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

[primer10]

Hi ppl,
   Managed to compile ccminer (linux) on a system without the nvidia card and ran the binary successfully on the rig with 750Ti. The speed is about 2110kh/s and seems to be similar to what others has posted here.

   My qn is that Windows binaries offered for download has various version for Compute 3, 3.5 and 5.0. I didn't see that option when I was compiling on Linux. How can we choose which compute to compile for? Does it solely depend on the CUDA toolkit version (was using v6)?

on Linux we build all Compute versions into one binary. Under Visual Studio the GUI doesn't have the flexibility to configure it in the same way.

Thanks Christian!

Btw, I remembered I saw the "gencode=compute .." string as part of the parameters to nvcc? But I think I didn't see the one saying "compute 5.0"? (I might have remembered wrongly but I cannot verify now since I am outside)

[1714885035]

1714885035

[1714885035]

1714885035

[1714885035]

1714885035

[1714885035]

1714885035

[gpuminer]

What will be the next step for ccminer going forward ? It seems that there is no news regarding new release or development. Can we expect to see anything new this month ?

I think even before the weekend. Adding that killer groestl should be an easy exercise.

Thank you ! Haven't have any good news for ccminer for quite a while.

[Madychoux]

What will be the next step for ccminer going forward ? It seems that there is no news regarding new release or development. Can we expect to see anything new this month ?

I think even before the weekend. Adding that killer groestl should be an easy exercise.

Would be great !!!

Groestl is very energy efficient, great algo.

[cbuchner1]

Btw, I remembered I saw the "gencode=compute .." string as part of the parameters to nvcc? But I think I didn't see the one saying "compute 5.0"? (I might have remembered wrongly but I cannot verify now since I am outside)

Compute 5.0 requires CUDA 6.0 which has some performance regressions. I'd rather wait for CUDA 6.5
I did not see any improvements by targeting sm_50 directly.

Christian

[DemosMirak]

What will be the next step for ccminer going forward ? It seems that there is no news regarding new release or development. Can we expect to see anything new this month ?

I think even before the weekend. Adding that killer groestl should be an easy exercise.

Would be great !!!

Groestl is very energy efficient, great algo.

Not to mention that it would speed up X11 as well.

[zelante]

Btw, I remembered I saw the "gencode=compute .." string as part of the parameters to nvcc? But I think I didn't see the one saying "compute 5.0"? (I might have remembered wrongly but I cannot verify now since I am outside)

Compute 5.0 requires CUDA 6.0 which has some performance regressions. I'd rather wait for CUDA 6.5
I did not see any improvements by targeting sm_50 directly.

http://www.alexstjohn.com/WP/2014/06/01/cuda-6-5/
Did you got access to download & try CUDA 6.5 ?

[nrg_wolf]

any news of getting cryptonight algo added for nvidia?

[noobminerd]

Anyone has tested ccminer or cudaminer on new mobile geforce 860m 880m laptops? just for reference

[cayars]

Yes, I've run it on these mobile laptops.

Any specific questions about them?

[tsiv]

Oooh, first post. Exciting. It's a "look what I made!" post.



Took some doing but finally got there. Nothing fancy, pretty much a copy&paste implementation of the hamsi and fugue hashes bolted on the rear end of ccminer's X11 code and voila, X13. Seems to work and the hashrate is around 80% of X11.

Will clean up the code a bit, probably going to fork ccminer on github and go for a source only release there. Haven't actually tried building it on Windows yet, VS project files aren't exactly my strongest suit but I doubt it'll be too much trouble.

[ivanlabrie]

X13, nice :p

[bigjme]

my reply.... meh, no words

but congrats on implementing it. Im not interested in x13 atall but glad you could do it

[gpuminer]

Oooh, first post. Exciting. It's a "look what I made!" post.



Took some doing but finally got there. Nothing fancy, pretty much a copy&paste implementation of the hamsi and fugue hashes bolted on the rear end of ccminer's X11 code and voila, X13. Seems to work and the hashrate is around 80% of X11.

Will clean up the code a bit, probably going to fork ccminer on github and go for a source only release there. Haven't actually tried building it on Windows yet, VS project files aren't exactly my strongest suit but I doubt it'll be too much trouble.

Christ gonna release killer goretl version probably before or during weekend. The new version will upgrade x11 speed gtx 750 ti to 2.8 - 3. Maybe u should wait for the new release and implement your code which will increase x13 speed significantly. Or maybe both of u can combine code and do a combine release ? So the new version will have killer goretl + x13. That is pretty exciting.

[myagui]

Oooh, first post. Exciting. It's a "look what I made!" post.

That's an awesome 1st post right there 
Welcome aboard! What really get's me smiling is to know that there is +1 contributor to this project.
Christian does great work, but does not look like there are many coding contributions from others (other than the group previously known as CC hash factory). I'd love to participate, but have not had a chance yet to start learning myself some cuda (or any coding for that matter  ).
With a larger development community, we're bound to greater mining achievements! (and greater profits?  )

Cheers,
~ Myagui

[Bombadil]

Yes, I've run it on these mobile laptops.

Any specific questions about them?

Yeah, can you tell us what hashrates you get with that

Oooh, first post. Exciting. It's a "look what I made!" post.



Took some doing but finally got there. Nothing fancy, pretty much a copy&paste implementation of the hamsi and fugue hashes bolted on the rear end of ccminer's X11 code and voila, X13. Seems to work and the hashrate is around 80% of X11.

Will clean up the code a bit, probably going to fork ccminer on github and go for a source only release there. Haven't actually tried building it on Windows yet, VS project files aren't exactly my strongest suit but I doubt it'll be too much trouble.

* Bombadil will hit "refresh" 20x/hour here for the next days That AND killer groestl, can't wait

* Bombadil also starts adding X13 to his profit switching GUI (far from done, but meh, a bit everyday)

[djm34]

Oooh, first post. Exciting. It's a "look what I made!" post.


Took some doing but finally got there. Nothing fancy, pretty much a copy&paste implementation of the hamsi and fugue hashes bolted on the rear end of ccminer's X11 code and voila, X13. Seems to work and the hashrate is around 80% of X11.

Will clean up the code a bit, probably going to fork ccminer on github and go for a source only release there. Haven't actually tried building it on Windows yet, VS project files aren't exactly my strongest suit but I doubt it'll be too much trouble.

Christ gonna release killer goretl version probably before or during weekend. The new version will upgrade x11 speed gtx 750 ti to 2.8 - 3. Maybe u should wait for the new release and implement your code which will increase x13 speed significantly. Or maybe both of u can combine code and do a combine release ? So the new version will have killer goretl + x13. That is pretty exciting.

no reason to wait  
Actually I just curious to know what I did wrong

[Bombadil]

So here's my profitability list including X13 (using 1.75MH as hashrate per 750ti, as tsiv screenshotted )
http://pastebin.com/S5mUXWhv
Again prices aren't averages, so it could change in an instant ^^" But all formulas should be OK now.

[tsiv]

Aaand it's out: https://github.com/tsiv/ccminer

Insert the standard "if your house burns down, totally not my fault" disclaimer here.

Windows build should be fine as well, at least I managed to compile it myself. I won't be releasing any binaries, feel free to do so yourselves though.

Christ gonna release killer goretl version probably before or during weekend. The new version will upgrade x11 speed gtx 750 ti to 2.8 - 3. Maybe u should wait for the new release and implement your code which will increase x13 speed significantly. Or maybe both of u can combine code and do a combine release ? So the new version will have killer goretl + x13. That is pretty exciting.

Should be an easy modification, right now my X13 is using the same code for the first 11 algos as the original X11. At the simplest it just takes a recompile after throwing in the new groestl version.

no reason to wait  
Actually I just curious to know what I did wrong

I have a very good idea about that I did end up messing with the original echo hash, it seemed to have an extraneous xor that made the full 512 bit hash come out wrong. Then using that result for the additional hamsi+fugue hashes kinda ends badly. The funny thing is that it worked fine for X11 because it was the last hashing algo in the chain and the X11 algo only takes the first 256 bits of the hash, discarding the rest and the flipped bit ended up in the discarded part. Was a bitch to track down for sure.

[cryptomines]

Great work! marvelous first few posts!

building in windows right now......

[reviks]

As a nvidia user again I would like to try and stress the importance of API scrypt use in our cuda/ccminer, as I myself am not capable of doing the programming myself.
Check out our great mining pool at https://www.minep.it/
https://www.minep.it/creatives/newDashWithGlare.jpg