[ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX]

[drakoin]

anyone got experience with how to configure this now (Linux Ubuntu 12 on x86_64):

Code:

./cudaminer -a keccak  -o stratum+tcp://maxpool.1gh.com:17333 -p x -u mymaxcoinaddress

[2014-02-07 12:33:11] GPU #0: GRID K520 with compute capability 3.0
[2014-02-07 12:33:11] GPU #0: interactive: 0, tex-cache: 0 , single-alloc: 0
[2014-02-07 12:33:11] GPU #0: 8 hashes / 1.0 MB per warp.
[2014-02-07 12:33:27] GPU #0: Performing auto-tuning (Patience...)
[2014-02-07 12:33:27] GPU #0: maximum total warps (BxW): 3959
[2014-02-07 12:36:25] Stratum detected new block

How much patience? 


Can I use the maxpool.1gh.com with this cudaminer?


What is the optimal hardware settings?



Thank you very much!

[1714065133]

1714065133

[1714065133]

1714065133

[1714065133]

1714065133

[djm34]

What's the difficulty of this maxcoin right now?

809 and decreasing

[LuckyLuke]

what is best setting for gtx680?

[kokes]

What's the difficulty of this maxcoin right now?

[dbbit]

Every now and again (once an hour or so) my CudaMiner on one of my machines goes off into la-la-land and starts spewing these lines, and won't stop until I restart it:

[2014-02-07 04:39:46] GPU #0: cudaError 2 (out of memory) calling 'cudaStreamSynchronize(context_streams[0][thr_id])' (C:/Users/Administration/Documents/Visual Studio 2010/Projects/cudaminer-2014-02-06/salsa_kernel.cu line 223)

[2014-02-07 04:39:46] GPU #0: cudaError 2 (out of memory) calling 'cudaStreamSynchronize(context_streams[1][thr_id])' (C:/Users/Administration/Documents/Visual Studio 2010/Projects/cudaminer-2014-02-06/salsa_kernel.cu line 224)

My settings are:
-i 0 -H 2 -C 0 -m 1 -L 64 -l K512x16

on a:
GeForce GTX 650


Is there a setting I should change so it doesn't run out of memory?

[djm34]

Every now and again (once an hour or so) my CudaMiner on one of my machines goes off into la-la-land and starts spewing these lines, and won't stop until I restart it:

[2014-02-07 04:39:46] GPU #0: cudaError 2 (out of memory) calling 'cudaStreamSynchronize(context_streams[0][thr_id])' (C:/Users/Administration/Documents/Visual Studio 2010/Projects/cudaminer-2014-02-06/salsa_kernel.cu line 223)

[2014-02-07 04:39:46] GPU #0: cudaError 2 (out of memory) calling 'cudaStreamSynchronize(context_streams[1][thr_id])' (C:/Users/Administration/Documents/Visual Studio 2010/Projects/cudaminer-2014-02-06/salsa_kernel.cu line 224)

My settings are:
-i 0 -H 2 -C 0 -m 1 -L 64 -l K512x16

on a:
GeForce GTX 650


Is there a setting I should change so it doesn't run out of memory?

Could be the reason, try to increase the L value

[cbuchner1]

i get about 20% less mhashs compared to the solo miner, is that normal?

I think so. I understand it's compiled with CUDA 5.0, and the boost in kepler cards (most 600 series) comes from CUDA 5.5
Let's hope Christian will share the official binary soon (he compiles with cuda 5.5)

In any case I am very grateful for the share, right now I'm doing ~50Mhs on with a GTX660 1GH pool.

EDIT: Also I am very grateful that I now can finally get up from my home pc and go to work. Still having a chance to keep my job there.

I posted this now to the first posting in thread. as always: careful with inofficial binaries :-)

Stratum-mine ahead guys!

[kokes]

i get about 20% less mhashs compared to the solo miner, is that normal?

I think so. I understand it's compiled with CUDA 5.0, and the boost in kepler cards (most 600 series) comes from CUDA 5.5
Let's hope Christian will share the official binary soon (he compiles with cuda 5.5)

In any case I am very grateful for the share, right now I'm doing ~50Mhs on with a GTX660 1GH pool.

EDIT: Also I am very grateful that I now can finally get up from my home pc and go to work. Still having a chance to keep my job there.

I posted this now to the first posting in thread. as always: careful with in official binaries :-)

You're the king of kings, from 86mh/s to 120 again. LONG LIVE CUDA

EDIT: AND LONG LIVE CHRISTIAN!!!!

[GordonSSS]

I decided to make my optimised Windows cudaminer build donation-ware instead

https://mega.co.nz/#!O05yEZTK!j8u4ZFHZxqif94ihKnOIgYsIl7cYAfSdgJsbDiSmtpk

Donations will be much appreciated!

MAX: mR6iiBz4NRBMDjo2zpY9fJr5yUn4EL6pgd
XPM: AWFyioszN3vsyQsPbAtCybqu3j5v6FqQTE
BTC: 1LXgRb1F6KZmVQBzcKsfpAAL57Se9EKeT6

Looks like Spoetnik just beat me to it, but at least on my machine my build is 23% faster

I have cudaminer Windows binaries, working on http://max.1gh.com !

This is waaaay faster than CPU mining!

Willing to sell for only 0.1 BTC or 30 XPM. Please PM me if interested.

[weedoge]

I decided to make my optimised Windows cudaminer build donation-ware instead

https://mega.co.nz/#!O05yEZTK!j8u4ZFHZxqif94ihKnOIgYsIl7cYAfSdgJsbDiSmtpk

Donations will be much appreciated!

MAX: mR6iiBz4NRBMDjo2zpY9fJr5yUn4EL6pgd
XPM: AWFyioszN3vsyQsPbAtCybqu3j5v6FqQTE
BTC: 1LXgRb1F6KZmVQBzcKsfpAAL57Se9EKeT6

Looks like Spoetnik just beat me to it, but at least on my machine my build is 23% faster

I have cudaminer Windows binaries, working on http://max.1gh.com !

This is waaaay faster than CPU mining!

Willing to sell for only 0.1 BTC or 30 XPM. Please PM me if interested.

Uh.. isn't this the version Chris just added to the OP?

Also does anyone have any idea of the best settings to start with for a Titan? ^^ I'm at 120MHashes atm but have seen better reported.
I don't have any clue what the "warps" and stuff actually mean.

Thanks in advance!

[cbuchner1]

Uh.. isn't this the version Chris just added to the OP?

Also does anyone have any idea of the best settings to start with for a Titan? ^^ I'm at 120MHashes atm but have seen better reported.
I don't have any clue what the "warps" and stuff actually mean.

Thanks in advance!

Very likely the same version, except that the code in the OP was compiled by me.

[kokes]

Uh.. isn't this the version Chris just added to the OP?

Also does anyone have any idea of the best settings to start with for a Titan? ^^ I'm at 120MHashes atm but have seen better reported.
I don't have any clue what the "warps" and stuff actually mean.

Thanks in advance!

Very likely the same version, except that the code in the OP was compiled by me.

Which is more trustworthy
Congratulations on your maxcoin man, 1000+ how much BTC did you get from that?

[GordonSSS]

I probably made about the same fixes to the code as you did. My build was much faster than the one previously posted here (not by you, by someone else), but not quite as fast as yours is now (hadn't seen you'd just added that to the OP before I posted )

Uh.. isn't this the version Chris just added to the OP?

Also does anyone have any idea of the best settings to start with for a Titan? ^^ I'm at 120MHashes atm but have seen better reported.
I don't have any clue what the "warps" and stuff actually mean.

Thanks in advance!

Very likely the same version, except that the code in the OP was compiled by me.

[boy2k]

Can anyone help me with some settings for my 295 GTX's?

i tried

Code:

cudaminer --algo=keccak -d 0 -i -0 -C 2 -L 1024 -H 2 -l K1000x16 -R 5 --no-autotune -o stratum+tcp://maxpool.1gh.com:17333 -u meKxs7NUf8SfXjznNFtmNhFfsyvVachisE -p x

Cuda tells me the card has only 1.3 capability so i can't use K but then hashes at some mental speeds... it doesn't find anything though so i think its just ticking away and not doing anything.




Can anyone help me tune it in?  

[Katatsuki]

I decided to make my optimised Windows cudaminer build donation-ware instead

https://mega.co.nz/#!O05yEZTK!j8u4ZFHZxqif94ihKnOIgYsIl7cYAfSdgJsbDiSmtpk

Donations will be much appreciated!

MAX: mR6iiBz4NRBMDjo2zpY9fJr5yUn4EL6pgd
XPM: AWFyioszN3vsyQsPbAtCybqu3j5v6FqQTE
BTC: 1LXgRb1F6KZmVQBzcKsfpAAL57Se9EKeT6

Looks like Spoetnik just beat me to it, but at least on my machine my build is 23% faster

I have cudaminer Windows binaries, working on http://max.1gh.com !

This is waaaay faster than CPU mining!

Willing to sell for only 0.1 BTC or 30 XPM. Please PM me if interested.

Thanks man! I went from 87 to 122, great job.

[cbuchner1]

Which is more trustworthy
Congratulations on your maxcoin man, 1000+ how much BTC did you get from that?

I've only sold off like 200 MAX so far. I am having trouble with an uncomfirmed 2.5 BTC withdrawal
from coins.e-com right now.

I heard mcxNOW has the coin for trading now. I shall try my luck there.

[mknlp]

Whats the command to see my Address ?

[cbuchner1]

Can anyone help me tune it in?  

Fermi-device needs Fermi-Kernel -l F1000x16 or so

[cbuchner1]

I probably made about the same fixes to the code as you did. My build was much faster than the one previously posted here (not by you, by someone else), but not quite as fast as yours is now (hadn't seen you'd just added that to the OP before I posted )

I know using CUDA 5.0 vs CUDA 5.5 toolkit makes a speed difference. That, and maybe the fact that I am using Visual Studio 2010
and the dependencies I made available in the OP.

[kokes]

Which is more trustworthy
Congratulations on your maxcoin man, 1000+ how much BTC did you get from that?

I've only sold off like 200 MAX so far. I am having trouble with an uncomfirmed 2.5 BTC withdrawal
from coins.e-com right now.

I heard mcxNOW has the coin for trading now. I shall try my luck there.

Nice, i also just signed up on mcxnow, seems like a fast and stable site! Good luck.