Posts by DoctorNow
|
41)
Message boards :
Number crunching :
post milkyway_windows_intelx86 problems here
(Message 7426)
Posted 4 Dec 2008 by  DoctorNow
Post: The Windoze version hasn't been added to the servers yet. Oh? Well as far as I see it is already... http://milkyway.cs.rpi.edu/milkyway/apps.php And my crunchers at home got work as well for this app. ...two days later, no change. Still getting *_stripe* WUs with extremely large estimated run times (>>100hrs). But these WUs get fully processed within an hour or so. Well, if it's still too high you could try to manually edit the client_state.xml-file. Open it with Notepad or something similar, search for the Milkyway part and then for the duration_correction_factor. If this value is very high, set it to 1, that should fix the high estimation time. Don't forget to restart BOINC after the editing. |
|
42)
Message boards :
Number crunching :
credit comparison to other projects
(Message 7241)
Posted 2 Dec 2008 by DoctorNow
Post: There you go: ...on my AMD... Is there no better optimised SETI-app for an AMD? That's pretty unfair in my eyes... ;-) |
|
43)
Message boards :
Number crunching :
credit comparison to other projects
(Message 7237)
Posted 2 Dec 2008 by DoctorNow
Post: Think I must be doing something wrong then, 41 credits an hour on Seti with the optimised app, 33 credits an hour with the new app here. Intel dual core @ 2.1GHz. Wow, what optimized app do you use for SETI? I never got a credit level better than 30 credits on my AMD with the app I used... |
|
44)
Message boards :
Number crunching :
Faster application (links inside)
(Message 7234)
Posted 2 Dec 2008 by DoctorNow
Post: Everyone is now on a level playing field! As it should be! Don't bet on that! This is an open-source project now and everyone can download the code and optimize it when he finds a way to make the app faster. |
|
45)
Message boards :
Number crunching :
credit comparison to other projects
(Message 7225)
Posted 2 Dec 2008 by DoctorNow
Post: 50/h but on which cpu? I'm getting nearly 50 credits/hour on my AMD X2 5200 under Windows 32-Bit, too. Would do a bit better under Linux 64-Bit I guess, but I haven't tried it out yet, my Linux-hard disk is a bit corrupt and I haven't repaired it yet. ;-) |
|
46)
Message boards :
Number crunching :
purge set to 1 day
(Message 7224)
Posted 2 Dec 2008 by DoctorNow
Post: The database should purge WUs after 1 day now (as opposed to immediately). Yeah, thanx from me, too, that's really nice. :-) |
|
47)
Message boards :
Number crunching :
credit comparison to other projects
(Message 7159)
Posted 1 Dec 2008 by DoctorNow
Post: Well, as far as I'm concerned I'm quite satisfied with the credits, you shouldn't change anything about that! I get a credit level from 40-50 on my AMD X2 5200 under Win 32-Bit. Have not yet checked it under Linux 64-Bit but I'm sure it's not that much different. Leave it like it is and forget about the credit whiners. ;-D Anyway I think, studying the MilkyWay is much more worth than searching for aliens right now. ;-) |
|
48)
Message boards :
Number crunching :
work availability
(Message 7033)
Posted 30 Nov 2008 by DoctorNow
Post: LOL, very good, Nicolas... ;-) Would be funny to really see that, would give a touch of Pirates. ;-D |
|
49)
Message boards :
Number crunching :
milkyway v0.3 performance
(Message 6724)
Posted 25 Nov 2008 by DoctorNow
Post: We'll be able to increase the amount of time WUs stay on the database after the new harddrives for milkyway get here and are installed. Okay, thanks. :-) Found out that they are up around 10 minutes now. Seems I was a bit slow the first time. ;-) |
|
50)
Message boards :
Number crunching :
milkyway v0.3 performance
(Message 6722)
Posted 25 Nov 2008 by DoctorNow
Post: Ok, got one so far, works fine here. :-) Seems they are a bit faster again like the previous ones. ;-) |
|
51)
Message boards :
Number crunching :
milkyway v0.3 performance
(Message 6714)
Posted 25 Nov 2008 by DoctorNow
Post: I just wanted to take a look at my first two crunched 0.3-WUs but it seems they're already purged from the database. How long do they stay now until they're gone? Can you please make it longer... |
|
52)
Message boards :
Number crunching :
testing assimilator
(Message 6697)
Posted 25 Nov 2008 by DoctorNow
Post: I think i've fixed this problem by updating the wu_template... there may be a few more WUs like this out there, but all nm_test22 should be ok. let me know if any of these break. Got some of the nm_test22-WUs but they all error out due to this: :-( 11/25/08 17:53:29|Milkyway@home|Giving up on download of nm_test22_astronomy_parameters.txt: file not found |
|
53)
Message boards :
Number crunching :
Error on connect can't attach shared memory
(Message 6593)
Posted 24 Nov 2008 by DoctorNow
Post: this still happening for people? Yep, can't report as well... |
|
54)
Message boards :
Number crunching :
Error on connect can't attach shared memory
(Message 6586)
Posted 24 Nov 2008 by DoctorNow
Post: It happens again now: 11/24/08 18:33:21|Milkyway@home|Message from server: Server error: can't attach shared memory Can't report due to that... |
|
55)
Message boards :
Number crunching :
validation testing
(Message 6581)
Posted 24 Nov 2008 by DoctorNow
Post: Are you sure about that? I'm getting 1.8 credits per hour consistently with a slightly optimized application. Well, looking at your results on this host and calculating the hour value (credits*3600 secs/crunch time in sec) I'm still getting about 108 per hour. Maybe you miscalculated them a bit. ;-) |
|
56)
Message boards :
Number crunching :
validation testing
(Message 6578)
Posted 24 Nov 2008 by DoctorNow
Post: Also, what would be a good granted credit for the test WUs? I'm pretty sure it's way too high atm :P Hehe, it is. Hits the credit limit again with 0.03 per sec and therefore 108 credits per hours. Not sure what to give elsewhere. Since the credits "should" go on line with the other credits my first suggestion is about 36 credits//hour now, is that good? ;-) So if the length at about 20 minutes stays, it would be nice to see a fixed level of 12 per WU. |
|
57)
Message boards :
Number crunching :
New App status
(Message 6350)
Posted 21 Nov 2008 by DoctorNow
Post: Oh, there are much more projects out there who use this function already. The server software supports it now anyway, but I think the standard if you build a server with it is, that the function is hidden. |
|
58)
Message boards :
Number crunching :
New App status
(Message 6345)
Posted 21 Nov 2008 by DoctorNow
Post: I cannot see it in either "Computing Preferences" or my "Community Preferences". Look in the MilkyWay Preferences. ;-) But since there is no "test" app so far it's not necessary yet. Nevertheless the idea from Brian isn't that bad, could be used for future apps I think. :-) |
|
59)
Message boards :
Number crunching :
No further Support for Milkyway at this time!
(Message 6256)
Posted 18 Nov 2008 by DoctorNow
Post: But regarding the credit limit's sense I think your wrong. Milksop somewhere posted two workarounds. And they are being used. So this even furthers the unbalance, as the way to manipulate the client is only usable by very few. the other way is easier but way less effective. Having read that, I did a closer look at all Milksop's posts here on the board again, but couldn't find such information. I could have overreaded it, but I don't think so. Can you please post a link to this "workaround" then? I don't want that everyone should be aware of that, but if he really posted it somewhere, I want to know (only for information if it's really true of course!), so PM it to me. As far as it is really being used to avoid the credit limit, there couldn't be much people who use it, as the stats of MilkyWay don't show a significant difference anywhere. Besides that I think it was wrong to introduce this limit because crunch3r actually produced that lot more of science. If the project didn't want it they should have prohibited the use of his appl. Well, to be honest I'm wondering now why the project admins made the project OpenSource, as far as I see there is only a big mess and it seems they are not capable of handling it right. Don't understand me wrong, I'm happy that the science is better now because some "outsiders" could improve the code, but in my eyes it's useless to make a project OpenSource when you can't deal with the consequences. I'm still hoping that the release of the new app - hopefully really this week as announced - shuts some people's voices down. ;-) |
|
60)
Message boards :
Number crunching :
No further Support for Milkyway at this time!
(Message 6240)
Posted 17 Nov 2008 by DoctorNow
Post: "cheater app" ? WTF ? There was no, and still is no cheater app. Well, I didn't meant it that way. I did not put the words within quotation marks without sense, you know? If I would have called it really cheating then I wouldn't have used the marks. It was meant as a "metaphor". That's all. I hope you understand it now better. |
Previous 20 · Next 20
©2024 Astroinformatics Group