- Joined
- Jan 4, 2005
- Location
- Vancouver, WA
I've been crunching 1 WU on a core from my Q6600 for over a day now, 28 hours, and it still says it has 10 hours left. Whats the deal? Usually WU's crunch insanely fast on this rig.
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
I don't think it works like that. It doesn't matter if I turn in a work unit on the stock e6400 or my overclocked one. If it is the same WU, I get the same amount of points.work done does = credits claimed.
if it takes you a week to read a book for a class, and someone else does it in a day, you've both done the same amount of work, and will both claim an A+, even if it did take you a different amount of time.
594348667 149295438 19 Aug 2007 7:59:40 UTC 28 Aug 2007 22:19:57 UTC Over Success Done 6,955.67 63.98 63.98
594348663 149295432 19 Aug 2007 7:59:40 UTC 29 Aug 2007 3:46:11 UTC Over Success Done 20,497.47 63.98 63.98
work done does = credits claimed.
The work done isn't based on how long it takes to crunch the unit, it is based on the number of calculations to crunch the unit. So if my pentium II takes 12 hours to finish a unit, and an overclocked P4 at 6.0Ghz takes 1 hour on the same unit, then they have still done the same amount of work, because they have both done the same number of calculations.
If you are being paid $10 to move a pile of bricks from point A to point B, you will still be paid the same amount whether you finish in 1 minute or an hour.
If you get 60 credits for finishing a work unit, you will get the same amount whether you finish in 1 minute or 1 hour (or 36 hours).
I tried finding a link supporting this on the SETI forums, but I couldn't, so my evidence is that if you look in the result files, there is a value "flops = XXXXXXXXXXXXXXXXX"
In the end, I'm not 100% sure that I'm right, but it does explain why everyone ALWAYS gets the same amount of credit for the same work unit. If you have a link that proves me wrong, lemme see it.
If nothing else, I think we can all agree that the credit system is really screwed up right now.
I couldn't get the second link to come up - "No such result" according to SETI?
Anyway, your first link I can see and it isn't over yet; "Over Success Done 8,219.66 89.55 pending". Since the last entry is "pending" you may not get all that credit once the other computer checks in (validates). I've seen several cases recently where "credit claimed" isn't the same as "credit granted".
I noticed that odd result, too, as I was looking over your other WUs. Let us know when it's validated - I'd like to see the final results on that one ...
Ah! It's obvious when you see it that way!
This is what it shows in the BOINC manager, sorry for the huge image -- don't have any JPEG compression software on this computer other than paint.
From your page 2 (now):Odd result? care to link?
That last WU is a "little" out of balance with the rest - I'd like to see if it corrects when the other computer validates it ...
Yeah this is really strange. The estimated times seem to drop surprisingly fast when the BOINC manager is unminimzed, whereas when it is minimized to the task bar as an icon, it doesn't seem to actually be doing anything but my CPU will be under load.
Another thing I forgot to mention is I'm using the KWSN_2.4_SSE3-Core2 SETI client
Rerunning the CPU benchmarks gives 2387 floating point MIPS per CPU and 5196 integer MIPS per CPU.
I wasn't so shocked by those comp times until I looked at the validations and they're waaayy under your times. Something is definitely wrong there.594425798 149331907 19 Aug 2007 12:06:52 UTC 1 Sep 2007 23:07:19 UTC Over Client error Compute error 171,995.41 38.10
594425646 149331862 19 Aug 2007 12:06:52 UTC 2 Sep 2007 0:01:37 UTC Over Success Done 158,899.34 85.21 85.20
594425580 149331806 19 Aug 2007 12:06:52 UTC 1 Sep 2007 23:34:12 UTC Over Client error Compute error 142,187.64 85.21 ---
Those are insane computation times, when its usually between 6 and 20 thousand seconds.