Thanks. That did the trick!

As of this morning, despite being set for PPLNS all my shares are going to PPS. I've tried changing method back and forth, creating new workers. Still can't get off PPS and back to PPLNS.

Any workaround recommendations or known issue?

Thanks. Pushing 430 on BTCGuild and running fine in GUIMiner. I can definitely isolate the problems with the hash rate crashing to deepbit specifically.

No issues w/ p2pool in CGminer either.

430 is what I'm targeting. What pool are you using w/ GUIMiner?  I still can't solve the problem of the GPU dropping out after 10-20 minutes regardless of setting w/ deepbit.

I have the same problem when using guiminer with deepbit and a 7870.

cgminer works fine though - averaging 415 Mh/s

I've been playing around with intensity setting in cgminer and noted on a 7870 when I went to I=13 I was suddenly getting mostly rejections.

So besides affecting hash rate, can a suboptimal intensity also lead to high rejection rates?

Running all night, I'm showing 413 Mh/s average in CGMiner at I 7 1200/750

Feels like I should be able to push this a bit higher. Curious about current hash rates of other 7870 owners.

I doubt the dollar is involved in any sort of tipping point w/ bitcoin. More likely the Euro or Yen between what's happened in Cyprus and the BOJ's insisting on a 2% inflation target for the yen "at all costs"

In case it helps anyone else, my problem seems to be with guiminer.

Switching to p2pool using cgminer I've not had any issues with GPU usage crashing to 0 or IO errors.  Manually upping fan to 80% also keeping gpu temp around 63. Noisy, but still trying to dial things in.

Still can't figure out how to run deepbit on cgminer to compare with guiminer hash rates. Will figure it out though.

Thanks!

Looking around more, perhaps my biggest issue is GUIMiner. Going to see if I can get CGMiner up and going.

Thanks again for giving me some new ideas to try.

Thanks for response.

I'm running AMD APP 2.8, perhaps that's the problem.

I don't think its an overheating issue now though - I cranked the fan to 100% and used your settings and kept the temp around 67C, but it still kicked the GPU down to 0% usage after about 10 minutes.

May I ask if you're using any extra flags?

I got a new MSI Radeon 7870 HD and have underclocked the memory to 750 and upped core to 1200.

Getting 390M at standard clock and 430M with above settings. Problem is regardless of settings, temp runs up to 80C, runs for 10-20minutes then GPU shuts down and hash drops to 400k.

So the newb question is my PSU has two 6 pin connectors Y'd on two separate cables - so total of 4. I'm using one Y'd pair for both required connections on the card. Could this be an issue?  Should I run one 6 pin connector from each cable? 

Feel like I've covered most other possibilities. Any thoughts are much appreciated.

happy to join the board after reading nearly every thread on here

look forward to joining the community