Quote:
Originally Posted by BUCK NASTY
Um, looks like the 912 and 925 point WU's are back  (P109xx), but who knows for how long. I think it may have been triggered by server issues. Don't forget to increase your overclocks to take advantage of the PPD. After the P680X return, you will want to back the clocks down to avoid EUE's and 24hr Pausing.
EDIT: Woot! GTS450 thinks it's a GTX460 again!!! |
My GTX460 experience so far. I replaced an 8800GTS(512) that was reading 185 watts for the whole system while Folding. The GTX460 system, OC'd to 821 on the core reads the same wattage with the P68xx WU and 191 with the P109xx. My points for the day would be much higher if it hadn't taken this weird 3 hour break in the middle of the night.
Quote:
[10:32:34] Folding@home Core Shutdown: FINISHED_UNIT
[10:32:38] CoreStatus = 64 (100)
[10:32:38] Sending work to server
[10:32:38] Project: 6801 (Run 2061, Clone 1, Gen 9)
[10:32:38] + Attempting to send results [January 29 10:32:38 UTC]
[10:32:38] Gpu type=3 species=30.
[13:16:59] + Could not connect to Work Server (results)
[13:16:59] (171.64.65.64:8080)
[13:16:59] + Retrying using alternative port
[13:17:20] - Couldn't send HTTP request to server
[13:17:20] + Could not connect to Work Server (results)
[13:17:20] (171.64.65.64:80)
[13:17:20] - Error: Could not transmit unit 01 (completed January 29) to work server.
[13:17:20] Keeping unit 01 in queue.
[13:17:20] Project: 6801 (Run 2061, Clone 1, Gen 9)
|
I woke up to my cards all working on their best PPD WU's, 353's on the G92's and a 925. There's a huge spike in the number of returned WU's in the last EOC update, which I think shows the effect of the 353's and the 9xx's.
Quote:
Originally Posted by Athlonite
so what does the avg PPD work to be for your PII x4 doing smp WU's I couldn't get shit out of mine
|
I think your problem is Folding with ATI GPU's, especially if you never set the environmental variables. (step 5 in the first post in
Setting up GPU2 on ATI cards! and post 24 tells you where). Even with them set, I would recommend leaving one core available to handle the GPU(s). Try it with SMP -3.