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

[AizenSou]

The new keccak code seems faster than a CPU I think.

Q9550 @3.4 Ghz (one core)
minerd-maxcoin --algo=keccak --benchmark
1151 khash/s.

i3-4130T @2.90GHz (one core)
minerd-maxcoin --algo=keccak --benchmark
599.59 khash/s.  (?? must be a fluke, lots of stuff running on this machine)

Celeron G1630 @ 2.80GHz (one core)
minerd-maxcoin --algo=keccak --benchmark
1020 khash/s

GTX560-448
cudaminer --algo=keccak -d 0 --benchmark -L16 -l F14x16
28468 Kh/s

I could not get cudaminer to run on my Quadro 295NVS. Not sure of the kernel parameters, it wouldn't start (not that I would want to run on this card anyway).

I get 85296 khash/s with cudaminer --algo=keccak -d gtx780 -L 16 -l K192x32 --benchmark.
Can't even believe my eyes  

Outstanding work, Christian. Could you post your VTC address so I can donate some VTCs to you?

[paulus51]

hi do you have a link for the lates build ? already compiled version ?

thanks in advanced

[bigjme]

The new keccak code seems faster than a CPU I think.

Q9550 @3.4 Ghz (one core)
minerd-maxcoin --algo=keccak --benchmark
1151 khash/s.

i3-4130T @2.90GHz (one core)
minerd-maxcoin --algo=keccak --benchmark
599.59 khash/s.  (?? must be a fluke, lots of stuff running on this machine)

Celeron G1630 @ 2.80GHz (one core)
minerd-maxcoin --algo=keccak --benchmark
1020 khash/s

GTX560-448
cudaminer --algo=keccak -d 0 --benchmark -L16 -l F14x16
28468 Kh/s

I could not get cudaminer to run on my Quadro 295NVS. Not sure of the kernel parameters, it wouldn't start (not that I would want to run on this card anyway).

I get 85296 khash/s with cudaminer --algo=keccak -d gtx780 -L 16 -l K192x32 --benchmark.
Can't even believe my eyes  

Outstanding work, Christian. Could you post your VTC address so I can donate some VTCs to you?

How are you managing those speeds? When did you download the github? And windows or linux?

[AizenSou]

hi do you have a link for the lates build ? already compiled version ?

thanks in advanced

check github. You can compile yourself. It is new so everyone is busy with their stuffs and I don't think Windows binary will be available immediately. I can help you with compiling in Linux but for Windows I'm sorry.

I get 85296 khash/s with cudaminer --algo=keccak -d gtx780 -L 16 -l K192x32 --benchmark.
Can't even believe my eyes  

Outstanding work, Christian. Could you post your VTC address so I can donate some VTCs to you?

How are you managing those speeds? When did you download the github? And windows or linux?

I compile the newest source in my Gentoo box.

[mrm0]

Current code cloned from github does not compile on Linux:

Code:

[...]
ptxas info    : Compiling entry function '_Z25titan_scrypt_core_kernelAILi1EL12MemoryAccess1EEvPKjii' for 'sm_35'
ptxas info    : Used 26 registers, 336 bytes cmem[0]
ptxas info    : Compiling entry function '_Z25titan_scrypt_core_kernelBILi1EL12MemoryAccess1EEvPjii' for 'sm_35'
ptxas info    : Used 30 registers, 336 bytes cmem[0]
g++  -O3 -pthread -L/usr/local/cuda/lib64  -o cudaminer cudaminer-cpu-miner.o cudaminer-util.o cudaminer-sha2.o cudaminer-scrypt.o cudaminer-scrypt-jane.o salsa_kernel.o sha256.o keccak.o fermi_kernel.o kepler_kernel.o test_kernel.o nv_kernel.o nv_kernel2.o titan_kernel.o -lcurl   compat/jansson/libjansson.a -lpthread  -lcudart -fopenmp 
cudaminer-cpu-miner.o: In function `miner_thread':
cpu-miner.c:(.text+0x3708): undefined reference to `scanhash_keccak'
collect2: ld returned 1 exit status
make[2]: *** [cudaminer] Error 1

[bigjme]

I compiled mine about 8 hours ago, there must have been some big updates?

EDIT: i just compiled again, and still getting 34Mhash/s
AizenSou i don't have a clue how your managing those speeds. is there anyway you can pm me your cudaminer so i can see if it runs the same for me?

[instacalm]

I compiled mine about 8 hours ago, there must have been some big updates?

cbuchner1 / CudaMiner
cbuchner1 authored 9 hours ago

[AizenSou]

I compiled mine about 8 hours ago, there must have been some big updates?

I don't see any new commits to the source because Christian is still sleeping. He worked very hard yesterday.
I think the main different lies in the cards itself. Even if the card from the same manufacturer has the different performance. For example my two GTX780s, I bought it the same time from the same manufacturer, one gets max 270khs and another one gets 290khs for Vertcoin.

[sippsnapp]

whoop, is that valid or am i missing something?

[2014-02-06 11:12:06] Total: 133856 khash/s
[2014-02-06 11:12:06] GPU #0: GeForce 9800 GTX/9800 GTX+, 134864 khash/s
[2014-02-06 11:12:06] Total: 134864 khash/s
[2014-02-06 11:12:06] GPU #0: GeForce 9800 GTX/9800 GTX+, 133981 khash/s
[2014-02-06 11:12:06] Total: 133981 khash/s
[2014-02-06 11:12:06] GPU #0: GeForce 9800 GTX/9800 GTX+, 133981 khash/s
[2014-02-06 11:12:06] Total: 133981 khash/s
[2014-02-06 11:12:06] GPU #0: GeForce 9800 GTX/9800 GTX+, 133731 khash/s
[2014-02-06 11:12:06] Total: 133731 khash/s
[2014-02-06 11:12:06] GPU #0: GeForce 9800 GTX/9800 GTX+, 133607 khash/s
[2014-02-06 11:12:06] Total: 133607 khash/s
[2014-02-06 11:12:06] GPU #0: GeForce 9800 GTX/9800 GTX+, 132495 khash/s
[2014-02-06 11:12:06] Total: 132495 khash/s
[2014-02-06 11:12:06] GPU #0: GeForce 9800 GTX/9800 GTX+, 135501 khash/s
[2014-02-06 11:12:06] Total: 135501 khash/s
[2014-02-06 11:12:06] GPU #0: GeForce 9800 GTX/9800 GTX+, 133856 khash/s
[2014-02-06 11:12:06] Total: 133856 khash/s
:~/CudaMiner$ ./cudaminer -a keccak -l K112x40 --benchmark

http://en.wikipedia.org/wiki/GeForce_9_Series#9800M_GTX_.5B6.5D

[bigjme]

yes but we're seeing 70Mhash/s, thats saying i am getting 1/3rd the performance of you which can not be right.
Can you pm me a copy of your cudaminer to try? and perhaps post a screen shot of yours running benchmark as thats a huge achievement

x40 is over the warp limit so its showing values way too high

[DrFingol]

If anyone has successfully compiled the newest code for Windows, please post a guide to how you did it! There's some sweet VTC in it for you if you do!

Christian, you're a godsend to us NVIDIA users! I'm going to donate!

[MarkEMarkEMark]

Hi all - a newbie here.

Firstly, I've just tried the 2014-02-04 version, and I'm getting 320+kh/s on my GTX 670. It was 270 ish with 2013-12-18, and I was already impressed with that.

I notice that you are all talking about even newer versions on Github which may be good for tonight's you-know-who-coin.

I've gone to https://github.com/cbuchner1/CudaMiner, but now I've no idea what to do.

I'm not a programmer, nor am I in any way Linux literate, but I do have an MSDN version of Visual Studio 2010 somewhere.

Would anyone be willing to assist with, or point out a guide to compiling this code for Windows for noobs?

Thanks!

[bigjme]

ok so i just tried to run
./cudaminer --algo=keccak -d gtx780 -L 16 -l K300x32 --benchmark

now thats using 2890MB memory of my 3GB, so all memory is being used, still only 34Mhash/s

[Mav2000]

Same here, new to the whole mining thing and I would love to see something like this for windows...I really have no clue what you guys ae talking about.

[bigjme]

This is for linux and windows. Just there arent as many of us that make it for windows

[sippsnapp]

What exactly is the second number (32) standing for? K300x32

Ive just played around a bit and figured my 9800gtx+ has 128 stream processors, K128/42 is the highest value i an run without errors.
But still like fishing in the dark, ive not really a clue what im doing here ^^.

[bigjme]

Its something christian put. Just to say, x40 is over the limit for the K kernel

[Mav2000]

This is for linux and windows. Just there arent as many of us that make it for windows

Thanks. So you just download all the files and run it or do we have to do anything else?

[bigjme]

You have to compile it. Which is something not many of us set up under windows. There are those of us that do, but most dont

[sippsnapp]

is it possible that the second number ie 32 on K112x42 is the Theoretical Shader Processing Rate which is on the geforce 9800gtx+ 420 ie 42 the which seems the max.

[2014-02-06 11:57:09] Total: 123688 khash/s
[2014-02-06 11:57:09] GPU #0: GeForce 9800 GTX/9800 GTX+, 123586 khash/s
[2014-02-06 11:57:09] Total: 123586 khash/s
[2014-02-06 11:57:09] GPU #0: GeForce 9800 GTX/9800 GTX+, 123993 khash/s
[2014-02-06 11:57:09] Total: 123993 khash/s
[2014-02-06 11:57:09] GPU #0: GeForce 9800 GTX/9800 GTX+, 124198 khash/s
[2014-02-06 11:57:09] Total: 124198 khash/s
[2014-02-06 11:57:09] GPU #0: GeForce 9800 GTX/9800 GTX+, 124096 khash/s
[2014-02-06 11:57:09] Total: 124096 khash/s
[2014-02-06 11:57:09] GPU #0: GeForce 9800 GTX/9800 GTX+, 124506 khash/s
[2014-02-06 11:57:09] Total: 124506 khash/s
[2014-02-06 11:57:09] GPU #0: GeForce 9800 GTX/9800 GTX+, 124609 khash/s
[2014-02-06 11:57:09] Total: 124609 khash/s
[2014-02-06 11:57:09] GPU #0: GeForce 9800 GTX/9800 GTX+, 124816 khash/s
[2014-02-06 11:57:09] Total: 124816 khash/s
~/CudaMiner$ ./cudaminer -a keccak -l K112x42 --benchmark

112= Stream Processors
42.0= Theoretical Shader Processing Rate (Gigaflops)

this would explain my K112 is max and x42 is max???

http://en.wikipedia.org/wiki/GeForce_9_Series#9800M_GTX_.5B6.5D