to tell the truth...
when i start mining
i was suprise that ati was better that nvidia...
but now happy to see support for real gpu...
Topic: [ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 442. (Read 3426976 times)
May 22, 2014, 01:08:04 AM
May 21, 2014, 11:49:30 PM
I'm trying to compile ccminer on Ubuntu Server 12.04 64bit, Googled but couldn't find an answer.
myriadgroestl.o cuda_myriadgroestl.o -L/usr/lib/x86_64-linux-gnu -lcurl -Wl,-Bsymbolic-functions -Wl,-z,relro compat/jansson/libjansson.a -lpthread -lcudart -static-libstdc++ -fopenmp -lcrypto -lssl -lcrypto -lssl
ccminer-cpu-miner.o: In function `miner_thread':
/home/ubuntu/ccminer-1.0/cpu-miner.c:903: undefined reference to `scanhash_nist5'
/home/ubuntu/ccminer-1.0/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/ubuntu/ccminer-1.0'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/ubuntu/ccminer-1.0'
make: *** [all] Error 2
I think it's chown related.
myriadgroestl.o cuda_myriadgroestl.o -L/usr/lib/x86_64-linux-gnu -lcurl -Wl,-Bsymbolic-functions -Wl,-z,relro compat/jansson/libjansson.a -lpthread -lcudart -static-libstdc++ -fopenmp -lcrypto -lssl -lcrypto -lssl
ccminer-cpu-miner.o: In function `miner_thread':
/home/ubuntu/ccminer-1.0/cpu-miner.c:903: undefined reference to `scanhash_nist5'
/home/ubuntu/ccminer-1.0/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/ubuntu/ccminer-1.0'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/ubuntu/ccminer-1.0'
make: *** [all] Error 2
I think it's chown related.
May 21, 2014, 09:53:03 PM
So for two days I've mined for SCIENCE only. Makes me feel all warm and fuzzy now.
I mined for two days as well, and I had my team number correct and still didn't see any coins whatsoever, even after waiting 3 days. But I'm sure my video cards are warm and fuzzy.
May 21, 2014, 09:03:04 PM
Christian, Just as a heads up, this may or may not be important for CUDA on X11. I'm not sure if your ccminer already implements this technique but the Dev of the newest X11 radeon miner used the following technique to get increased speeds. He split the opencl code into 11 different kernels in 11 different folders instead of having them in a single long kernel. Doing it this way with shorter kernels made better use of SIMD which brings a speed improvement. Not sure if you can do the same for CUDA X11 or even if you already did something similar, but maybe its a way to get even more improvement out of CUDA?
I am no programmer but this sounds very familiar.
May 21, 2014, 07:37:15 PM
I honestly haven't been following MRO, but with only 200 sync coins left to be mined I was just exploring options.
May 21, 2014, 07:34:54 PM
Christian, Just as a heads up, this may or may not be important for CUDA on X11. I'm not sure if your ccminer already implements this technique but the Dev of the newest X11 radeon miner used the following technique to get increased speeds. He split the opencl code into 11 different kernels in 11 different folders instead of having them in a single long kernel. Doing it this way with shorter kernels made better use of SIMD which brings a speed improvement. Not sure if you can do the same for CUDA X11 or even if you already did something similar, but maybe its a way to get even more improvement out of CUDA?
May 21, 2014, 05:43:57 PM
Any whisperings about -a cryptonight?
I think its dead on arrival. It just got quite clear that ring signatures will bloat a blockchain to 8 times the size of Bitcoin's.
I don't really care if it also bloats the rates you can sell MRO at exchanges
Strictly speaking, bitcoin's blockchain also isn't sustainable.
Christian
May 21, 2014, 05:42:27 PM
Any whisperings about -a cryptonight?
I think its dead on arrival. It just got quite clear that ring signatures will bloat a blockchain to 8 times the size of Bitcoin's.
May 21, 2014, 05:34:37 PM
Any whisperings about -a cryptonight?
Anyone who whispered it has since lost his voice.
May 21, 2014, 05:33:17 PM
Any whisperings about -a cryptonight?
May 21, 2014, 05:01:50 PM
Does ccminer run with Tesla cards?
Could you estimate a hash rate for X11 on a K20?
Could you estimate a hash rate for X11 on a K20?
I don't know if anyone tried a K20 before, so I can't say
But I can tell you that you'll need to use ccminer35.exe or ccminer30.exe (whatever turns out best).
May 21, 2014, 04:54:11 PM
Does ccminer run with Tesla cards?
Could you estimate a hash rate for X11 on a K20?
Could you estimate a hash rate for X11 on a K20?
May 21, 2014, 04:51:14 PM
Team number is 224497, if it's missing you won't get anything...
So for two days I've mined for SCIENCE only. Makes me feel all warm and fuzzy now.
Do you have a curecoin address?
May 21, 2014, 04:14:11 PM
Some helpful stuff for those who mining with our batch solution on NiceHash.
Now this is console program can receive prices from NiceHash
https://github.com/zelante/NiceHashPrices
Now this is console program can receive prices from NiceHash
https://github.com/zelante/NiceHashPrices
Are you reading the prices from the website or stratum?
If its the Website, maybe you can add https://www.ltcrabbit.com/index.php?page=x11 and https://ipominer.com/stats too.
They are offereing similar Informations for X11.
May 21, 2014, 04:11:02 PM
@zelante
Maybe we need another cudaminer/ccminer patch for the failover script.
The stratum timeout is too long.
We are loosing mining time if the price changes while mining on nicehash.
Maybe cudaminer/ccminer should exit after 5 seconds so that the script can check the price again.
Stratum time outs exactly after recconnect?Maybe we need another cudaminer/ccminer patch for the failover script.
The stratum timeout is too long.
We are loosing mining time if the price changes while mining on nicehash.
Maybe cudaminer/ccminer should exit after 5 seconds so that the script can check the price again.
I have stratum timeouts without reconnects, so i think it is idle on they side.
I can change algo so that the script check price 2 or more times and then connect to NiceHash.
I think the price was falling a bit under my level only for some seconds and then there is always a 2 minutes timeout.
May 21, 2014, 04:10:43 PM
I think it is a good idea to just let him speak to these discussions about supporting AMD/ATI cards and not just engage in banter one way or another.
the last ATI (now AMD) card I owned was a Radeon 9550 I think (2005 timeframe)
Also I am not familiar at all with OpenCL. But I consider myself really proficient in CUDA.
I own about 20 nVidia cards (but not all are currently in service).
So why is it I should support/improve AMD mining? It would cut into my nVidia mining profits.
Like, seriously. Supporting both vendors would have the same effect as supporting none.
It wouldn't shift the balance of power in favor of anybody.
Christian
Another bullish statement read this week. Majority of x11 hashrate now belongs to amd card. Improving their hashrate would seriously cut down nvida advantage and profit margin
(I think I am already done with x11 coin... they have redefined the term of shitcoin with 2 or 3 launch/ day even during business days... totally insane)Those two guys WERE GPU/CPU mining XPM at ypool, if it makes you feel any better, so the smaller XPM miners are not unhappy to see them on something else. Then again, I'm on Curecoin myself now (from XPM) so I share your pain.
looks like everyone who was mining xpm moved to curecoin (even primegpu... ) hopefully nobody will resell a better folding@home for 500k curecoins (the funny thing is that there were all complaining about botnets when they got huge resources themselves... ) May 21, 2014, 04:10:28 PM
Team number is 224497, if it's missing you won't get anything...
So for two days I've mined for SCIENCE only. Makes me feel all warm and fuzzy now.
May 21, 2014, 04:07:02 PM
@zelante
Maybe we need another cudaminer/ccminer patch for the failover script.
The stratum timeout is too long.
We are loosing mining time if the price changes while mining on nicehash.
Maybe cudaminer/ccminer should exit after 5 seconds so that the script can check the price again.
Stratum time outs exactly after recconnect?Maybe we need another cudaminer/ccminer patch for the failover script.
The stratum timeout is too long.
We are loosing mining time if the price changes while mining on nicehash.
Maybe cudaminer/ccminer should exit after 5 seconds so that the script can check the price again.
I have stratum timeouts without reconnects, so i think it is idle on they side.
I can change algo so that the script check price 2 or more times and then connect to NiceHash.
I think the price was falling a bit under my level only for some seconds and after there is always a 2 minutes timeout.
May 21, 2014, 03:25:52 PM
I think it is a good idea to just let him speak to these discussions about supporting AMD/ATI cards and not just engage in banter one way or another.
the last ATI (now AMD) card I owned was a Radeon 9550 I think (2005 timeframe)
Also I am not familiar at all with OpenCL. But I consider myself really proficient in CUDA.
I own about 20 nVidia cards (but not all are currently in service).
So why is it I should support/improve AMD mining? It would cut into my nVidia mining profits.
Like, seriously. Supporting both vendors would have the same effect as supporting none.
It wouldn't shift the balance of power in favor of anybody.
Christian
Another bullish statement read this week. Majority of x11 hashrate now belongs to amd card. Improving their hashrate would seriously cut down nvida advantage and profit margin
(I think I am already done with x11 coin... they have redefined the term of shitcoin with 2 or 3 launch/ day even during business days... totally insane)Those two guys WERE GPU/CPU mining XPM at ypool, if it makes you feel any better, so the smaller XPM miners are not unhappy to see them on something else. Then again, I'm on Curecoin myself now (from XPM) so I share your pain.
May 21, 2014, 03:18:07 PM
@zelante
Maybe we need another cudaminer/ccminer patch for the failover script.
The stratum timeout is too long.
We are loosing mining time if the price changes while mining on nicehash.
Maybe cudaminer/ccminer should exit after 5 seconds so that the script can check the price again.
Stratum time outs exactly after recconnect?Maybe we need another cudaminer/ccminer patch for the failover script.
The stratum timeout is too long.
We are loosing mining time if the price changes while mining on nicehash.
Maybe cudaminer/ccminer should exit after 5 seconds so that the script can check the price again.
I have stratum timeouts without reconnects, so i think it is idle on they side.
I can change algo so that the script check price 2 or more times and then connect to NiceHash.
Jump to: