Welcome to MilkyWay@home

Posts by Martin P.

21) Message boards : Number crunching : Website unreachable (Message 3994)
Posted 4 Jul 2008 by Martin P.
Post:
This is the first time I can access the MilkyWay website since many days. Usually I just can't get in. Also uploads/downloads are slow and are often timed out. What's going on? Bandwidth problem? Server getting short on capacity?


tutta55,

don't be surprised. Weekend is coming. This is the usual time when Milkyway@Home breaks down. It will worsen during weekend and (probably) will be resolved Monday.
22) Message boards : Number crunching : 40,000 credits missing (Message 3456)
Posted 5 May 2008 by Martin P.
Post:
Hi Martin....

I am on my 4th computer since 2000...and I had forgotten at least twice now to merge my computers inside the user's area ....just a thought...


Stan,

same situation here. Unfortunately there is nothing I could merge. The previous system simply dissapeared.
23) Message boards : Number crunching : 40,000 credits missing (Message 3438)
Posted 1 May 2008 by Martin P.
Post:
Hi,

my total credit number is o.k. However, when I add up the credits of my computers there are 40,000 credits missing! On Tuesday I had a major problem with my new MacPro and had to reinstall the system. I migrated all data from a bootable clone. Since that time the 40,000 credits that were done on that computer do not appear in the computers stats anymore although they are correctly displayed in the total credit number.


If you had to reinstall stuff on the computer, boinc included, the computer stats file was probably started new since it seemed that it was a "new" host.


stefsaber,

I understand that. However, in this case the newly installed computer/system would receive a new computer ID and the previous installation would remain in the stats with the old number. But this is not the case. The 40,000 credits that were done on that computer before the new system install are simply gone. Very weird!
24) Message boards : Number crunching : 40,000 credits missing (Message 3422)
Posted 30 Apr 2008 by Martin P.
Post:
Hi,

my total credit number is o.k. However, when I add up the credits of my computers there are 40,000 credits missing! On Tuesday I had a major problem with my new MacPro and had to reinstall the system. I migrated all data from a bootable clone. Since that time the 40,000 credits that were done on that computer do not appear in the computers stats anymore although they are correctly displayed in the total credit number.
25) Message boards : Number crunching : Os-X Time comparisons PPC vs. Intel (Message 3402)
Posted 28 Apr 2008 by Martin P.
Post:
OK... so here we go...

OSX optimized applcations

Have fun :)


Hi Crunch3r,

thanks a lot! Do these apps also work with the Xeons in the MacPro?


Yes. Use the Core 2 (SSSE3) version for the Xeons.


Thanks a lot! These results are fantastic!!!! 239 seconds versus 536 seconds before.
26) Message boards : Number crunching : Os-X Time comparisons PPC vs. Intel (Message 3400)
Posted 28 Apr 2008 by Martin P.
Post:
OK... so here we go...

OSX optimized applcations

Have fun :)


Hi Crunch3r,

thanks a lot! Do these apps also work with the Xeons in the MacPro?
27) Message boards : Number crunching : Os-X Time comparisons PPC vs. Intel (Message 3253)
Posted 20 Apr 2008 by Martin P.
Post:
Paul, Click on my username, then computers, then view. My Q6600 does them in 4:10(OCed to 2.9), my AMD 9600 black (OCed to 2.6) 3:50, my AMD 64 2800 clawhammer 6:10, my AMD64 X2 4800 in 4:27, my AMD64 5200 3:59, and my AMD64 X2 6000 in 3:37. all using 64b Mandriva linux.

Basically, It looks like the mac app is grossly under optimized. Sorry.

tony

{edit} my AMD64 3700 (single core) using 32b windows is doing them in 8 min. I just checked the wife's MOBILE AMD64 3700 and it's doing them in 6:29 with 32b win.


Astro,

unfortunately you are completely right: My 8-core MacPro 3.0 GHz takes more than twice as much time (535 sec/WU) for a work-unit than the 2.4 GHz Linux machines (190-210 sec./WU) and almost exactly the same time as my 3-year-old G5 Dual 2.7 GHz machine. Any chance we will see optimized Mac-apps soon?
28) Message boards : Number crunching : application v1.21/v1.22 errors/memory leaks/crashes here (Message 3094)
Posted 11 Apr 2008 by Martin P.
Post:
Hey Dave and crew....30 days no reports of problems...unsticky in order? ( the mod in me coming out ) ;=)


Well, there is a (continued) problem: My 8-core MacPro 3.0 GHz takes more than twice as much time (535 sec/WU) for a work-unit than the 2.4 GHz Linux machines (190-210 sec./WU)

I'll suspend Milkyway on my new machine until this problem gets solved.
29) Message boards : Number crunching : Server Outages (Message 2164)
Posted 11 Mar 2008 by Martin P.
Post:
Man, these outages are getting really annoying... >:(
Is there nothing you can do about it? ;-)


Yes, there is: tell us that they work on it...

30) Message boards : Number crunching : App 1.16 (Message 1835)
Posted 4 Mar 2008 by Martin P.
Post:
Just finished the first work-units with (Mac) App 1.16. They validated o.k. However, the timing bug is still there: the timer goes to 100%, then restarts at 0 and goes back to 0.150%. At that point it suddenly jumps to 100% and uploads.
31) Message boards : Number crunching : No work from project ... (Message 1463)
Posted 13 Jan 2008 by Martin P.
Post:
As the Clash said.... Calling Travis!

So, when do we get new work?


There's something very wrong! Travis had a very good record of keeping us informed about what's going on, but this time he did not post for several days, and if he does post it is very vague. Weird.
32) Message boards : Number crunching : Hard to get new work ! (Message 1437)
Posted 11 Jan 2008 by Martin P.
Post:
i'm hoping these are just because of labstaff taking down the machine for awhile. are these problems still happening?


Travis, I have not received a single new work unit on any of my 3 computers since Jan. 9th!
33) Message boards : Number crunching : error in progress display (Message 1119)
Posted 19 Dec 2007 by Martin P.
Post:
The new and longer work units have an error in displaying the progress. The progress bar advances to 100% and then restarts at 0%. It then keeps crunching for another minute or 2 until progress is again at 0.15% and then jumps directly to 100%. This happens with both Windows and Mac clients.

Results validate o.k. and do not show an error.
34) Questions and Answers : Wish list : Navigation bar on top of pages (Message 1102)
Posted 16 Dec 2007 by Martin P.
Post:
Please insert a navigation bar on top of all pages (e.g. My account) or at least the Milkyway-logo. When you run several projects it is not obvious on which pages you currently are and it also makes returning to the mainpage easier than scolling all the way down to the bottom of the page.
35) Message boards : Number crunching : Started Particle Swarm Searches (Message 1062)
Posted 14 Dec 2007 by Martin P.
Post:
Isn't it 20 WU's at a time, instead? I'm under the impression that it's:
{(Unstarted WU's)+(WU You're working on)+(Finished WU not yet reported)=20} at any particular time.

I dont know, because I crunched 20 WUs and I am not getting any other new WUs


They were having problems with power/internet outages last night so it seems like they've had another overnight, won't be any work for a few more hours most likely.


tahanko,

the daily quota is 2,000 work-units, not 20. The 20 only refers to the maximum number of work-units that you can have on your computer at any time. As soon as you finish a few work units they will be transmitted and the missing number of new WUs to 20 will then be transferred to your computer.
36) Message boards : Number crunching : No work from project ... (Message 1060)
Posted 14 Dec 2007 by Martin P.
Post:
And again ...
37) Message boards : Number crunching : No work from project ... (Message 982)
Posted 10 Dec 2007 by Martin P.
Post:
No work again. Pipeline seems to be empty.

38) Message boards : Number crunching : Account page error (Message 929)
Posted 4 Dec 2007 by Martin P.
Post:
Catchable fatal error: Object of class stdClass could not be converted to string in /export/share0/www/boinc/milkyway/html/inc/db_forum.inc on line 220


Thats what my account page now reads:( and hate to keep bugging about the stats update but going on 2 days again. Write yourself a sticky perhaps?


Same problem here.

39) Message boards : Number crunching : No work from project ... (Message 846)
Posted 30 Nov 2007 by Martin P.
Post:
Sorry about that everyone, one of the files on the server that was used to generate work units went missing (of course while i was sleeping), and everything crashed to a screeching halt. on another note, work is back and i'm updating the assimilator tonight so we'll be starting to do particle swarm optimization in addition to genetic search.


Travis,

work units are flowing again. However, these are extremly short units, just 22 seconds each. These might overflow the servers, since BOINC contacts the server almost continuosly.
40) Message boards : Number crunching : server really back up (Message 807)
Posted 29 Nov 2007 by Martin P.
Post:
Most WU are completing and receiving credit. However, all my successful completions have this error in the output:

<core_client_version>5.10.28</core_client_version>
<![CDATA[
<stderr_txt>
APP: error writing checkpoint (resolving checkpoint file) -108
output_path: astronomy_checkpoint

</stderr_txt>

It doesn''t seem to affect validation or credit, but thought I should mention it anyway.

C


The same error message also appears on the Mac (PPC) version.


Previous 20 · Next 20

©2024 Astroinformatics Group