I tried a new .conf file, reverting to stock clocks, different video cards, etc. And this has been happening on all of my rigs, not just one.
Topic: OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.0 - page 425. (Read 5805983 times)
December 12, 2012, 02:57:25 PM
I'm also having a similar issue where CGMiner freezes after a certain amount of time. Sometimes it can go fine for a few days but lately its been crashing more frequently.
I tried a new .conf file, reverting to stock clocks, different video cards, etc. And this has been happening on all of my rigs, not just one.
I tried a new .conf file, reverting to stock clocks, different video cards, etc. And this has been happening on all of my rigs, not just one.
December 12, 2012, 01:45:43 AM
I upgraded to Win8 this past weekend, but haven't been mining since. I DL 2.9.7 last night, and tried it with my Single. It froze about 30 min after I started it. I restarted it this morning before I left for work, and now that worker is already offline, so I'm assuming it did it again.
I'm running a .bat file with a loop to restart if it freezes, but it's weird cuz nothing happens when it freezes. It just stops, it doesn't exit, give any warning messages, error codes, etc.
I'm assuming this has something to do with Win8 rather than cgminer, but any ideas?
Did you download 2.9.7 or 2.9.7-1? The latter has a Windows crashing bug fixed.I'm running a .bat file with a loop to restart if it freezes, but it's weird cuz nothing happens when it freezes. It just stops, it doesn't exit, give any warning messages, error codes, etc.
I'm assuming this has something to do with Win8 rather than cgminer, but any ideas?
I've tried removing anything having to do with a GPU from my config file, and I realized that while I was using ozcoin's servers, one of my backkup pools was a direct IP for the ozcoin server. They had dns issues a while back, so I set this. Could having a backup pool be the same pool as your main pool cause any issues, if one is set to a typical name, and the other is a direct IP?
December 12, 2012, 12:42:08 AM
...
cgminer times how long it takes for the gpu code to "return" after it's been given work. Unfortunately with windows, the timer resolution is so shithouse at 15ms that I have to sample many many iterations of the GPU code and then average them to see how long it took.
...
cgminer times how long it takes for the gpu code to "return" after it's been given work. Unfortunately with windows, the timer resolution is so shithouse at 15ms that I have to sample many many iterations of the GPU code and then average them to see how long it took.
...
cklovias, there are several ways to overcome this drawback of the simple windows time functions. Probably the two easiest ways are using the code from this explanation http://msdn.microsoft.com/en-us/magazine/cc163996.aspx, or by using the .Net stopwatch class (for your code, you would have to call into a c++/cli dll wrapper around the class, then you have to deal with people needing the .Net framework installed too...maybe make it optional?)
I like using dynamic on my main windows computer, but I have been sticking with 2.7.5 because it has better dynamic behaviour for me. I have just been thinking, 'oh well, make it until the ASICs come out and that's good enough!'
December 12, 2012, 12:28:47 AM
...
cgminer times how long it takes for the gpu code to "return" after it's been given work. Unfortunately with windows, the timer resolution is so shithouse at 15ms that I have to sample many many iterations of the GPU code and then average them to see how long it took.
...
cgminer times how long it takes for the gpu code to "return" after it's been given work. Unfortunately with windows, the timer resolution is so shithouse at 15ms that I have to sample many many iterations of the GPU code and then average them to see how long it took.
...
cklovias, there are several ways to overcome this drawback of the simple windows time functions. Probably the two easiest ways are using the code from this explanation http://msdn.microsoft.com/en-us/magazine/cc163996.aspx, or by using the .Net stopwatch class (for your code, you would have to call into a c++/cli dll wrapper around the class, then you have to deal with people needing the .Net framework installed too...maybe make it optional?)
I like using dynamic on my main windows computer, but I have been sticking with 2.7.5 because it has better dynamic behaviour for me. I have just been thinking, 'oh well, make it until the ASICs come out and that's good enough!'
This is where I've found more useful info (than msdn) to solve the problem - since MS themselves really don't seem to care, it seems pointless looking at msdn
http://www.geisswerks.com/ryan/FAQS/timing.html
December 11, 2012, 11:29:40 PM
upgraded to 2.10.0 on my BAMT
in my config was I suppose to switch
to something else now? Im guessing this might have changed since 2.6.1 
in my config was I suppose to switch
Code:
"kernel" : "scrypt,scrypt,scrypt,scrypt",
December 11, 2012, 10:16:22 PM
...
cgminer times how long it takes for the gpu code to "return" after it's been given work. Unfortunately with windows, the timer resolution is so shithouse at 15ms that I have to sample many many iterations of the GPU code and then average them to see how long it took.
...
cgminer times how long it takes for the gpu code to "return" after it's been given work. Unfortunately with windows, the timer resolution is so shithouse at 15ms that I have to sample many many iterations of the GPU code and then average them to see how long it took.
...
cklovias, there are several ways to overcome this drawback of the simple windows time functions. Probably the two easiest ways are using the code from this explanation http://msdn.microsoft.com/en-us/magazine/cc163996.aspx, or by using the .Net stopwatch class (for your code, you would have to call into a c++/cli dll wrapper around the class, then you have to deal with people needing the .Net framework installed too...maybe make it optional?)
I like using dynamic on my main windows computer, but I have been sticking with 2.7.5 because it has better dynamic behaviour for me. I have just been thinking, 'oh well, make it until the ASICs come out and that's good enough!'
December 11, 2012, 08:32:22 PM
why would i get a "all devices disabled, cannot mine" error?
win 7 geforce gtx 260
installed amd app sdk v2.8 windows 64 bit, ati 11.12
thanks
win 7 geforce gtx 260
installed amd app sdk v2.8 windows 64 bit, ati 11.12
thanks
Well helps if you use the right drivers geforce uses nvidia drivers not ati, btw those cards suck at mining.
December 11, 2012, 08:20:43 PM
why would i get a "all devices disabled, cannot mine" error?
win 7 geforce gtx 260
installed amd app sdk v2.8 windows 64 bit, ati 11.12
thanks
win 7 geforce gtx 260
installed amd app sdk v2.8 windows 64 bit, ati 11.12
thanks
December 11, 2012, 02:40:54 PM
It is intended, when set "failover" but not "failover only" that work is split to 2 pools? I have set 3 backup p2pool nodes in my config and I just realized that half of my work is sent to another node O_o. My node is fast enough to keep 170MH worker ;] ... lol
See my worker "toy.gpu"
http://rav3n.dtdns.net:9332/static/graphs.html?Hour
and backup 1KurD1za... on
http://bitpoppool.geekgalaxy.com:9332/static/graphs.html?Day
See my worker "toy.gpu"
http://rav3n.dtdns.net:9332/static/graphs.html?Hour
and backup 1KurD1za... on
http://bitpoppool.geekgalaxy.com:9332/static/graphs.html?Day
December 11, 2012, 11:46:35 AM
I upgraded to Win8 this past weekend, but haven't been mining since. I DL 2.9.7 last night, and tried it with my Single. It froze about 30 min after I started it. I restarted it this morning before I left for work, and now that worker is already offline, so I'm assuming it did it again.
I'm running a .bat file with a loop to restart if it freezes, but it's weird cuz nothing happens when it freezes. It just stops, it doesn't exit, give any warning messages, error codes, etc.
I'm assuming this has something to do with Win8 rather than cgminer, but any ideas?
Did you download 2.9.7 or 2.9.7-1? The latter has a Windows crashing bug fixed.I'm running a .bat file with a loop to restart if it freezes, but it's weird cuz nothing happens when it freezes. It just stops, it doesn't exit, give any warning messages, error codes, etc.
I'm assuming this has something to do with Win8 rather than cgminer, but any ideas?
December 11, 2012, 11:32:08 AM
I upgraded to Win8 this past weekend, but haven't been mining since. I DL 2.9.7 last night, and tried it with my Single. It froze about 30 min after I started it. I restarted it this morning before I left for work, and now that worker is already offline, so I'm assuming it did it again.
I'm running a .bat file with a loop to restart if it freezes, but it's weird cuz nothing happens when it freezes. It just stops, it doesn't exit, give any warning messages, error codes, etc.
I'm assuming this has something to do with Win8 rather than cgminer, but any ideas?
I'm running a .bat file with a loop to restart if it freezes, but it's weird cuz nothing happens when it freezes. It just stops, it doesn't exit, give any warning messages, error codes, etc.
I'm assuming this has something to do with Win8 rather than cgminer, but any ideas?
Did you download 2.9.7 or 2.9.7-1? The latter has a Windows crashing bug fixed.
December 11, 2012, 11:26:57 AM
I upgraded to Win8 this past weekend, but haven't been mining since. I DL 2.9.7 last night, and tried it with my Single. It froze about 30 min after I started it. I restarted it this morning before I left for work, and now that worker is already offline, so I'm assuming it did it again.
I'm running a .bat file with a loop to restart if it freezes, but it's weird cuz nothing happens when it freezes. It just stops, it doesn't exit, give any warning messages, error codes, etc.
I'm assuming this has something to do with Win8 rather than cgminer, but any ideas?
I'm running a .bat file with a loop to restart if it freezes, but it's weird cuz nothing happens when it freezes. It just stops, it doesn't exit, give any warning messages, error codes, etc.
I'm assuming this has something to do with Win8 rather than cgminer, but any ideas?
December 11, 2012, 04:36:30 AM
A couple of people who are usually in the #cgminer IRC (as per the commit names) are the origin of the latest ztex code changes.
One is also the author of the libusb code.
Since you dropped ztex and left it hanging, it's been nice that others have taken an interest in working on it.
If you have any help for them, then come visit #cgminer and discuss with them.
One is also the author of the libusb code.
Since you dropped ztex and left it hanging, it's been nice that others have taken an interest in working on it.
If you have any help for them, then come visit #cgminer and discuss with them.
Ok, so I 'left it hanging'
nice... still keeping it civil, I see.I don't think this is ztex specific, but I humbly appreciate the invitation.
December 11, 2012, 04:17:38 AM
I gave 2.9.10 a quick try yesterday on OpenWRT (MIPSBE) and when mining with ztex fpgas on deepbit the hashrate drops to 1/5 of what it should be. On bitminter I would dare say it is faster than it was on 2.9.7.
Going back to 2.9.7 on deepbit works fine. It might be another endianess issue, I have not debugged this yet, but just thought I'd report in case someone gets a 'doh!' moment and immediately knows what's wrong
A couple of people who are usually in the #cgminer IRC (as per the commit names) are the origin of the latest ztex code changes.Going back to 2.9.7 on deepbit works fine. It might be another endianess issue, I have not debugged this yet, but just thought I'd report in case someone gets a 'doh!' moment and immediately knows what's wrong
One is also the author of the libusb code.
Since you dropped ztex and left it hanging, it's been nice that others have taken an interest in working on it.
If you have any help for them, then come visit #cgminer and discuss with them.
December 11, 2012, 04:09:12 AM
I gave 2.9.10 a quick try yesterday on OpenWRT (MIPSBE) and when mining with ztex fpgas on deepbit the hashrate drops to 1/5 of what it should be. On bitminter I would dare say it is faster than it was on 2.9.7.
Going back to 2.9.7 on deepbit works fine. It might be another endianess issue, I have not debugged this yet, but just thought I'd report in case someone gets a 'doh!' moment and immediately knows what's wrong
Going back to 2.9.7 on deepbit works fine. It might be another endianess issue, I have not debugged this yet, but just thought I'd report in case someone gets a 'doh!' moment and immediately knows what's wrong
December 11, 2012, 02:21:59 AM
Having trouble compiling it for mips.
I updated curl, as It was missing some references, but now it's missing even more with version 7.28.1.
Not sure what's going on.
EDIT:
So it seems that -lcurl isn't being included!
I updated curl, as It was missing some references, but now it's missing even more with version 7.28.1.
Not sure what's going on.
EDIT:
So it seems that -lcurl isn't being included!
December 10, 2012, 07:37:54 PM
cgminer times how long it takes for the gpu code to "return" after it's been given work. Unfortunately with windows, the timer resolution is so shithouse at 15ms that I have to sample many many iterations of the GPU code and then average them to see how long it took. This means that between giving the GPU code, there are short periods dependent on CPU code, and windows' scheduler does not seem to give this CPU code priority even though folding@home should be idle priority. So it ends up counting delays of the CPU scheduling as though the GPU had spent more time on its work. So it's a combination of the shit windows timer resolution and the shit windows scheduling that makes dynamic not work that well for your test case.
No prob. I can live with it, just curious.
so, best senario: linux!
'cept the gaming part, which keeps me in windows. ah well..I LOVE the stratum support! dunno how you do it but cgminer just keeps getting better!
Quote from: ckolivas
Thanks 
youre welcome. donation coming your way, check your PM. make sure it goes onto the stratum side of the scoreboard if youre still keeping count
December 10, 2012, 07:36:35 PM
On the new version 2.10.0 My BFL boxes all are showing the same MHash rates, but all my 7970's have gone from 680Mhash to 603MHash. Any thoughts?
My 7970s are performing exactly the same with 2.10 as with 2.9. Coincidence related to some other problem for you perhaps?I figured it out. Drivers! They crashed when i first started CGMiner for some reason, so I had to reset all the clocks. Thanks for the replies.
December 10, 2012, 07:18:27 PM
question on how cgminer figures intensity.
on dynamic with both 2.9.7.0/1 and 2.10.0 when I fire up folding@home on the cpu (which loads the 4 cores to 100%) cgminer drops intensity down to -8 or so on my 6870. stop folding and intensity goes to 4 or 5 where it should be.
cgminer times how long it takes for the gpu code to "return" after it's been given work. Unfortunately with windows, the timer resolution is so shithouse at 15ms that I have to sample many many iterations of the GPU code and then average them to see how long it took. This means that between giving the GPU code, there are short periods dependent on CPU code, and windows' scheduler does not seem to give this CPU code priority even though folding@home should be idle priority. So it ends up counting delays of the CPU scheduling as though the GPU had spent more time on its work. So it's a combination of the shit windows timer resolution and the shit windows scheduling that makes dynamic not work that well for your test case.on dynamic with both 2.9.7.0/1 and 2.10.0 when I fire up folding@home on the cpu (which loads the 4 cores to 100%) cgminer drops intensity down to -8 or so on my 6870. stop folding and intensity goes to 4 or 5 where it should be.
I LOVE the stratum support! dunno how you do it but cgminer just keeps getting better!
Thanks December 10, 2012, 07:08:33 PM
question on how cgminer figures intensity.
on dynamic with both 2.9.7.0/1 and 2.10.0 when I fire up folding@home on the cpu (which loads the 4 cores to 100%) cgminer drops intensity down to -8 or so on my 6870. stop folding and intensity goes to 4 or 5 where it should be.
dunno if previous versions of cgminer do this as I just upgraded to an i5 3570k; I didnt fold on the other CPU.
AFAIK folding does not touch the GPU in any way when not displaying folding graphics (which i never do) and it runs @ idle priority.
does cgminer figure CPU load into dynamic intensity?
win7 64 bit
cgminer 2.10.0
11.11 driver
12.4 SDK
HD6870
anyhow, setting "I" manually is fine. and I LOVE the stratum support! dunno how you do it but cgminer just keeps getting better!
on dynamic with both 2.9.7.0/1 and 2.10.0 when I fire up folding@home on the cpu (which loads the 4 cores to 100%) cgminer drops intensity down to -8 or so on my 6870. stop folding and intensity goes to 4 or 5 where it should be.
dunno if previous versions of cgminer do this as I just upgraded to an i5 3570k; I didnt fold on the other CPU.
AFAIK folding does not touch the GPU in any way when not displaying folding graphics (which i never do) and it runs @ idle priority.
does cgminer figure CPU load into dynamic intensity?
win7 64 bit
cgminer 2.10.0
11.11 driver
12.4 SDK
HD6870
anyhow, setting "I" manually is fine. and I LOVE the stratum support! dunno how you do it but cgminer just keeps getting better!
Jump to: