Mxrider420z
Sr. Member
Offline
Activity: 619
Merit: 289
Ⓚ KORE TEAM Ⓚ ★ Portuguese/SPA
|
|
July 02, 2014, 01:14:07 AM |
|
I have updated my repository with x14 and qubit: https://github.com/djm34/ccminerIt is doing rather well solo mining on Deepcoin, I already found 6 blocks... (which is quite funny as running on hashharder just bring some scrap... ), which are worth close to nothing For the moment my 780ti is 2MHash/s faster than my R9 290x (and the R9 didn't find any blocks yet...) any chance for windows binary ? +1
|
|
|
|
djm34
Legendary
Offline
Activity: 1400
Merit: 1050
|
|
July 02, 2014, 01:14:49 AM |
|
I have updated my repository with x14 and qubit: https://github.com/djm34/ccminerIt is doing rather well solo mining on Deepcoin, I already found 6 blocks... (which is quite funny as running on hashharder just bring some scrap... ), which are worth close to nothing For the moment my 780ti is 2MHash/s faster than my R9 290x (and the R9 didn't find any blocks yet...) any chance for windows binary ? tomorrow... takes an hour and half to compile... (and people are always complaining about my binaries... )
|
djm34 facebook pageBTC: 1NENYmxwZGHsKFmyjTc5WferTn5VTFb7Ze Pledge for neoscrypt ccminer to that address: 16UoC4DmTz2pvhFvcfTQrzkPTrXkWijzXw
|
|
|
|
yellowduck2
|
|
July 02, 2014, 01:39:33 AM |
|
Anyone keeping track of ccminer ? There seems to be many version running around. We need unify ccminer.
|
|
|
|
Stratobitz
Legendary
Offline
Activity: 1022
Merit: 1010
|
|
July 02, 2014, 02:09:17 AM |
|
Do any of the cuda miners available now have the ability to mine X11, X13, or X15 with the lower compute cards, like the GTX 570?
Just curious if I missed something.
Thanks,
Strato
|
|
|
|
cayars
|
|
July 02, 2014, 03:43:00 AM |
|
Anyone keeping track of ccminer ? There seems to be many version running around. We need unify ccminer.
Well we have official Christian release 1.2. We have tsiv CryptoNight release based on release 1.1 We have djm34 release based on 1.2 djm34 has everything except for CryptoNight in it. djm34, any chance you would consider adding tsiv's cryptonight code into your miner? If you were to do this your miner would have everything currently available in ccminer. Could you/would you? Carlo PS Yellowduck2: Just use djm34's release which has everything Christian's last release has plus more. It has all ccminer algos except CryptoNight. Use tsiv's release for CryptoNight.
|
|
|
|
ap1
Newbie
Offline
Activity: 41
Merit: 0
|
|
July 02, 2014, 04:05:41 AM |
|
I have updated my repository with x14 and qubit: https://github.com/djm34/ccminerIt is doing rather well solo mining on Deepcoin, I already found 6 blocks... (which is quite funny as running on hashharder just bring some scrap... ), which are worth close to nothing For the moment my 780ti is 2MHash/s faster than my R9 290x (and the R9 didn't find any blocks yet...) any chance for windows binary ? tomorrow... takes an hour and half to compile... (and people are always complaining about my binaries... ) I tried compiling djm34's source under Linux but it stayed at the point shown below for more than 2 hours already... I killed it as I thought it was hung but sounds like it is normal to wait a couple of hours? It is a Xeon 1230V2 but I was running cudaminer in parallel while compiling.. nvcc -g -O2 -I . -Xptxas "-abi=no -v" -gencode=arch=compute_30,code=\"sm_30,compute_30\" -gencode=arch=compute_35,code=\"sm_35,compute_35\" --maxrregcount=80 --ptxas-options=-v -I./compat/jansson -o x13/cuda_whirlpool512.o -c x13/cuda_whirlpool512.cu x13/cuda_whirlpool512.cu(59): warning: function "HIWORD" was declared but never referenced
x13/cuda_whirlpool512.cu(68): warning: function "REPLACE_HIWORD" was declared but never referenced
x13/cuda_whirlpool512.cu(73): warning: function "LOWORD" was declared but never referenced
x13/cuda_whirlpool512.cu(91): warning: function "REPLACE_LOWORD" was declared but never referenced
x13/cuda_whirlpool512.cu(59): warning: function "HIWORD" was declared but never referenced
x13/cuda_whirlpool512.cu(68): warning: function "REPLACE_HIWORD" was declared but never referenced
x13/cuda_whirlpool512.cu(73): warning: function "LOWORD" was declared but never referenced
x13/cuda_whirlpool512.cu(91): warning: function "REPLACE_LOWORD" was declared but never referenced
x13/cuda_whirlpool512.cu(59): warning: function "HIWORD" was declared but never referenced
x13/cuda_whirlpool512.cu(68): warning: function "REPLACE_HIWORD" was declared but never referenced
x13/cuda_whirlpool512.cu(73): warning: function "LOWORD" was declared but never referenced
x13/cuda_whirlpool512.cu(91): warning: function "REPLACE_LOWORD" was declared but never referenced
|
|
|
|
|
restless
Legendary
Offline
Activity: 1151
Merit: 1001
|
|
July 02, 2014, 07:55:25 AM |
|
Any chance of seeing an algo-switching feature? sgminer currently supports this, for example using with NiceHash you can see the config here: https://www.nicehash.com/multialgo/Before algo-switching ccminer needs failover pool switching... BTW, what happened with "killer-groestl" is this being used in some algos atm?
|
|
|
|
Balitorium
Newbie
Offline
Activity: 23
Merit: 0
|
|
July 02, 2014, 08:05:03 AM |
|
yeah, I managed to get 190 H/s out of the K520 - still not profitable even if you include the CPU... but it was worth a try ;-)
|
|
|
|
djm34
Legendary
Offline
Activity: 1400
Merit: 1050
|
|
July 02, 2014, 08:31:01 AM |
|
I have updated my repository with x14 and qubit: https://github.com/djm34/ccminerIt is doing rather well solo mining on Deepcoin, I already found 6 blocks... (which is quite funny as running on hashharder just bring some scrap... ), which are worth close to nothing For the moment my 780ti is 2MHash/s faster than my R9 290x (and the R9 didn't find any blocks yet...) any chance for windows binary ? tomorrow... takes an hour and half to compile... (and people are always complaining about my binaries... ) I tried compiling djm34's source under Linux but it stayed at the point shown below for more than 2 hours already... I killed it as I thought it was hung but sounds like it is normal to wait a couple of hours? It is a Xeon 1230V2 but I was running cudaminer in parallel while compiling.. I am afraid it is normal (if you killed after 2 hours... it was almost done ). This part (as many other in ccminer... but this one is still badly coded) uses texture memory, which means that calculations involving constants stored in this memory are precalculated by the compiler so they can be processed faster when the program runs and it is some giant loops... I found a way to make it faster, but it is slower on the 750ti (while it works better on 780ti, probably kepler...) Anyone keeping track of ccminer ? There seems to be many version running around. We need unify ccminer.
Well we have official Christian release 1.2. We have tsiv CryptoNight release based on release 1.1 We have djm34 release based on 1.2 djm34 has everything except for CryptoNight in it. djm34, any chance you would consider adding tsiv's cryptonight code into your miner? If you were to do this your miner would have everything currently available in ccminer. Could you/would you? Carlo I could, I guess... but you realise that tsiv has just removed all the other algo from his miner (which actually is better especially if you compile often... ), and at the moment it is still a work in progress. may-be once it is final... (assuming it doesn't make me looks like I want to step on everybody's toes...) However, I think it would be better if everything was moving to Christian official release
|
djm34 facebook pageBTC: 1NENYmxwZGHsKFmyjTc5WferTn5VTFb7Ze Pledge for neoscrypt ccminer to that address: 16UoC4DmTz2pvhFvcfTQrzkPTrXkWijzXw
|
|
|
bigjme
|
|
July 02, 2014, 08:41:02 AM |
|
I feel like shopping again, this time in the Intel HPC camp...
has knight's landing potential for mining? even a TI pocket calculator can mine :-) ... and with a "killer" hashrate, I bet!! certainly kills one's patience... BTW Knight's landing is Q3 2015, all you get now is Knight's Corner. I take it by Knights your referring to the Xeon Phi Coprocessors?
|
Owner of: cudamining.co.uk
|
|
|
PVmining
|
|
July 02, 2014, 08:47:53 AM |
|
djm34, any chance you would consider adding tsiv's cryptonight code into your miner? If you were to do this your miner would have everything currently available in ccminer.
...this is a good question... is it a clever idea to plug every algo in ccminer and compile it all together? it takes quite long at the moment... what would it take in a few months? is there a possibility to get a plug-in / add-on based structure, where you just have to compile the changed algo?
|
|
|
|
bigjme
|
|
July 02, 2014, 08:52:35 AM |
|
It may be poissible to tell the miner to check for given files in its located folder and build dynamically from that. Not easy atall And it would have to include the files for the hashing as a separate file the same way it includes .dll's
very very complex
|
Owner of: cudamining.co.uk
|
|
|
AizenSou
|
|
July 02, 2014, 08:59:20 AM |
|
nope, but i honestly dont know what else to go mine anymore
britcoin is profitable right now, 0.03 a day with 6x 750ti first rule here: always mine new coins, old coins are doomed to be not profitable i've already made roi with my 6 x 750ti lmao, took me only two weeks, too easy I was naive at first when I joined crypto and I got this number too, but it only works if you have a low number of rigs. So you could join the instamine party and dumped whatever coin for some high price because in the end you can only dump a little amount of coin. Don' think this strategy works if you expand your farm and want to earn big money. In my experience, your farm grows with your ROI time, unfortunately.
|
|
|
|
Amph
Legendary
Offline
Activity: 3206
Merit: 1069
|
|
July 02, 2014, 09:02:31 AM |
|
i made the "fresh algo" based on djm34 qubit, probably it don't work but i'm learning new stuff
|
|
|
|
S_tring
Full Member
Offline
Activity: 252
Merit: 102
OPEN Platform - Powering Blockchain Acceptance
|
|
July 02, 2014, 09:08:50 AM |
|
There's too many versions of ccMiner from different sources. Can't we have one github with one version mining all algos?
It's really good that there are so many active devs, but it's getting a little confusing!
|
|
|
|
bigjme
|
|
July 02, 2014, 09:16:45 AM |
|
nope, but i honestly dont know what else to go mine anymore
britcoin is profitable right now, 0.03 a day with 6x 750ti first rule here: always mine new coins, old coins are doomed to be not profitable i've already made roi with my 6 x 750ti lmao, took me only two weeks, too easy I was naive at first when I joined crypto and I got this number too, but it only works if you have a low number of rigs. So you could join the instamine party and dumped whatever coin for some high price because in the end you can only dump a little amount of coin. Don' think this strategy works if you expand your farm and want to earn big money. In my experience, your farm grows with your ROI time, unfortunately. my 780 hit ROI pretty quick thanks to maxcoin Because ive been sitting on older coins my 750Ti's still havent hit ROI in over 2 months now, plus to get those i had to buy £300 worth of PSU's sooo its going to be a while till they have cost me nothing. Its all down to luck, and Amph seems to have a lot of it! the only thing is that right now, Britcoin is calculating to be more along the lines of 0.005btc thats with 9.5MH/s, or just under 5 x 750Ti's
|
Owner of: cudamining.co.uk
|
|
|
djm34
Legendary
Offline
Activity: 1400
Merit: 1050
|
|
July 02, 2014, 09:17:12 AM |
|
There's too many versions of ccMiner from different sources. Can't we have one github with one version mining all algos?
It's really good that there are so many active devs, but it's getting a little confusing!
there are 3.... it is still possible to count them with only one hand i made the "fresh algo" based on djm34 qubit, probably it don't work but i'm learning new stuff did they release what they will use ? Actually you can check if it works on any pools (if you gets boo, it works, if you gets "don't validate on cpu", it doesn't)
|
djm34 facebook pageBTC: 1NENYmxwZGHsKFmyjTc5WferTn5VTFb7Ze Pledge for neoscrypt ccminer to that address: 16UoC4DmTz2pvhFvcfTQrzkPTrXkWijzXw
|
|
|
Amph
Legendary
Offline
Activity: 3206
Merit: 1069
|
|
July 02, 2014, 09:26:50 AM |
|
ah nevermind is more complex, 5 rounds with 3 algo, like quark 9 rounds with 6 algo
|
|
|
|
|