I just took this to be a fact of life for a while, but now that I am trying to automate restarting the polcbm process it's becoming really annoying. If I X out of the cmd window or do a taskkill to the exe, everything goes black for a few seconds and I get a "The display driver has stopped working and has recovered " message. I'm on an NVIDIA card on Windows 7 Ultimate and have the latest drivers. When this happens my fan control gets reset and I have to set it again, not to mention that it's annoying when it blacks out like that. In fact sometimes it even has rebooted after the black screen so I'd really like this not to happen. To get around it, I sent a keyboard interrupt (Ctrl C) in the cmd window a couple of times and it would exit ok.

Does everyone have this problem? Is there some sort of other way I can kill the process without killing the display adapter every time?

I'm using a mix of Phoenix/polcbm-mod on my machines, but unfortunately my main rig is having hardware issues and will error out after several hours. I tried resetting my bios to factory settings (no overclocking) and maxing out my fan speed. That seemed to help a little, but still after maybe 5 hours it will inevitably decrease by a factor of 10 (!). Obviously I don't want to keep it mining at that rate, and there don't seem to be any ill effects when I restart it. The only thing I haven't tried is underclocking the GPU, which I'd rather not do since it should be running fine at stock settings. My case has very good airflow, so I don't know what's going on. Anyway this is a separate issue so...

I'm this close to creating a python/ruby wrapper just for this machine that will kill/autorestart when this happens, but it would be nice to have some built in support (warnings, stderr) if we get a hardware failure or our Mh/s is very low. Is there any tool that does this?

If I have to write one myself, I'll release it to the forum in hopes it will help other people with hardware woes.

I've got a lot of machines I don't really use all the time so I decided to set them mining. They mine from about 1-5 Mhs/s each.

I recently switched to Bitcoins.cl to try it out, as I was having issues with Slush's pool not giving me good payouts given my hashrate. Bitcoins.cl does show avg. hashrate which is helpful (although it was taken down for tuning I guess recently).

Anyway I don't know if this is an issue because I switched pools but is it just a fact of life that slow miners will get stale shares a lot of the time?

Edit: Forgot to add that I'm using Phoenix or polcbm mod for all of the machines, which SHOULD support long polling, so in theory I wouldn't get this problem as much.

I don't actually pay electricity at my rental so I figured I'd try putting all of my computers into Slush's pool. Only one of them has a decent hash rate, at ~50,000kh/s . The rest are contributing maybe an extra 10,000kh/s.

My max reward on Slush's pool was about 0.05 for a day, but due to issues with my main rig dropping hash rate, now I'm only getting around 0.01.

I'm trying to figure out if my rig going down to 3000kh/s is the real cause of this.  On June 6th and 7th I got ~0.05 and I must have still been having the same setup, if not worse. It's hard to explain how this is happening! On June 5th I started in the evening and got close to what I have gotten for a whole day the past few days.

Would you suggest switching to another pool? It's hard to see what my average is or if I was just lucky for  few days.

Please someone help, this problem is driving me crazy. I only have one computer that is worth mining on, and I'm having stability issues. I was overclocking my CPU a bit, and I thought it was due to this. I analyzed the log and whenever I had hardware failures it would drop, so I reset to factory defaults.

Now using both poclbm and polcbm-mod even with default settings on my machine, I drop from 50ms/s to 4mhs/s. There are no errors in the log.

Does anyone have any idea what could be going on? The computer is literally running idle for hours and I come back ant it is stuck at 4mh/s. I haven't been watching in enough to know if it drops suddenly or gradually down to 4.

This is an NVIDIA card and it is running the latest drivers. My internal temps only ever go up to 56C max, which is not that hot for this machine.

At the very least, is there some sort of flag or utility that will kill the process and restart if your mh/s go way down?

Hi everyone, I am new to mining. I recently set up a new rig because I have some nice hardware I don't want to go to waste. I don't pay electricity at my rental so I figured why not! I have about 3 computers set up around the house which are doing light mining, around maybe 3-1 Mhash/s. But my main computer can get ~50. I'm using the GUI Miner with slush's pool, I've tried the python one but as far as I can tell this does the same thing under the hood.

It was a hot day this weekend and I was actually getting some "hardware failed" errors. I am slightly overclocking my CPU so it could be that there's not enough stability there. However today it is much cooler and I haven't been getting the errors.

What I have been getting though, is that the 50Mhash drops to around 3 if I leave it sitting around and come back. If I restart mining, it goes back up to ~50.

I noticed this performance drop if something is using the CPU at 100%, but when it starts reporting this there is not much load on the CPU. I suppose it could have spiked but it wasn't at 100% at the time when it consistently goes down to 3 Mhash/s.

How can I debug this problem? Are there any known issues with the miner that ships with Guiminer? I updated the binary to the latest one and then rolled back to the one in GuiMiner to make sure I wasn't just having odd issues.

Additionally whether it is GUIMiner or what, is there any client that supports auto recovery? I hate that I have to baby this machine and make sure it hasn't dropped again. It's sad that 3 crappy computers are contributing more than this main rig just due to these issues. Thanks!