CCminer X11 mining on NVidia cards

Lzeppelin

Member
Feb 27, 2014
283
57
88
Great news if true, it will boost the number of potential darkcoin miners. With more miners each coin will be harder to get and hopefully receive a boost in price.
 

yibble

Member
Apr 8, 2014
63
9
48
Absolutely! I've been unable to compile it so far, so I'll have another look tomorrow.

Code:
/home/-/build/ccminer/cpu-miner.c:903: undefined reference to `scanhash_nist5'
/home/-/build/ccminer/cpu-miner.c:908: undefined reference to `scanhash_x11'
collect2: ld returned 1 exit status
make[2]: *** [ccminer] Error 1
make[2]: Leaving directory `/home/-/build/ccminer'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/-/build/ccminer'
make: *** [all] Error 2
 

hu398

New Member
May 13, 2014
12
2
3
I'm getting 1.8mhash on P2Pool with an NVidia GTX 760 (ccMiner v1.0 windows binary). Apparently x11 hasn't been optimized in this release, so future releases of ccMiner might improve the has rate further. Either way it is a huge improvement on CPU only mining and Christian deserves credit for the work he's put in to allow NVidia owners to mine Darkcoin.
 
  • Like
Reactions: noruka and yibble

rumple

New Member
Apr 3, 2014
13
0
1
Build issue for me (64-bit 13.10):

Stack dump:
0. Running pass 'Live Variable Analysis' on function '@_Z24myriadgroestl_gpu_sha256Pj'
Segmentation fault (core dumped)
make[2]: *** [cuda_myriadgroestl.o] Error 139

---
cicc crashed with SIGSEGV in llvm::LiveVariables::HandleVirtRegUse()
reason: writing NULL VMA
with: nvidia-cuda-toolkit 5.0.35-7ubuntu1
 

noruka

New Member
Apr 2, 2014
15
1
3
I'm getting 1.8mhash on P2Pool with an NVidia GTX 760 (ccMiner v1.0 windows binary). Apparently x11 hasn't been optimized in this release, so future releases of ccMiner might improve the has rate further. Either way it is a huge improvement on CPU only mining and Christian deserves credit for the work he's put in to allow NVidia owners to mine Darkcoin.
anyone got a 750ti running x11?
 

aminkorea

New Member
May 23, 2014
5
0
1
Is there a windows binary for cuda anywhere to be found? I have 7 nVida 750Ti that I would like to start mining with but I am not a linux person.
 

aminkorea

New Member
May 23, 2014
5
0
1
I just install ccminer from the link you provided and when I start the program I get an error that says the miner can't start because MSVCR100.dll is missing from my computer. Any suggestions?

Edit: I googled the problem and installed the missing dll file but still get the same error. Can anyone offer some help?
 
Last edited by a moderator:

hu398

New Member
May 13, 2014
12
2
3
That's the Microsoft Visual C run time library v10, if you just downloaded the single dll file from somewhere then maybe try downloading the full runtime installer from microsoft and installing that. This dll file usually lives in the windows\system32 folder but would probably be picked up if you copied it to the ccminer x86 folder.
 

xavier04

New Member
May 24, 2014
1
0
1
Build issue for me (64-bit 13.10):

Stack dump:
0. Running pass 'Live Variable Analysis' on function '@_Z24myriadgroestl_gpu_sha256Pj'
Segmentation fault (core dumped)
make[2]: *** [cuda_myriadgroestl.o] Error 139

---
cicc crashed with SIGSEGV in llvm::LiveVariables::HandleVirtRegUse()
reason: writing NULL VMA
with: nvidia-cuda-toolkit 5.0.35-7ubuntu1
I have the exact same issue. Have you had any luck finding a solution?
 

aminkorea

New Member
May 23, 2014
5
0
1
That's the Microsoft Visual C run time library v10, if you just downloaded the single dll file from somewhere then maybe try downloading the full runtime installer from microsoft and installing that. This dll file usually lives in the windows\system32 folder but would probably be picked up if you copied it to the ccminer x86 folder.
I installed the MS Visual C and still get the same error.
 

aminkorea

New Member
May 23, 2014
5
0
1
Maybe try KBomba's build? It uses the most recent copy of the official ccMiner source (faster the the current official) and also includes an optional downgraded ccminer21 for older nvidia cards.
https://github.com/KBomba/ccminer-KBomba/releases/
I will give this a try, but my nVidia cards are not older, they are the new 750Ti Maxwell cards. Will post again once I have tried this version. Thanks for ur help.
 

Syntor

Member
May 26, 2014
60
7
48
New Orleans, LA
I am running (2) GeForce GTX 760's..
This is my string: ccminer30.exe -t 2 -a x11 -o stratum+tcp://mining.darkcointalk.org:3333 -u -p
This is my clocking: Power Limit: 84%, Core Clock(MHz): +192, Memory Clock(MHz) +192, Fan Speed: 93%
I am currently getting a VRel cap that is only letting me get @78% power% (not sure why I am getting that.. may need bigger power supply)
My temps are 63c on my Evga, and 77c on my Galaxy (doesn't have nearly the cooling fin size as the Evga)
I am currently getting @2040khash/s from the Evga and @2010khash/s on the Galaxy!!
I am pretty happy with the output!!
 

Burb

New Member
May 26, 2014
1
0
1
Hello,

I've been mining with ccminer on my GTX 780 over the past 3 or 4 days. However, it has stopped working.
It's been working fine over the weekend, getting around 2000 khash/s. But now, all I get is this error:
GPU #0: result for nonce $00000000 does not validate on CPU!
For reference, here is my .bat:
ccminer35.exe --algo=x11 -o stratum+tcp://212.227.91.190:7903 -u [wallet]+0.001 -p 123 --quiet
I've tried with two different pools, and I get the same error on both. Neither the GPU or the CPU is overclocked.
Using minerd-corei7-avx instead of ccminer works, so there is nothing wrong with the pools. However I only get around 500 khash/s with cpu mining.. I'd rather use ccminer.

I'm also getting artifacts on the windows desktop sometimes when I try to start up ccminer, little green or white dots appear at times then disappears. And sometimes when I try to start up ccminer the screen blacks out and I get a "nvidia driver crashed and has been restored error".

I've been using MSI Afterburner while mining this whole time and set the fan speed to 100%, so temperatures have been low.

All was working fine yesterday... Did my graphics card break or something?
 

Syntor

Member
May 26, 2014
60
7
48
New Orleans, LA
Following.. I get the same error maybe 3-4 times an hour, and just chalk it up to the fact that the program is just a tweak off of a cpu mining program..I just get that as a random error.. but the next line it goes right back to accepting..
 

scavenger7

New Member
May 25, 2014
17
7
3
KBomba's build with 2.1 is nice for the older nVidia cards like my GTS 450. I'm averaging about 100ish khash/s which is nicer than the 40-50ish from my CPU. Together it's nicer. definitely nice to see it working now.
 

dchoquette

New Member
May 29, 2014
1
0
1
Im having some trouble with ccminer. The program loads and is supposedly hashing but no shares are accepted/found and the pool doesn't detect the work even if it was working. Here is what is in my BAT file: "ccminer35.exe --algo=x11 -o stratum+tcp://mining.darkcointalk.org:3333 -u dchoquette.1 -p stuff"
I would really appreciate some help!
I gotta say this is really confusing. and I was excited to use my gpu again (evga gtx760)
 

aadje93

New Member
May 23, 2014
37
4
8
Do you see hw errors dchoquette?
I have overclocked my 780 a little and it has zero error with ccminer 1.0beta (not yet tryed the new version, saw some comments that it doesnt work that good)
 

Uleepera

New Member
May 30, 2014
7
0
1
Has anyone tried this? I started out with darkcoin the other night running two systems whit the cpuminer just to see how everything worked. Had zero issues so I'm moving over to one of my GPU setups. It seems gpu support for darkcoin is sketchy at best. I've tried sgminer (4.10) both regular and x11 versions with no luck as they just keep crashing.

I found in a different forum a suggestion to use ccminer (1.0) which is supposed to specifically support nvidia gpus. I'm using drkpool.com and in the cpuminer I get a message indicating when I've completed a hash (yay!!!) but with CC I don't see any indication when I've completed a hash nor does the drkpool website seem to register any progress either on a kh/sec or valid hash count. the start-up for seems easy "ccminer35.exe --algo=x11 -o stratum+tcp://drkpool.com:3333" plus -u/-p tags.

drkpool doesn't even seem to realize that ccminer is logged in although it doesn't seem to have any issues with logging on.

Any thoughts? I'd really like to get beyond 200ish kh/sec numbers.

Thank you for reading.
 

rumple

New Member
Apr 3, 2014
13
0
1
Maybe try KBomba's build? It uses the most recent copy of the official ccMiner source (faster the the current official) and also includes an optional downgraded ccminer21 for older nvidia cards.
https://github.com/KBomba/ccminer-KBomba/releases/
Thanks for the suggestion, but I get the same failure during the build. The make output shows it is using options compute_30 and compute_35, but no sign of anything like compute_21. Do I need to add some option to the "configure" command?
 

Uleepera

New Member
May 30, 2014
7
0
1
I appreciate the suggestion. Why is that pool better than the one I'm using?

I double checked and that is the same version of ccminer I've been trying to use.
 
Last edited by a moderator:

Syntor

Member
May 26, 2014
60
7
48
New Orleans, LA
Double check your ccminer version inside of 1.0, there are ccminer30, ccminer35, and ccminer50 depending on your card computing.. I use 2 GTX 760's and have to use ccminer30, I was getting the same thing you were getting before I switched.. I think the default is ccminer35, check your string.. Here's mine for my 2 cards:
ccminer30.exe -t 2 -a x11 -o stratum+tcp://mining.darkcointalk.org:3333 -u -p
 

Uleepera

New Member
May 30, 2014
7
0
1
What is the difference between them? The documentation doesn't seem to have been updated for this.
 

Syntor

Member
May 26, 2014
60
7
48
New Orleans, LA
You want to change yours to ccminer30, I am running 2 GTX 760's and with a little tweaking I am getting @ 2.05mhash/s out of each.. Here is my string:
ccminer30.exe -t 2 -a x11 -o stratum+tcp://mining.darkcointalk.org:3333 -u -p
MSI Afterburner settings: Power Limit : 84% ( I keep getting VRel cap limit @ 75%, think I need a bigger power supply), Core Clock: +192, Memory Clock: +192, Fan Speed: +84, My temps are at 73c and 67c ( I am running one EVGA and one Galaxy), the Galaxy has half the cooling vents as the EVGA hence the higher temp..
 

Syntor

Member
May 26, 2014
60
7
48
New Orleans, LA
Not 100% sure of the explanation, but it has to do with the computing power of the card itself.. sort of like when dealing with cudaminer.. some cards use K6x24 etc.. and some use T6x24 etc.. It has to do with how the card is rated or something like that.. just change the beginning of your string and try it, if you get the same error, change it to the other one.. I promise once you get the right one.. you will know!!