Message boards :
News :
testing new validator
Message board moderation
Previous · 1 . . . 3 · 4 · 5 · 6 · 7 · Next
Author | Message |
---|---|
Send message Joined: 18 Feb 08 Posts: 4 Credit: 110,489 RAC: 0 |
My second WU to be marked invalid: Workunit 91877832 The cuda and the HD5800 validated and left my poor Mac's CPU out in the cold. In case it's purged: de_s222_3s_13_360570_1270920195 application MilkyWay@Home created 10 Apr 2010 17:23:15 UTC canonical result 99637572 granted credit 213.76 minimum quorum 2 initial replication 3 max # of error/total/success tasks 3, 9, 6 This is displayed on the workunit page Task ID click for details Computer Sent Time reported or deadline explain Status Run time (sec) CPU time (sec) Claimed credit Granted credit Application 99637572 121302 10 Apr 2010 17:27:11 UTC 11 Apr 2010 21:10:34 UTC Completed and validated 813.94 24.32 0.18 213.76 MilkyWay@Home v0.24 (cuda23) 99698813 42872 11 Apr 2010 23:32:02 UTC 14 Apr 2010 11:42:03 UTC Completed, marked as invalid 27,759.43 25,764.73 195.20 0.00 MilkyWay@Home v0.26 101340563 155415 14 Apr 2010 11:43:12 UTC 14 Apr 2010 11:59:27 UTC Completed and validated 78.36 6.51 0.05 213.76 MilkyWay@Home v0.23 (ati13ati) |
Send message Joined: 11 Sep 09 Posts: 5 Credit: 40,080,262 RAC: 0 |
wann funkrioniert die scheisse hier wieder es kann nicht sein das 50% für den arsch sind |
Send message Joined: 17 Feb 08 Posts: 363 Credit: 258,227,990 RAC: 0 |
wann funkrioniert die scheisse hier wieder es kann nicht sein das 50% für den arsch sind Die "Scheisse" funktionert so wie es soll. Es liegt mit hoher Warscheinlichkeit an deinem Scheiss übertakten...macht wohl deine Scheisskarte nicht mit und produziert nur falsche Resultate. Darum bekommst du auch keine Scheiss Credits mehr... So ne Scheisse aber auch ... Join Support science! Joinc Team BOINC United now! |
Send message Joined: 11 Sep 09 Posts: 5 Credit: 40,080,262 RAC: 0 |
ob übertaktet oder nicht das Ergebnis ist nicht fürs Gesicht |
Send message Joined: 22 Mar 09 Posts: 99 Credit: 503,422,495 RAC: 0 |
Bitte, Bitte, wir machen hier Wissenschaft und da kommt es vor, das manchmal Dinge nicht funktionieren. Ich möchte nicht daran zurückdenken, wie viele Experimente in meiner Diss (organische Chemie) schief gelaufen sind, und der schwarze Schleim im Kolben sah wirklich aus wie Sch... Also bitte die Haltung bewahren :-) Please, we are all making science here and so it can happen that sometimes things go wrong. I don´t want to remember, how many experiments in my PhD-thesis (organic chemistry) went wrong, and the black slime in the flask looked like Sch... So please save your countenance :-) |
Send message Joined: 11 Sep 09 Posts: 5 Credit: 40,080,262 RAC: 0 |
wenn meine karte bei allen anderen Projekten funktioniert, nur hier nicht liegt das nicht an meiner karte |
Send message Joined: 27 Aug 07 Posts: 647 Credit: 27,592,547 RAC: 0 |
Kann man so allgemein auch nicht sagen. Vielleicht vertragen andere Projekte einfach grössere Abweichungen bei den Ergebnissen. Eine geringere Fehlertoleranz ist aber kein "Fehler" des Projektes. :-P Lovely greetings, Cori |
Send message Joined: 11 Sep 09 Posts: 5 Credit: 40,080,262 RAC: 0 |
ist ja nicht so als das meine karte hier noch nie funktioniert hat. sie hat ja schon mal zu 100% funktioniert |
Send message Joined: 8 Feb 08 Posts: 261 Credit: 104,050,322 RAC: 0 |
Ein paar Leute haben geschrieben, dass sie mit der neuen ATI app ihre Ãœbertaktung etwas reduzieren mussten. Different programs are stressing the GPU in a different way. If you are running at the limit of the card, even a new prog version can push it over the line. Suggestion: Set your card back to default speed for the moment. Wait until the new programm (MW3) is out and stable, than see where your new oc limit is. |
Send message Joined: 22 Mar 08 Posts: 1 Credit: 2,071,306 RAC: 0 |
Maybe it could be a good idea to send WUs for validation to different types of systems if possible (for example no more than 1 to 58X0, etc.) or no more than one to systems that have many validation problems recently (58X0 whould currently qualify for that, but in future that could of course change). |
Send message Joined: 11 Sep 09 Posts: 5 Credit: 40,080,262 RAC: 0 |
aber wenn die fehlertoleranz so gering ist das man die mouse nicht mehr bewegen kann ist es vieleicht etwas übertrieben. man hat auch mal was zu tun am pc |
Send message Joined: 26 Jul 08 Posts: 627 Credit: 94,940,203 RAC: 0 |
aber wenn die fehlertoleranz so gering ist das man die mouse nicht mehr bewegen kann ist es vieleicht etwas übertrieben. man hat auch mal was zu tun am pc Wenn Deine Karten schon fehlerhafte Resultate ausspucken, wenn Du nur die Maus bewegst, würde ich mir ernsthaft Gedanken machen. Bei mir ist das vollkommen egal. Ich lasse MW auch auf meinem normalen Arbeits-PC laufen (habe da eine 3870X2 reingesteckt), da kann ich machen was ich will und es läuft ohne ungültige Ergebnisse. Hilfreich wäre vielleicht, wenn Du mal ein paar Angaben über Deinen Rechner machst (GPU, Takte, OS, Treiber, MW-Version). PS: Wenn es geht, sind übrigens englischsprachige Posts zu empfehlen. |
Send message Joined: 6 Jul 09 Posts: 1 Credit: 1,714,690 RAC: 0 |
So, am I the only one running nVidia that's still having troubles? Host reference : http://milkyway.cs.rpi.edu/milkyway/show_host_detail.php?hostid=161353 I've been keeping up w/the news about the validation problem. I'm using the stock applications from MW@Home, and I'm running nVidia (dual GTX 295). I don't overclock the CPU or GPU because I don't want to void the warranty. I uploaded 24 WUs a little bit ago, only to have 9 of them go Validation inconclusive. Any help would be appreciated. - Safari Hood |
Send message Joined: 1 Jul 09 Posts: 8 Credit: 1,734,500 RAC: 0 |
I uploaded 24 WUs a little bit ago, only to have 9 of them go Validation inconclusive. For some obscure reason WUs are listed as "Completed, validation inconclusive" instead of "Completed, waiting for validation" even if there is no other deviating result that would make the validation inconclusive. Currently I see 8 of 10 of your WUs in the "validation inconclusive" state although there are no deviating results. One WU is labeled as "Completed, waiting for validation" with one task "Completed, waiting for validation" and the other task "Completed, validation inconclusive". The validator is still buggy in this regard. |
Send message Joined: 26 Apr 08 Posts: 87 Credit: 64,801,496 RAC: 0 |
I posted the following message earlier, but I may have stuck it in the wrong thread: I don't know if these are the workunits that I am having problems with but I am using the stock app with a stock Mac mini core duo (plus a few on a core 2 duo) and more than half of the units I return are marked invalid. Many of the units start with a quorum of one and then after they are returned the quorum jumps to two. I usually end up paired with a GPU and then it will validate against a third machine (also running a GPU) leaving mine as odd man out. Shouldn't the stock app on a stock CPU return (mostly) valid results? I like this project and I realize that I am not helping very much compared with the GPU folks but I would really like to be able to return valid units. Any help or insight would be appreciated. I have tried to keep up with the postings on the forum boards here, so if I missed something vital somewhere please point me in the right direction. Thank you. Plus SETI Classic = 21,082 WUs |
Send message Joined: 12 Nov 07 Posts: 2425 Credit: 524,164 RAC: 0 |
I uploaded 24 WUs a little bit ago, only to have 9 of them go Validation inconclusive. "Completed, waiting for validation" = waiting on another task "Completed, validation inconclusive" = wu wasn't quite on target, and another was sent out and waiting for it to come back. Doesn't expecting the unexpected make the unexpected the expected? If it makes sense, DON'T do it. |
Send message Joined: 1 Jul 09 Posts: 8 Credit: 1,734,500 RAC: 0 |
I uploaded 24 WUs a little bit ago, only to have 9 of them go Validation inconclusive. The entry "Completed, validation inconclusive" should only be set if there are deviating results and not if the result is the only one sent back. If the other WU is unsent or no result is reported back the entry should be set to "Completed, waiting for validation". Just my two cents... |
Send message Joined: 27 Nov 09 Posts: 108 Credit: 430,760,953 RAC: 0 |
I uploaded 24 WUs a little bit ago, only to have 9 of them go Validation inconclusive.Offhand, except for Workunit 90594183, all of the pending WU's you submitted appear to be in the expected state based on what my own units do. All results require a confirmation from another client (quorum=2) and your work units are showing 'Validation inconclusive' because none of the other results have come back yet. In fact, some of those confirmation tasks haven't even been sent out yet. You should post a complaint about 90594183 in the "Number Crunching: Waiting for Validation..." thread (http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=1678) to let Travis know there is a problem. A lot of us had problems with validation around the same time. |
Send message Joined: 30 Aug 07 Posts: 2046 Credit: 26,480 RAC: 0 |
I uploaded 24 WUs a little bit ago, only to have 9 of them go Validation inconclusive. Sadly this is just how the BOINC server relays messages. If your WU gets set to Completed, validation inconclusive, that just means the server has sent out more results and is waiting to complete validation. Completed, waiting for validation means that the result hasn't gone through the validator at all yet. |
Send message Joined: 17 Mar 08 Posts: 165 Credit: 410,228,216 RAC: 0 |
OOPS it would seem that the server is out of work.. Ouch.. :) |
©2024 Astroinformatics Group