Do you have it overclocked? If yes, What software are you using?
No overclocking involved, everything is stock.
Topic: Not able to mine consistently on my 4870x2 using any gpu miner. (Read 3422 times)
No overclocking involved, everything is stock.
Do you have it overclocked? If yes, What software are you using?
Guess it was a driver issue. Updated drivers to Catalyst 11.5 and the kernel errors went away as well as more frequent accepted shares along with almost all the rejected shares.
Now only problem is that one core gets about 83-84 Mhash/sec, while second one only gets 54-55 Mhash/sec. I think before when I was having problems, both cores ran at around 80 Mhash/sec. Any ideas on this or should I start a new topic?
Now only problem is that one core gets about 83-84 Mhash/sec, while second one only gets 54-55 Mhash/sec. I think before when I was having problems, both cores ran at around 80 Mhash/sec. Any ideas on this or should I start a new topic?
Try different drivers or a different SDK version. Simple, but who knows.
Not able to mine consistently on my 4870x2 using sdk 2.4. with deepbit pool.
Tried GUIMiner, but within a few min both cores shut down with an error message saying "verification failed, check hardware". Read that Pheonix continues hashing after error is found, but it too gives me an error "Kernel error: Unusual behavior from OpenCL. Hardware problem?". Even though it contintues to process, the error shows up quite frequently.
Here are the things I've tried so far, to no avail.
1. Checked "Disable Catalyst A.I." under ccc 3d settings
2. Ran phoenix from two seperate dir's
3. Unplugged second monitor, so that second dvi port has dummy plug in it
4. Tried running phoenix with poclbm and phatk kernals
5. Tried reducing workload and agression parms
Results over a 5 hour period running on one core only with these parms.
phoenix -u http://user@host:[email protected]:8332 -k poclbm DEVICE=1 VECTORS AGGRESSION=3 WORKSIZE=128
- 20 accapted shares
- 10 rejected shares
- Kernel error popping up several times an hour, though there was an hour and a half stretch without error
- Phoenix Log shows "LP: New work pushed" every three to five minutes, but a result accepted or rejected message sparadically. Sometimes response comes every 5 mintues, sometimes once an hour, and other times several responses within a few seconds.
Full log is here
http://pastebin.com/eVS4P5pE
Can anyone offer any advice on what I can try to make this work properly?
Tried GUIMiner, but within a few min both cores shut down with an error message saying "verification failed, check hardware". Read that Pheonix continues hashing after error is found, but it too gives me an error "Kernel error: Unusual behavior from OpenCL. Hardware problem?". Even though it contintues to process, the error shows up quite frequently.
Here are the things I've tried so far, to no avail.
1. Checked "Disable Catalyst A.I." under ccc 3d settings
2. Ran phoenix from two seperate dir's
3. Unplugged second monitor, so that second dvi port has dummy plug in it
4. Tried running phoenix with poclbm and phatk kernals
5. Tried reducing workload and agression parms
Results over a 5 hour period running on one core only with these parms.
phoenix -u http://user@host:[email protected]:8332 -k poclbm DEVICE=1 VECTORS AGGRESSION=3 WORKSIZE=128
- 20 accapted shares
- 10 rejected shares
- Kernel error popping up several times an hour, though there was an hour and a half stretch without error
- Phoenix Log shows "LP: New work pushed" every three to five minutes, but a result accepted or rejected message sparadically. Sometimes response comes every 5 mintues, sometimes once an hour, and other times several responses within a few seconds.
Full log is here
http://pastebin.com/eVS4P5pE
Can anyone offer any advice on what I can try to make this work properly?
Jump to: