Posts by [BAT] tutta55

21) Message boards : Number crunching : Work Unit Credit? (Message 3674) Posted 6 Jun 2008 by [BAT] tutta55 Post: I just finished a work unit yesterday that took 15 hours. So far there is no credit. That's probably one of the bad batch discussed here http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=320 If you see any WU that runs noticeably longer than usual, abort it. These WU start with gs_3XXXXXX instead of the current gs_5XX
22) Message boards : Cafe MilkyWay : Nevermind (Message 3671) Posted 6 Jun 2008 by [BAT] tutta55 Post: Nevermind! I don't quite see how to delete this thread, but I made a Newbie mistake. Sorry Draco ATA members like newbie mistakes :D
23) Message boards : Number crunching : Best completion times (Message 3659) Posted 4 Jun 2008 by [BAT] tutta55 Post: I just found this wu that was completed last night on 1 of my rigs ... Is this the fastest so far ? I'm not sure http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=32260933 Intel 9450 @ 3.008Ghz ddr2 800 on kubuntu 64 Some tasks seem to take a bit shorter than others. Don't know why. On average your machine appears to do 160-165s/WU. Q9450 doing well ;)
24) Message boards : Cafe MilkyWay : Word Link (Message 3643) Posted 2 Jun 2008 by [BAT] tutta55 Post: aid
25) Message boards : Number crunching : How to get FreeBSD work? (Message 3464) Posted 7 May 2008 by [BAT] tutta55 Post: Hmm, I'm a bit overwhelmed by all the response I guess there's not many people here who run FreeBSD (I don't either). Have you looked at this page? There's quite a few OpenBSD, NetBSD and FreeBSD BOINC clients listed there. Yes, I looked there. i386 FreeBSD clients enough, but no amd64. OpenBSD and NetBSD binaries are afaik not compatible with FreeBSD. The only thing I did find was a package from FreeBSD. That lets you compile your own Boinc client. I tried that. Through package dependencies it also downloaded some widget libraries it had to compile. Just when I thought it might actually work... compiler errors.
26) Message boards : Number crunching : How to get FreeBSD work? (Message 3462) Posted 7 May 2008 by [BAT] tutta55 Post: Hmm, I'm a bit overwhelmed by all the response. :p Meanwhile I wanted to repeat the exercise with 64-bit FreeBSD. But I can't find a 64-bit FreeBSD Boinc client anywhere. Which makes me wonder. Since the project's applications page shows "FreeBSD running on amd64 CPU", who tested this version, and how?
27) Message boards : Number crunching : How to get FreeBSD work? (Message 3461) Posted 6 May 2008 by [BAT] tutta55 Post: I went in with a blunt axe :p Finally got it to work, but in an unorthodox way. I opened the crunch3r Boinc client in a binary editor and replaced the platform identification string i386-portbld-freebsd by i386-unknown-freebsd. And yes, it now downloads and runs. But the WU take 2 minutes longer than in Windows 32-bit. 12 minutes as opposed to 10. Bummer. Am I missing something? According to http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=297&nowrap=true#3388 the FreeBSD version would be by far the fastest. But it appears to be the slowest. Is there an unofficial optimized FreeBSD application? EDIT: I must add this for completeness, as it may influence performance: - I run DesktopBSD, and not FreeBSD. DesktopBSD is basically FreeBSD with the KDE GUI added to it (similar to Kubuntu) - I run the OS in a VMWare virtual machine
28) Message boards : Number crunching : How to get FreeBSD work? (Message 3460) Posted 6 May 2008 by [BAT] tutta55 Post: I installed the FreeBSD client prepared by crunch3r. It runs, and I attached to the project. But when I try to ask for work, I get "06-May-2008 13:06:58 [Milkyway@home] Message from server: platform 'i386-portbld-freebsd' not found" Any clues? Obviously the server expects another identification string.
29) Message boards : Number crunching : 20 workunit limit (Message 3432) Posted 30 Apr 2008 by [BAT] tutta55 Post: The key concept is "server share" what works for one may very well overload the server if too many quads get on this bus. Think son. Voltron Son? :p Ok dad, I see the point you want to make ;) Hmmmmm 1st crash in weeks overnight .....hmmmm.....Guy I don't know where 'dad' is but down here in Texas and sw US thats just the way we talk(watch some old westerns) :) I didn't take offense. I just thought it funny being addressed as son. And I do appreciate Voltron's (and Yoda's) reply. I hadn't thought of that.
30) Message boards : Number crunching : 20 workunit limit (Message 3419) Posted 30 Apr 2008 by [BAT] tutta55 Post: The key concept is "server share" what works for one may very well overload the server if too many quads get on this bus. Think son. Voltron Son? :p Ok dad, I see the point you want to make ;)
31) Message boards : Number crunching : 20 workunit limit (Message 3411) Posted 29 Apr 2008 by [BAT] tutta55 Post: sysfried, my quad goes idle too if I let it download "the full" 20 work units. I got around that by setting a very small cache (0.02 days). Boinc then downloads only a few work units, and doesn't defer communication. When 2 or 3 WU are finished, it just downloads new ones. As long as there are WU available from the server, this scheme works fine. I could have thougth of that! Darn. But hey, thanks for the heads up. I'll fix it tomorrow.. btw, wrong host id: here's the right one: 15945 Sysfried Hmmmmmmmmmmmmmmmmm... Can anyone think of a downside to this? I can. Voltron Please tell us, because I can't. As long as the server supplies work, this scheme works fine. When the server does run out of work, usually something is seriously wrong and it takes several hours to fix. 20 WU in cache or just 5 or 6 doesn't make much difference then.
32) Message boards : Number crunching : 20 workunit limit (Message 3406) Posted 29 Apr 2008 by [BAT] tutta55 Post: I think the 20 wu limit is ok. Its only that the re-connect is timed to 20 min after each connect, which will cause my big puter to go idle.... http://milkyway.cs.rpi.edu/milkyway/show_host_detail.php?hostid=15565 16 cores... 12 min per wu..--> 8 min idle / 20 min Just a note since the Admin posted that he didn't expect more than 8 core machines here! ;-) sysfried sysfried, my quad goes idle too if I let it download "the full" 20 work units. I got around that by setting a very small cache (0.02 days). Boinc then downloads only a few work units, and doesn't defer communication. When 2 or 3 WU are finished, it just downloads new ones. As long as there are WU available from the server, this scheme works fine.
33) Message boards : Number crunching : Best completion times (Message 3392) Posted 27 Apr 2008 by [BAT] tutta55 Post: That's all and there's no way toargue about OS cpu combination... that's the way it is and that'S it ... all further talk is BS.. Please humour us, and let us post our slowest or fastest times anyway. Us, meaning those of us who like bragging. Bragging is all this thread was intended for :p
34) Message boards : Number crunching : Best completion times (Message 3387) Posted 26 Apr 2008 by [BAT] tutta55 Post: Q9550 speed up to 3.65GHz. Completion time down to 130-132s/WU. Hard to say, but switching from Ubuntu 7.04 to 8.04 may also have improved a few seconds. http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=14685
35) Message boards : Number crunching : Milestones (Message 3359) Posted 25 Apr 2008 by [BAT] tutta55 Post: I have now counted 100000 stars in our milky way :)
36) Message boards : Cafe MilkyWay : Team Recruitment thread - CLOSED (Message 3330) Posted 24 Apr 2008 by [BAT] tutta55 Post: Belgium, a small country with a great team: BOINC.BE Today BOINC.BE entered the top 20 of all projects combined. A remarkable achievement for a team from a relatively small country like Belgium. With our recent average of over 600K/day we rank even higher, at position 16. This means that we can rise even further. As a national team we only let 5 countries before us: the US, Germany, Canada, Australia and the Czech Republic. But we want to do better. We reach the vast majority of crunchers in the north of our country. We would like to launch a special appeal to our friends in the south to join our effort. Help us build a top team where all Belgians feel at home in a friendly, multilingual team and forum.
37) Message boards : Number crunching : Best completion times (Message 3274) Posted 21 Apr 2008 by [BAT] tutta55 Post: Q9550 overclocked 20% to 3.4GHz now. Completion times 155-158s. See http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=14685 That's really amazing, given a Q6600 at higher speed takes longer (190s at 3.8GHz). Must be the extra L2 cache that makes all the difference? Or does it have very fast RAM (like DDR3/1800)? The RAM is DDR2 PC8000 (1000MHz). But I run it 1:1, so currently running at 800MHz. From reviews I read, I gather that for most Boinc projects the large L2 cache will make a difference. But that can't be the only difference. I also read that at stock speed the 2.5GHz Q9300 (6MB L2) outperforms the 2.4GHz Q6600 (8MB L2) in most applications.
38) Message boards : Number crunching : Best completion times (Message 3256) Posted 20 Apr 2008 by [BAT] tutta55 Post: Q9550 overclocked 20% to 3.4GHz now. Completion times 155-158s. See http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=14685
39) Message boards : Number crunching : Best completion times (Message 3244) Posted 17 Apr 2008 by [BAT] tutta55 Post: My new baby after a first small overclock to 3.06GHz: 173s in 64-bit Linux under VMWare. See http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=16443737
40) Message boards : Number crunching : Results purged too quickly (Message 1015) Posted 11 Dec 2007 by [BAT] tutta55 Post: Yes, that would be fine, Travis. If the results are purged too quickly, we cannot see if any of our machines have problems.

Previous 20 · Next 20