Author

Topic: OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.0 - page 762. (Read 5805728 times)

donator
Activity: 1218
Merit: 1079
Gerald Davis
The 5870 instant jumped to the max which caused a gpu error screen Cheesy

So i lowered the setting to one which i knew did function some time ago with very low memory speeds ( 920 core 450 memory )

If you want to save some power you likely can get stable @ 200 MHz memclock for the 5870s.  Yet another area the 5xxx series is superior to the 6xxx is their tolerance of much lower memory.  Kill-a-watt indicates about a 10 watt savings between 200 and 400 MHz.
c_k
donator
Activity: 242
Merit: 100
Code:
[2011-10-10 07:26:25] Failed to init GPU thread 0, disabling device 0
[2011-10-10 07:26:25] Restarting the GPU from the menu is unlikely to fix this.
[2011-10-10 07:26:25] Try stopping other applications using the GPU like afterburner.
[2011-10-10 07:26:25] Then restart cgminer.

I get this when ever I start cgminer on a machine with 7 cards

It works fine on other machines with 1 or 2 cards

I have no other GPU apps running on the machine that it won't work on

Switching back to 2.0.5 works fine on the machine that it won't work on
hero member
Activity: 774
Merit: 500
Lazy Lurker Reads Alot
Solved it myself and no Sekioh sorry it was just not working with older drivers.

I was running catalyst 10.10 driver so today i downloaded 11.9 since i needed the support for a game issue

And guess what, poof the --gpu-engine 850-950 function worked on both my ati rigs.

The 5870 instant jumped to the max which caused a gpu error screen Cheesy

So i lowered the setting to one which i knew did function some time ago with very low memory speeds ( 920 core 450 memory )

And my 4770 shows a massive increase when i set it to run at 910 - 400 it even seem to run fine with 950 but lol i am not sure if thats wise xD
But for such a slow card it really does well doing up to 97 MH/s

I tried with poclbm but it does not come near the performance of cgminer, just a little more stales popping up.
And somehow pocblm lost the connect with the server which i not have seen on cgminer yet

Still leaves a question on the side of developer when a core error appears should the program not try to lower the clocks the next time it starts/restarts ?


 

sr. member
Activity: 435
Merit: 250
Can not compile cgminer-2.0.5:

Code:
make[2]: Entering directory `/home/administrativo/cgminer-2.0.5'
  CC     cgminer-main.o
main.c: In function ‘main’:
main.c:5418:12: error: ‘struct cgpu_info’ has no member named ‘gpu_engine’
main.c:5419:12: error: ‘struct cgpu_info’ has no member named ‘gpu_memclock’
main.c:5420:12: error: ‘struct cgpu_info’ has no member named ‘gpu_vddc’
main.c:5421:12: error: ‘struct cgpu_info’ has no member named ‘gpu_fan’
main.c:5422:12: error: ‘struct cgpu_info’ has no member named ‘gpu_powertune’
make[2]: *** [cgminer-main.o] Error 1
make[2]: Leaving directory `/home/administrativo/cgminer-2.0.5'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/administrativo/cgminer-2.0.5'
make: *** [all] Error 2

Ubuntu 11.04 32 bits.

CGMiner 2.0.4 compiles just fine.

What can I do?!

Thanks!
Thiago

Still no solution?
It's fixed in git.

Indeed, it is.
Cheers, thanks for the info.
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
Got the "not found" error when trying to download the Windoze version.
Sam
Link fixed, thanks.
legendary
Activity: 3583
Merit: 1094
Think for yourself
Got the "not found" error when trying to download the Windoze version.
Sam
-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
Thanks to continued donations  Grin Much appreciated.

New version: 2.0.6
- Expire shares as stale with a separate timeout from the scantime, defaulting
to 120 seconds.
- Retry pools after a delay of 15 seconds if none can be contacted on startup
unless a key is pressed.
- Don't try to build adl features without having adl.
- Properly check shares against target difficulty - This will no longer show
shares when solo mining at all unless they're considered to be a block solve.
- Add altivec 4 way (cpu mining) support courtesy of Gilles Risch.
- Try to use SSL if the server supports it.
- Display the total solved blocks on exit (LOL if you're lucky).
- Use ADL activity report to tell us if a sick GPU is still busy suggesting it
is hard hung and do not attempt to restart it.
- Don't make donation work interfere with block change detection allowing
donation to work regardless of the block chain we're mining on.
full member
Activity: 181
Merit: 100
This far i think its a nice program but the auto-engine seems to fail
Manually had my 5870 set to its basic clocks at 850 core 1200 memory
When i add the parameters for core 850-950 and memory 450 nothing happens when i watch over time with gpu-z it keeps running at the speed it started with so no change at all
Now i certain this card can run at 890 Mhz but i have not seen any increase... maybe i am missing something here


I had issues getting my cards to clock at all until a little bit ago so haven't tested, but from what I read you manually set the cards to the HIGH side, and the range will clock DOWN based on throttling.
hero member
Activity: 774
Merit: 500
Lazy Lurker Reads Alot
This far i think its a nice program but the auto-engine seems to fail
Manually had my 5870 set to its basic clocks at 850 core 1200 memory
When i add the parameters for core 850-950 and memory 450 nothing happens when i watch over time with gpu-z it keeps running at the speed it started with so no change at all
Now i certain this card can run at 890 Mhz but i have not seen any increase... maybe i am missing something here
legendary
Activity: 2576
Merit: 1186
Can not compile cgminer-2.0.5:

Code:
make[2]: Entering directory `/home/administrativo/cgminer-2.0.5'
  CC     cgminer-main.o
main.c: In function ‘main’:
main.c:5418:12: error: ‘struct cgpu_info’ has no member named ‘gpu_engine’
main.c:5419:12: error: ‘struct cgpu_info’ has no member named ‘gpu_memclock’
main.c:5420:12: error: ‘struct cgpu_info’ has no member named ‘gpu_vddc’
main.c:5421:12: error: ‘struct cgpu_info’ has no member named ‘gpu_fan’
main.c:5422:12: error: ‘struct cgpu_info’ has no member named ‘gpu_powertune’
make[2]: *** [cgminer-main.o] Error 1
make[2]: Leaving directory `/home/administrativo/cgminer-2.0.5'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/administrativo/cgminer-2.0.5'
make: *** [all] Error 2

Ubuntu 11.04 32 bits.

CGMiner 2.0.4 compiles just fine.

What can I do?!

Thanks!
Thiago

Still no solution?
It's fixed in git.
sr. member
Activity: 435
Merit: 250
Can not compile cgminer-2.0.5:

Code:
make[2]: Entering directory `/home/administrativo/cgminer-2.0.5'
  CC     cgminer-main.o
main.c: In function ‘main’:
main.c:5418:12: error: ‘struct cgpu_info’ has no member named ‘gpu_engine’
main.c:5419:12: error: ‘struct cgpu_info’ has no member named ‘gpu_memclock’
main.c:5420:12: error: ‘struct cgpu_info’ has no member named ‘gpu_vddc’
main.c:5421:12: error: ‘struct cgpu_info’ has no member named ‘gpu_fan’
main.c:5422:12: error: ‘struct cgpu_info’ has no member named ‘gpu_powertune’
make[2]: *** [cgminer-main.o] Error 1
make[2]: Leaving directory `/home/administrativo/cgminer-2.0.5'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/administrativo/cgminer-2.0.5'
make: *** [all] Error 2

Ubuntu 11.04 32 bits.

CGMiner 2.0.4 compiles just fine.

What can I do?!

Thanks!
Thiago

Still no solution?

sr. member
Activity: 378
Merit: 250
I tried
Code:
-d 0 -v 2 -w 128 --auto-fan --auto-gpu --gpu-fan 25-50 --gpu-engine 600-900 --gpu-memclock 350 --temp-target 60
for my HD5870, which is my 1st card, which works, but the 2nd card gets the fan ramped up to 85% even if it is disabled. What about this? Bug or a stupid config error from me?

By the way, is there a chance to reset the cards values to default if clicking the close button on Windows? CGMINER only resets the values, if I press Q.

Thanks,
Dia

I don't think that's possible honestly.  Clicking the X on a Dos window just flat out kills the process, so there's no way the program will be able to reset things.

You can create a second batch file to launch your miner again with all the defaults or just make sure you press Q Smiley
hero member
Activity: 772
Merit: 500
I tried
Code:
-d 0 -v 2 -w 128 --auto-fan --auto-gpu --gpu-fan 25-50 --gpu-engine 600-900 --gpu-memclock 350 --temp-target 60
for my HD5870, which is my 1st card, which works, but the 2nd card gets the fan ramped up to 85% even if it is disabled. What about this? Bug or a stupid config error from me?

By the way, is there a chance to reset the cards values to default if clicking the close button on Windows? CGMINER only resets the values, if I press Q.

Thanks,
Dia
hero member
Activity: 742
Merit: 500
I wish I had found this program weeks ago.  I get more hash/s than phoenix and all my GPUs are setup with one command!

Thanks for your great work!
sr. member
Activity: 404
Merit: 250
I just want to say thanks for all the hard work! I just donated a bit more than 4 btc. I don't have a request, only thanks.
c_k
donator
Activity: 242
Merit: 100
2 request for features please, since we've been donating a while:

1.) SSL support (to the extent of what bitcoind offers)
2.) SOCKS proxy support (to the same extent bitcoind supports)

Cheers!  Cheesy
legendary
Activity: 4634
Merit: 1851
Linux since 1997 RedHat 4
...
second, can you help me or point me up to instructions to install CGMINER on Ubuntu 11 x64?
Try my xubuntu 11.04 script in the main directory ... linux-usb-cgminer
Start from step 6)
donator
Activity: 446
Merit: 262
Interesting.
Hello again, i have 2 questions:

I have a mixed rig, with
-GPU0:5850
-GPU1:5870
-GPU2:6950
-GPU3:5850
-GPU4:6950

when running: cgminer -o url -u user -p pass -w 256 -v 2 -I 8,8,9,8,9 --gpu-engine 930,930,930,930,930 --gpu-memclock 300 --donation 1

Only the first 2 mine correcly, the other 3 kind of tries to mine from gpu0 and gpu1. i think gpu2, gpu3 and gp4 are getting wrongly detected.
Maybe it's something to do with the 5800 series being detected 2 devices, and the 6900 series being detected 3? (bad english, hope you understand)

EDIT: removed -w 256 and -v 2 and it looks fine now.

second, can you help me or point me up to instructions to install CGMINER on Ubuntu 11 x64?
sr. member
Activity: 278
Merit: 250
Can you copy and paste your AMDOverDriver configuration file, with overvolt, for your 5970?!

I've never been able to overvolt with AMDOC, sorry if that wasn't clear.  I was saying you could potentially do better if *you* have the ability to overvolt (using a different tool like Trixx or Afterburner).

I do know that running the fans at 100% is risky.  I've siezed up four of them on my 5850's (my 5970s are all fine so far).  I just remove the fan, lay the card down and put a std 80mm fan over the hole, problem solved.  If all my fans melt down it'll be a problem, but I needed the extra air flow over the summer.  With colder weather coming I'll probably try reducing them all in a few weeks.

Here is the AMDOverdriveCtrl file for one half of one 5970:

Code:


 
 
 
 
 
 
 
 
 
 
 
 
 
 


IIRC, I just took the default file and tweaked the gpu and mem speeds... when I tried the voltage AMDOC hung because it couldn't increase it.
legendary
Activity: 1862
Merit: 1011
Reverse engineer from time to time
So, i tried compiling on mingw32 and

Quote
In file included from sigaction.c:21:0:
./signal.h:616:3: error: expected specifier-qualifier-list before 'uid_t'
./signal.h:659:1: error: conflicting types for 'restrict'
./signal.h:659:1: note: previous definition of 'restrict' was here
sigaction.c:130:58: error: expected ';', ',' or ')' before 'act'
sigaction.c:84:1: warning: 'sigaction_handler' defined but not used

This on windows. The configure i use is

Quote
CFLAGS="-g -O3 -Wall -msse2 -march=amdfam10 -msse4a -mtune=amdfam10 -mabm -co
mbine -funroll-all-loops -ffast-math -fprefetch-loop-arrays -ftree-parallelize-
loops=4 -IC:/APP/include -IC:\MinGW\msys\1.0\home\curl\include\curl" LDFLAGS="-
LC:/APP/lib/x86 -LC:\MinGW\msys\1.0\local\lib\libcurl.a" ./configure
and then just make and i get that error.
Jump to: