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

[Amph]

Thanks for your work, djm  

Has anyone tried deepcoin with the new version? what kind of speeds do you get for a 750Ti?


**edit: I'm getting about 8000k with a 750ti and 6500k with 750 non-ti Havent tried OC yet

there is no exchange for that coin, what's the point? just testing?

[1714730667]

1714730667

[1714730667]

1714730667

[1714730667]

1714730667

[1714730667]

1714730667

[1714730667]

1714730667

[mmmaybe]

Thanks for your work, djm  

Has anyone tried deepcoin with the new version? what kind of speeds do you get for a 750Ti?


**edit: I'm getting about 8000k with a 750ti and 6500k with 750 non-ti Havent tried OC yet

there is no exchange for that coin, what's the point? just testing?

Yup, always interesting to see how we are doing compared to amd 

[ltcnim]

can't compile under Ubuntu 14.04:

Code:

ptxas fatal   : Unresolved extern function 'vprintf'
make[2]: *** [x13/cuda_mul.o] Error 255

I guess djm34 forgets to fix it. Here is my fixes:

1. Comment out the printf line 107-109 @ x13/cuda_mul.cu

2. //typedef unsigned long long uint64_t;  
@ line 35 @ qubit/qubit_luffa512.cu
@ line 55 @ x13/cuda_x13_hamsi512.cu
->  invalid redeclaration

there is no type redefinition in qubit_luffa512 (at least not in the current version)

I updated the files on github (made a change in makefile rather than commenting out the printf... as it is there on purpose, not working well actually...)

hi djm34, thanks for the latest fix which resolves the vprintf issue, but unfortunately it's now stuck at another place:

Code:

g++: error: @MPIRLIB@: No such file or directory
make[2]: *** [ccminer] Error 1

I compiled & installed mpir 2.6.0 from source and header & .so files are located under /usr/local/

[djm34]

new version compatible with low vram device:
https://mega.co.nz/#!cEMAja7S!TtFipbqNKZ4Mf0f7I6Fgx3bKVpOLXJOwbCK0JJZr5R0

It should be equivalent or even slightly better also the standard 750ti.
I decreased the number of thread on deepcoin, it should decrease a bit screen lag (might impact the perf though... )

[djm34]

can't compile under Ubuntu 14.04:

Code:

ptxas fatal   : Unresolved extern function 'vprintf'
make[2]: *** [x13/cuda_mul.o] Error 255

I guess djm34 forgets to fix it. Here is my fixes:

1. Comment out the printf line 107-109 @ x13/cuda_mul.cu

2. //typedef unsigned long long uint64_t; 
@ line 35 @ qubit/qubit_luffa512.cu
@ line 55 @ x13/cuda_x13_hamsi512.cu
->  invalid redeclaration

there is no type redefinition in qubit_luffa512 (at least not in the current version)

I updated the files on github (made a change in makefile rather than commenting out the printf... as it is there on purpose, not working well actually...)

hi djm34, thanks for the latest fix which resolves the vprintf issue, but unfortunately it's now stuck at another place:

Code:

g++: error: @MPIRLIB@: No such file or directory
make[2]: *** [ccminer] Error 1

I compiled & installed mpir 2.6.0 from source and header & .so files are located under /usr/local/

try addind -lmpir directly at the end of this line in makefile.am

Code:

ccminer_LDADD		= @LIBCURL@ @JANSSON_LIBS@ @PTHREAD_LIBS@ @WS2_LIBS@ @CUDA_LIBS@ @OPENMP_CFLAGS@ @LIBS@ -lmpir

Not sure if it will work

[ltcnim]

can't compile under Ubuntu 14.04:

Code:

ptxas fatal   : Unresolved extern function 'vprintf'
make[2]: *** [x13/cuda_mul.o] Error 255

I guess djm34 forgets to fix it. Here is my fixes:

1. Comment out the printf line 107-109 @ x13/cuda_mul.cu

2. //typedef unsigned long long uint64_t; 
@ line 35 @ qubit/qubit_luffa512.cu
@ line 55 @ x13/cuda_x13_hamsi512.cu
->  invalid redeclaration

there is no type redefinition in qubit_luffa512 (at least not in the current version)

I updated the files on github (made a change in makefile rather than commenting out the printf... as it is there on purpose, not working well actually...)

hi djm34, thanks for the latest fix which resolves the vprintf issue, but unfortunately it's now stuck at another place:

Code:

g++: error: @MPIRLIB@: No such file or directory
make[2]: *** [ccminer] Error 1

I compiled & installed mpir 2.6.0 from source and header & .so files are located under /usr/local/

try addind -lmpir directly at the end of this line in makefile.am

Code:

ccminer_LDADD		= @LIBCURL@ @JANSSON_LIBS@ @PTHREAD_LIBS@ @WS2_LIBS@ @CUDA_LIBS@ @OPENMP_CFLAGS@ @LIBS@ -lmpir

Not sure if it will work

that did the trick, thanks a lot!

for the record, old version from catia @ 5x750Ti (EVGA SC, stock):

19100khash @ 330W

your version:

23500khash @ 360W

[RavenXBR]

6rig - G1620 + H61 Pro BTC + 4Gb DDR3 + TT 550W + 6xMSI 750Ti 2Gb
Works 1 month, the temperature does not rise above 70 degrees

Pretty awesome, congratz. 

[RavenXBR]

binaries from the new m7/deepcoin release:
https://mega.co.nz/#!FIsUXDSA!_tcR1JhLDGQbpcCt078dWFj6fz7K8IThZc3HAph0SEY

Concerning my addresses, I can confirm right now, the BTC, XCN, JPC addresses for the others I need to check...

please don't send me your BBL 

Hashrate improved from around 4.5 to 4.8MHs on 750Ti on my 5 gpu rig.
+135/+200 OC on Gibabyte cards.

Cards running around 55ºC and rig consuming 350W at wall.



Let's see if it will be stable as previous version.

Big thanks again DJM... 

[Schleicher]

new source with M7 (at last,  ) and also new deepcoin algo.

To be able to compile it you will need to instal and compile mpir-2.6.0
(placed in the same directory as the other dependencies  (eg: openssl, curl etc...))
go to build.vc10, click on mpir.sln, then build dll_mpir.gc (or lib_mpir_cxx) for Win32 Release.

Then you will need to move "mpir.h" to build.vc10/Win32/Release
(I had to move it to avoid conflict between other files present in that package...)

Once it is done, it should compile fine. (I just tried it, it seems to work).
You will also need to be able to run ccminer mpir.dll that you will have compiled.
[...]

For Visual Studio you have an absolute path for the mpir include in the release version.
The path for the debug version looks ok.

[djm34]

new source with M7 (at last,  ) and also new deepcoin algo.

To be able to compile it you will need to instal and compile mpir-2.6.0
(placed in the same directory as the other dependencies  (eg: openssl, curl etc...))
go to build.vc10, click on mpir.sln, then build dll_mpir.gc (or lib_mpir_cxx) for Win32 Release.

Then you will need to move "mpir.h" to build.vc10/Win32/Release
(I had to move it to avoid conflict between other files present in that package...)

Once it is done, it should compile fine. (I just tried it, it seems to work).
You will also need to be able to run ccminer mpir.dll that you will have compiled.
[...]

For Visual Studio you have an absolute path for the mpir include in the release version.
The path for the debug version looks ok.

yes, just saw that... I thought I fixed that for the release (since I don't use the debug version..) but only the debug version got fixed...

[Skarfur]

Thanks again djm for your efforts  .
Sent you 4000 XCN as appreciation for your work.

[djm34]

Thanks again djm for your efforts  .
Sent you 4000 XCN as appreciation for your work.

thanks a lot,
cheers,
djm

[ilovecudacompute]

Thanks again djm for your efforts  .
Sent you 4000 XCN as appreciation for your work.

That is  what i call a donation

I tested all djm versions for my Asus 760 and XCN mining
v5 is still the fastest for you out there with the 760 for XCN mining

Deepcoin i am at 8MH with the 760

[CapnBDL]

Hi,

I have a ccminer v1.2 and one CUDA dated 2/28 and would like to know if the djm m7v6 is the same one or newer or someone elses,etc. Guess I'm confused as to what and whos. Also, plz explain what the low mem is? It was just the .exe file downloaded and need to know what files need to be in the DIR for it to work.

This damn GTX650 (nonTi) still locks on me. anyone have any info on this card besides that it's junk? I was hoping for better mining performance, but it doesn't even run CUDA. Don't see any difference on desktop/web speed from the GT6402G.   It's not the Ti, just a GTX6502G. It's all I have as funding doesn't allow a purchase.  Info would be nice. Thanks

[cayars]

nvMiner 1.2-D8 has been released.

Includes Deep, GoalCoin and M7/CryptoNite

Includes small speedups for some other algos.

www.cudamining.co.uk/url/releases/member/8

Carlo

[jpouza]

nvMiner 1.2-D8 has been released.

Includes Deep, GoalCoin and M7/CryptoNite

Includes small speedups for some other algos.

www.cudamining.co.uk/url/releases/member/8

Carlo

Hi!

I´ve downloaded this latest version and testing, so far so good, just a few noob questions cause I was used to mine with ccminer:

1) Scrypt-N
scrypt   Hashrate: 10.93 GH/s   Mining: litecoin   vsLTC: 88%
nscrypt   Hashrate: 424.38 MH/s   Mining: vertcoin   vsLTC: 304%
x11   Hashrate: 21.28 GH/s   Mining: darkcoin   vsLTC: 170%


Scrypt-N looks better than X11 in WafflePool, mining it is possible with this version? What´s the ideal command line for a 750Ti? Scrypt-N consumes a lot more than X11? I´m not used with Scrypt-N mining.


2)What is this counter for found blocks? (50, 49, 48, 47 etc...??) / CCminer didn´t had this.

X11 mining:

[2014-08-24 22:42:42] Stratum detected new block
[2014-08-24 22:42:42] GPU #2: GeForce GTX 750 Ti, 2726 khash/s
[2014-08-24 22:42:42] GPU #0: GeForce GTX 750 Ti, 2648 khash/s
[2014-08-24 22:42:42] GPU #3: GeForce GTX 750 Ti, 2681 khash/s
[2014-08-24 22:42:42] GPU #1: GeForce GTX 750 Ti, 2642 khash/s
[2014-08-24 22:42:56] GPU #1: GeForce GTX 750 Ti, 2631 khash/s
[2014-08-24 22:42:56] accepted: 1/1 (100.00%), 10687 kh/s (10687-50) - Yay
[2014-08-24 22:43:31] Stratum detected new block
[2014-08-24 22:43:31] GPU #3: GeForce GTX 750 Ti, 2676 khash/s
[2014-08-24 22:43:31] GPU #1: GeForce GTX 750 Ti, 2613 khash/s
[2014-08-24 22:43:31] GPU #0: GeForce GTX 750 Ti, 2648 khash/s
[2014-08-24 22:43:31] GPU #2: GeForce GTX 750 Ti, 2728 khash/s
[2014-08-24 22:43:33] GPU #3: GeForce GTX 750 Ti, 2608 khash/s
[2014-08-24 22:43:33] accepted: 2/2 (100.00%), 10597 kh/s (10642-49) - Yay
[2014-08-24 22:44:22] GPU #0: GeForce GTX 750 Ti, 2591 khash/s
[2014-08-24 22:44:22] accepted: 3/3 (100.00%), 10540 kh/s (10608-48) - Yay
[2014-08-24 22:44:31] GPU #1: GeForce GTX 750 Ti, 2587 khash/s
[2014-08-24 22:44:32] GPU #2: GeForce GTX 750 Ti, 2683 khash/s
[2014-08-24 22:44:33] GPU #3: GeForce GTX 750 Ti, 2634 khash/s
[2014-08-24 22:44:58] Stratum detected new block
[2014-08-24 22:44:58] GPU #0: GeForce GTX 750 Ti, 2647 khash/s
[2014-08-24 22:44:58] GPU #2: GeForce GTX 750 Ti, 2726 khash/s
[2014-08-24 22:44:58] GPU #3: GeForce GTX 750 Ti, 2679 khash/s
[2014-08-24 22:44:58] GPU #1: GeForce GTX 750 Ti, 2631 khash/s
[2014-08-24 22:45:36] GPU #1: GeForce GTX 750 Ti, 2546 khash/s
[2014-08-24 22:45:37] accepted: 4/4 (100.00%), 10598 kh/s (10605-47) - Yay
[2014-08-24 22:45:52] GPU #0: GeForce GTX 750 Ti, 2592 khash/s
[2014-08-24 22:45:52] accepted: 5/5 (100.00%), 10543 kh/s (10593-46) - Yay
[2014-08-24 22:45:59] GPU #3: GeForce GTX 750 Ti, 2638 khash/s
[2014-08-24 22:45:59] GPU #2: GeForce GTX 750 Ti, 2686 khash/s
[2014-08-24 22:46:08] GPU #2: GeForce GTX 750 Ti, 2555 khash/s
[2014-08-24 22:46:08] accepted: 6/6 (100.00%), 10330 kh/s (10549-45) - Yay
[2014-08-24 22:46:35] GPU #1: GeForce GTX 750 Ti, 2605 khash/s
[2014-08-24 22:46:51] GPU #0: GeForce GTX 750 Ti, 2625 khash/s
[2014-08-24 22:46:58] GPU #3: GeForce GTX 750 Ti, 2660 khash/s
[2014-08-24 22:47:05] GPU #2: GeForce GTX 750 Ti, 2722 khash/s
[2014-08-24 22:47:36] GPU #1: GeForce GTX 750 Ti, 2587 khash/s
[2014-08-24 22:47:38] Stratum detected new block
[2014-08-24 22:47:38] GPU #0: GeForce GTX 750 Ti, 2591 khash/s
[2014-08-24 22:48:38] GPU #0: GeForce GTX 750 Ti, 2586 khash/s
[2014-08-24 22:48:39] GPU #3: GeForce GTX 750 Ti, 2648 khash/s
[2014-08-24 22:48:39] GPU #1: GeForce GTX 750 Ti, 2587 khash/s
[2014-08-24 22:48:39] GPU #2: GeForce GTX 750 Ti, 2695 khash/s
[2014-08-24 22:48:41] Stratum detected new block
[2014-08-24 22:48:41] GPU #3: GeForce GTX 750 Ti, 2682 khash/s
[2014-08-24 22:48:41] GPU #1: GeForce GTX 750 Ti, 2642 khash/s
[2014-08-24 22:48:41] GPU #2: GeForce GTX 750 Ti, 2716 khash/s
[2014-08-24 22:48:41] GPU #0: GeForce GTX 750 Ti, 2649 khash/s
[2014-08-24 22:49:41] GPU #2: GeForce GTX 750 Ti, 2724 khash/s
[2014-08-24 22:49:41] GPU #3: GeForce GTX 750 Ti, 2675 khash/s
[2014-08-24 22:49:41] GPU #0: GeForce GTX 750 Ti, 2642 khash/s
[2014-08-24 22:49:41] GPU #1: GeForce GTX 750 Ti, 2623 khash/s
[2014-08-24 22:50:13] GPU #3: GeForce GTX 750 Ti, 2617 khash/s
[2014-08-24 22:50:13] accepted: 7/7 (100.00%), 10605 kh/s (10557-44) - Yay
[2014-08-24 22:50:42] GPU #2: GeForce GTX 750 Ti, 2694 khash/s
[2014-08-24 22:50:42] GPU #0: GeForce GTX 750 Ti, 2619 khash/s
[2014-08-24 22:50:42] GPU #1: GeForce GTX 750 Ti, 2592 khash/s
[2014-08-24 22:51:11] GPU #3: GeForce GTX 750 Ti, 2687 khash/s

Tks in advance!

[DougB62]

2)What is this counter for found blocks? (50, 49, 48, 47 etc...??) / CCminer didn´t had this.

The counter is showing the average speed for your miner/s - It counts the first fifty to let you know that there aren't really enough to get a good average yet, and once it counts down all the way, the blockcount disappears, and only the avg speed is displayed.

[cayars]

Question 1:
Not included in nvMiner.
nvMiner for now only includes all known ccminer algos that all of the different devs have released.
But it doesn't include the algos in CudaMiner with the exception of keccak

Question 2:
The number is the "average" hash rate submitted.  It is more useful on some algos as the hash rate fluctuates a lot.
Inside the parens is a number dash number or just a number.
When there is a dash you will see the number to the right decrease for each submitted share.
Typically you won't see the "average" hash rate stabilize until this number gets close to zero. But it depends on the algo you are running.

Hope that helps,
Carlo

EDIT: DougB62 responded before me and I think his explanation is better.

[750Ti 1GB]

nvMiner 1.2-D8 has been released.

Includes Deep, GoalCoin and M7/CryptoNite

Includes small speedups for some other algos.

www.cudamining.co.uk/url/releases/member/8

Carlo

The 750 Ti 1 GB model is not working

[cayars]

On what coin?

wow, there is a 1GB 750ti model?