CCminer X11 mining on NVidia cards

[flare]

I don't think I was able to install the cuda driver correctly as got this issue after the following command.
user@user-System-Product-Name ~ $ sudo apt-get install libcurl4-openssl-dev autoconf git screen libncurses5-dev libdb4.8-dev -y
Reading package lists... Done
Building dependency tree
Reading state information... Done
Package libdb4.8-dev is not available, but is referred to by another package.
This may mean that the package is missing, has been obsoleted, or
is only available from another source

E: Package 'libdb4.8-dev' has no installation candidate

find / | grep nvcc still brings back permission denied so im assuming I didn't install this properly, which would explain why I get more errors and stuff when trying to start the program. Sorry I know your instructions were prob bang on but I've been a windows user my whole life and may have to just dual boot because a week into linux and I still can't install a program like this. I really appreciate the help though.


edit: I don't even know how to what to do to start the miner if it was installed properly. My best guess would be
LD_LIBRARY_PATH=/usr/local/cuda-5.5/lib64 ./ccminer -a X11 -o pool -u worker -p pass -B
but that's because I'm copying what I use to start my cpuminer pretty much and I don't know if it's supposed to be an "a" before X11 or another letter. I'm just completely lost.

Did you continue after the libdb4.8-dev error? Actually this package is not needed if only running a miner and it can be removed safely from the commandline, so just

sudo apt-get update
sudo apt-get install freeglut3-dev build-essential libx11-dev libxmu-dev libxi-dev libgl1-mesa-glx libglu1-mesa libglu1-mesa-dev gcc make -y
sudo apt-get install libcurl4-openssl-dev autoconf git screen libncurses5-dev -y
sudo apt-get install linux-headers-$(uname -r) -y
wget http://developer.download.nvidia.com/compute/cuda/5_5/rel/installers/cuda_5.5.22_linux_64.run
sudo sh cuda_5.5.22_linux_64.run -silent -driver -toolkit

Please check /usr/local directory after the full procedure for existence of a cuda directory

$ ls -l /usr/local

 

[ROYALEwitCHEESE]

nothing happens when I type in
sudo sh cuda_5.5.22_linux_64.run -silent -driver -toolkit
and there's nothing in the usr/local directory except for the usual folders.

 

[flare]

nothing happens when I type in
sudo sh cuda_5.5.22_linux_64.run -silent -driver -toolkit
and there's nothing in the usr/local directory except for the usual folders.

Try to run

sudo sh cuda_5.5.22_linux_64.run -driver -toolkit

to get some output from the installer.

 

[flare]

nothing happens when I type in
sudo sh cuda_5.5.22_linux_64.run -silent -driver -toolkit
and there's nothing in the usr/local directory except for the usual folders.

Or are you on 32bit Linux MINT? Then you need to run the 32bit version of CUDA installer instead :smile:

Try

wget http://developer.download.nvidia.com/compute/cuda/5_5/rel/installers/cuda_5.5.22_linux_32.run
sudo sh cuda_5.5.22_linux_32.run -silent -driver -toolkit

 

[ROYALEwitCHEESE]

sudo sh cuda_5.5.22_linux_64.run -driver -toolkit
brough up some info on the toolkit so I think it's installed and I do have 64 bit. Thanks again I'll be back shortly just have to walk my dog.

 

[flare]

sudo sh cuda_5.5.22_linux_64.run -driver -toolkit
brough up some info on the toolkit so I think it's installed and I do have 64 bit. Thanks again I'll be back shortly just have to walk my dog.

Perhaps this info is revealing where nvcc was installed :smile:

 

[ROYALEwitCHEESE]

I'm confused and feel like an idiot

edit: it says
/tmp/cuda_install_14677.log
so nvcc installed to tmp? I should go back to windows if its this hard for me to install a program lol.

edit again: I just realized that it brought up tonnes of stuff and it mentions user/local again but this seems to all be over my head I got linux because I don't do a whole lot on my computer but I would like to mine darkcoins and it looks like I'll have to wipe it all and go back to windows7.

 

[ROYALEwitCHEESE]

Im sorry to be pessimistic it's just that it starts to get infuriating when you've spent 5+ hours trying to mine dark coins....

 

[flare]

I'm confused and feel like an idiot

edit: it says
/tmp/cuda_install_14677.log
so nvcc installed to tmp? I should go back to windows if its this hard for me to install a program lol.

edit again: I just realized that it brought up tonnes of stuff and it mentions user/local again but this seems to all be over my head I got linux because I don't do a whole lot on my computer but I would like to mine darkcoins and it looks like I'll have to wipe it all and go back to windows7.

Never mind - btw: have you ever tried to run the windows distribution file of ccminer? It's ready compiled and runs out of the box on my windows 8 Laptop

No need for Linux at all when mining...

 

[flare]

t looks like I'll have to wipe it all and go back to windows7.

If you want to revert to Windows 7 now: here is the ccminer binary for Windows. https://github.com/cbuchner1/ccminer/releases/download/v1.2/ccminer-v1.2.zip

If you need further assistance let me know :smile:

 

[ROYALEwitCHEESE]

Is there a guide anywhere as I don't know where to find the cpu miner for windows as I can only find linux guides on installing darkcoin-cpuminer-1.3-avx-aes-master is this the miner I want for windows as well?

downloaded the cpuminr 1.3 and I think I should be using minerd-core-avx-i but I don't know where the config file is to set it up.

 

[flare]

here is the ccminer binary for Windows. https://github.com/cbuchner1/ccminer/releases/download/v1.2/ccminer-v1.2.zip

Is there a guide anywhere as I don't know where to find the cpu miner for windows as I can only find linux guides on installing darkcoin-cpuminer-1.3-avx-aes-master is this the miner I want for windows as well?

downloaded the cpuminr 1.3 and I think I should be using minerd-core-avx-i but I don't know where the config file is to set it up.

Actually i posted the link to windows ccminer in my last post

 

[ROYALEwitCHEESE]

Actually i posted the link to windows ccminer in my last post

I'd use the ccminer for cpu mining as well or ccminer for gpu and cpuminr 1.3 for the cpu?

 

[flare]

I'd use the ccminer for cpu mining as well or ccminer for gpu and cpuminr 1.3 for the cpu?

ccminer: x11 gpu miner for nVidia cards
cpuminer: x11 cpu miner

 

[ROYALEwitCHEESE]

All looks to be good getting ~2800 khash/s with my 680 and ~100 khash per core on my 3570k thanks for the help flare.

 

[ROYALEwitCHEESE]

All looks to be good getting ~2800 khash/s with my 680 and ~100 khash per core on my 3570k thanks for the help flare.

I'm actually getting an unstable hash rate on my gpu it's not staying at 2800 and dipping as low as 400 is that something to do with my setup potentially? edit: sorry I'm a complete newbie to mining.

 

[ROYALEwitCHEESE]

Nevermind I didn't have percision installed yet as this windows installation is fresh and once I set an aggressive fan curve and got the temps down my hash rate went back up.

 

[flare]

I'm actually getting an unstable hash rate on my gpu it's not staying at 2800 and dipping as low as 400 is that something to do with my setup potentially? edit: sorry I'm a complete newbie to mining.

There are a lot of tweaks you can do with the parameters, but thats were it's becoming very ccminer and graphic card specific. Maybe ask in the official ccminer forum --> http://www.cudaminers.net/forum/ccminer-discussions-forum.html

 

[erikasas]

Hello, So i am kinda new to mining, and decided to give it a go.

The problem is that my two gpu's (i know they are old) are giving me ridiculously high khash, but the other one gpu, is working properly. Can anybody help me fix this or tell me whats the problem ?
Also it is not giving me any real khash.

Here is what i did, reinstalled the drivers, and installed cuda.

Here is what the miner shows me:

[2014-07-18 18:35:35] Binding thread 1 to cpu 1
[2014-07-18 18:35:35] 3 miner threads started, using 'x11' algorithm.
[2014-07-18 18:35:35] Binding thread 0 to cpu 0
[2014-07-18 18:35:35] Binding thread 2 to cpu 2
[2014-07-18 18:35:36] GPU #1: GeForce 210, 9279 khash/s
[2014-07-18 18:35:36] GPU #2: GeForce 8500 GT, 7943 khash/s
[2014-07-18 18:35:36] GPU #0: GeForce GTX 660, 1413 khash/s
[2014-07-18 18:35:37] GPU #2: GeForce 8500 GT, 365015 khash/s
[2014-07-18 18:35:37] GPU #1: GeForce 210, 362475 khash/s
[2014-07-18 18:35:39] GPU #1: GeForce 210, 448174 khash/s
[2014-07-18 18:35:42] GPU #1: GeForce 210, 462750 khash/s
[2014-07-18 18:35:45] GPU #1: GeForce 210, 563015 khash/s
[2014-07-18 18:35:47] GPU #1: GeForce 210, 564347 khash/s
[2014-07-18 18:35:50] GPU #1: GeForce 210, 669042 khash/s
[2014-07-18 18:35:52] GPU #1: GeForce 210, 666860 khash/s
[2014-07-18 18:35:55] GPU #1: GeForce 210, 507173 khash/s
[2014-07-18 18:35:57] GPU #0: GeForce GTX 660, 1836 khash/s
[2014-07-18 18:35:57] accepted: 1/1 (100.00%), 874024 khash/s (yay!!!)
[2014-07-18 18:35:57] GPU #1: GeForce 210, 582960 khash/s
[2014-07-18 18:35:59] GPU #1: GeForce 210, 634078 khash/s
[2014-07-18 18:36:02] GPU #1: GeForce 210, 576853 khash/s
[2014-07-18 18:36:05] GPU #1: GeForce 210, 499389 khash/s

 

[ROYALEwitCHEESE]

I'm not positive but I can tell you this happens to me when the card driver has crashed. The gpu usage goes down and the hash rate goes up to number like the ones that you listed. It could be a voltage/heat/driver issue IMO but I'm not sure.

Edit: Just realized some of your cards are probably too old as it it has to support at least Cuda 2.0 I believe and it probably makes sense then why only your 660 is making realistic numbers.