Topic: Bounty for Open-Sourced XMR/Cryptonight GPU Miner Bounties Thread - page 8. (Read 170965 times)
September 22, 2014, 06:30:46 PM
GTX 970 is probably the better card for mining, for the money.
August 27, 2014, 05:54:33 PM
? September 22, 2014, 04:34:01 PM
Anyone settings / updated miner for GTX 980?
Send me the card and I will test it out for you.
Nice GPU.
My lil 750Ti, I use this 8x30
Maybe try that and then try other settings to compare to.
No matter what settings i try the card is barely warm with 60c and 30% cooler. Its not used to full. Hash is 530 with 8x128 (best ive gotten)
Tsiv, any ideas? It should get at least 3x 750ti perf @1000hash
September 20, 2014, 09:56:06 PM
GTX 980 makes GPU mining profitable again.
How? It's absurdly expensive, and at first (naive) testing, appears to be similar to the 750ti, plus or minus a bit. Ok, you save pci-e slots. That's good. But the ratios of hash vs power draw are nothing new.. now maybe that changes as developers start really optimizing for the cards.. but I don't think we're going to see massive improvement when the first pass was basically this good.
Maybe there are new silicon features that can be exploited, though..
August 15, 2014, 02:17:22 AM
set -l ?
nomal is 8x40 i can't run now i use 6x40 can run on 750ti 220 h/s
nomal is 8x40 i can't run now i use 6x40 can run on 750ti 220 h/s
Known bug - cayars used my code to fix it in nvMiner.
September 20, 2014, 06:38:27 PM
GTX 980 makes GPU mining profitable again.
September 20, 2014, 05:56:22 PM
Anyone settings / updated miner for GTX 980?
Send me the card and I will test it out for you.
Nice GPU.
My lil 750Ti, I use this 8x30
Maybe try that and then try other settings to compare to.
September 20, 2014, 09:46:34 AM
Anyone settings / updated miner for GTX 980?
August 04, 2014, 02:56:42 PM
I have worked on CryptoNight ATI sgminer based on sph-sgminer.
I have already lost more then one week for this full-time job
, but currently I don't have time for this project. 
Here is repo: https://github.com/Atrides/sgminer-monero
What is done: all except of right cryptonight.cl
Miner works currently via RPCv1, available on test-port 8001 on erebor.dwarfpool.com
I have integrated djm's cpu-miner to compare hash, so, debuggers can see it in logs as "DIFFERENT_HASH" if OpenCL gives wrong hash.
In "cryptonight.cl_withaes_noworking" I have added AES implementation from djm. But with some errors.
"cryptonight.cl" == "cryptonight.cl_withkeccak_noaes" does not have AES therefore bad hashes.
So, if anybody can check and make normal "cryptonight.cl", you can use test port 8001.
I have already lost more then one week for this full-time job
, but currently I don't have time for this project. Here is repo: https://github.com/Atrides/sgminer-monero
What is done: all except of right cryptonight.cl
Miner works currently via RPCv1, available on test-port 8001 on erebor.dwarfpool.com
I have integrated djm's cpu-miner to compare hash, so, debuggers can see it in logs as "DIFFERENT_HASH" if OpenCL gives wrong hash.
In "cryptonight.cl_withaes_noworking" I have added AES implementation from djm. But with some errors.
"cryptonight.cl" == "cryptonight.cl_withkeccak_noaes" does not have AES therefore bad hashes.
So, if anybody can check and make normal "cryptonight.cl", you can use test port 8001.
Awesome, I have a CL file, but no good host! If we put it together, we might get something.
See my repo and code here: https://bitcointalksearch.org/topic/unfinished-cryptonight-opencl-amd-miner-671784
August 28, 2014, 03:19:22 AM
I got similar off-limit readings using my old GeForce GT 430 card -- even without overclocking.
August 22, 2014, 01:15:49 PM
What is it?
This problem with card?
?
This problem with card?
? August 08, 2014, 06:59:52 AM
set -l ?
nomal is 8x40 i can't run now i use 6x40 can run on 750ti 220 h/s
nomal is 8x40 i can't run now i use 6x40 can run on 750ti 220 h/s
July 29, 2014, 04:37:12 AM
I am starting to take a serious look at this algo but that cannot happen until the playing field is level.
If the guy who made the miner kept it a secret and just mined , like other things with other coins have gone down in the past he would have been better off but he wants a piece of everyone's action.
Developers deserve compensation for thier work , but equal to what they develop.
The rest is for the community.
I would urge those who have gpu miners to make the code open source for the good of all.
I think the fame you will get and respect and potential for future work will far outweigh your current strategy.
Please consider this.
95% of the community doesn't do shit for you when you release a 2x faster miner. They take their profit and forget about you, almost as if they'd rather be mining at half speed.
August 06, 2014, 01:46:09 AM
Bump 
July 31, 2014, 08:18:10 PM
thanks, cool , i forked your repo ill check it out probably this weekend should have some time
July 31, 2014, 05:59:08 AM
I have worked on CryptoNight ATI sgminer based on sph-sgminer.
I have already lost more then one week for this full-time job , but currently I don't have time for this project.
Here is repo: https://github.com/Atrides/sgminer-monero
What is done: all except of right cryptonight.cl
Miner works currently via RPCv1, available on test-port 8001 on erebor.dwarfpool.com
I have integrated djm's cpu-miner to compare hash, so, debuggers can see it in logs as "DIFFERENT_HASH" if OpenCL gives wrong hash.
In "cryptonight.cl_withaes_noworking" I have added AES implementation from djm. But with some errors.
"cryptonight.cl" == "cryptonight.cl_withkeccak_noaes" does not have AES therefore bad hashes.
So, if anybody can check and make normal "cryptonight.cl", you can use test port 8001.
I have already lost more then one week for this full-time job
, but currently I don't have time for this project. Here is repo: https://github.com/Atrides/sgminer-monero
What is done: all except of right cryptonight.cl
Miner works currently via RPCv1, available on test-port 8001 on erebor.dwarfpool.com
I have integrated djm's cpu-miner to compare hash, so, debuggers can see it in logs as "DIFFERENT_HASH" if OpenCL gives wrong hash.
In "cryptonight.cl_withaes_noworking" I have added AES implementation from djm. But with some errors.
"cryptonight.cl" == "cryptonight.cl_withkeccak_noaes" does not have AES therefore bad hashes.
So, if anybody can check and make normal "cryptonight.cl", you can use test port 8001.
July 29, 2014, 01:53:56 AM
I am starting to take a serious look at this algo but that cannot happen until the playing field is level.
If the guy who made the miner kept it a secret and just mined , like other things with other coins have gone down in the past he would have been better off but he wants a piece of everyone's action.
Developers deserve compensation for thier work , but equal to what they develop.
The rest is for the community.
I would urge those who have gpu miners to make the code open source for the good of all.
I think the fame you will get and respect and potential for future work will far outweigh your current strategy.
Please consider this.
July 23, 2014, 12:50:30 AM
Why isn't the donation addressees in the OP?
There has been no collection done for this bounty. Everything so far has been done by pledges on the honor system, with payments directly to the eventual winner.
July 23, 2014, 12:37:34 AM
As of today I'll be considering the bounty for the Nvidia GPU miner to be fulfilled by Tsiv. It seems as if it's in widespread use and there are few problems. Thank you very much for your effort!
Tsiv: The bounty addresses below will be used. It seems that the BTC part (.2 BTC) has already been paid by HardwarePal.
XMR: 42uasNqYPnSaG3TwRtTeVbQ4aRY3n9jY6VXX3mfgerWt4ohDQLVaBPv3cYGKDXasTUVuLvhxetcuS16 ynt85czQ48mbSrWX
BTC: 1JHDKp59t1RhHFXsTw2UQpR3F9BBz3R3cs
There will be no collection, and everyone who has pledged should just send to those addresses and then post the tx info here if you'd like.
-updated by kbm
Tsiv: The bounty addresses below will be used. It seems that the BTC part (.2 BTC) has already been paid by HardwarePal.
XMR: 42uasNqYPnSaG3TwRtTeVbQ4aRY3n9jY6VXX3mfgerWt4ohDQLVaBPv3cYGKDXasTUVuLvhxetcuS16 ynt85czQ48mbSrWX
BTC: 1JHDKp59t1RhHFXsTw2UQpR3F9BBz3R3cs
There will be no collection, and everyone who has pledged should just send to those addresses and then post the tx info here if you'd like.
-updated by kbm
Just popping in to confirm having received the BTC part from HardwarePal a while back and the 150 and 300 XMR from kbm and smooth respectively. Cheers, gents.
July 17, 2014, 06:32:47 PM
Why isn't the donation addressees in the OP?
Jump to: