times. They would run 55-100 min on mine then. Or it would be 125-250 min (Coppermine-P3 and G4) up to 570-1140 min (G3) on not so recent computers.
18) Message boards : Number crunching : Cache Limit (Message 2112) Posted 8 Mar 2008 by MB Atlanos Post: something else i want to do is add a line search capability to the work units. with that in place the amount of time for a WU should increase in the range of 10x - 50x, which should keep almost everyone happy i'd hope :) Please not so much increase. Milkyway is one of the very few project, that run acceptable on older G3-Macs (ca 1,5 h per WU) or older comps in general.