[ANN][YAC] Yacminer GPU miner for Yacoin

[ivanlabrie]

I get 11,95 kH/s on my 4xR9 280x

In my opinion this is somehow a little?

Or is this normal?

SET GPU_MAX_ALLOC_PERCENT=100
yacminer.exe --scrypt-chacha -o stratum+tcp://yac.coinmine.pl:9088 -u barbados-2 -p 1 --worksize 256 --lookup-gap 4 -g 1 -R 2048 --thread-concurrency 48640 --buffer-size 2656 --auto-fan --temp-cutoff 90 --temp-overheat 85 --gpu-fan 85-95 2>logfile.txt

It's a bit low...try my settings:

yacminer --scrypt-chacha -R 4424 -g 1 --buffer-size 2560 --lookup-gap 7 --worksize 256 --gpu-engine 1020 --gpu-memclock 1500 --temp-cutoff 95 --gpu-fan 0-100 --auto-gpu --temp-target 85 --temp-overheat 95 -o stratum+tcp://yac.coinmine.pl:9088 -u ivanlabrie.1 -p x

[barbados-fs]

I get 11,95 kH/s on my 4xR9 280x

In my opinion this is somehow a little?

Or is this normal?

SET GPU_MAX_ALLOC_PERCENT=100
yacminer.exe --scrypt-chacha -o stratum+tcp://yac.coinmine.pl:9088 -u barbados-2 -p 1 --worksize 256 --lookup-gap 4 -g 1 -R 2048 --thread-concurrency 48640 --buffer-size 2656 --auto-fan --temp-cutoff 90 --temp-overheat 85 --gpu-fan 85-95 2>logfile.txt

It's a bit low...try my settings:

yacminer --scrypt-chacha -R 4424 -g 1 --buffer-size 2560 --lookup-gap 7 --worksize 256 --gpu-engine 1020 --gpu-memclock 1500 --temp-cutoff 95 --gpu-fan 0-100 --auto-gpu --temp-target 85 --temp-overheat 95 -o stratum+tcp://yac.coinmine.pl:9088 -u ivanlabrie.1 -p x

Thanks, I got 18,25 kH/s

What is a good figure for cards 280x?

[ivanlabrie]

3.6kh/s per card in my case.

[meodows92]

Hi i am back again i am using the github from Thirtybird. But still have error as below:



checking minix/config.h usability... no
checking minix/config.h presence... no
checking for minix/config.h... no
checking whether it is safe to define __EXTENSIONS__... yes
checking whether to enable maintainer-specific portions of Makefiles... no
checking for gcc... (cached) gcc
checking whether we are using the GNU C compiler... (cached) yes
checking whether gcc accepts -g... (cached) yes
checking for gcc option to accept ISO C89... (cached) none needed
checking dependency style of gcc... (cached) gcc3
checking for ranlib... ranlib
checking whether gcc needs -traditional... no
checking whether gcc and cc understand -c and -o together... yes
checking for ranlib... (cached) ranlib
checking sys/mman.h usability... yes
checking sys/mman.h presence... yes
checking for sys/mman.h... yes
checking wchar.h usability... yes
checking wchar.h presence... yes
checking for wchar.h... yes
checking for stdint.h... (cached) yes
checking for mprotect... yes
checking for sigaction... yes
checking for sigaltstack... yes
checking for siginterrupt... yes
checking for mmap... yes
checking for MAP_ANONYMOUS... yes
checking whether memchr works... yes
checking whether memmem is declared... yes
checking for memmem... yes
checking whether memmem works... yes
checking for C/C++ restrict keyword... __restrict
checking for uid_t in sys/types.h... yes
checking for inline... inline
checking whether the preprocessor supports include_next... yes
checking whether system header files limit the line length... no
checking for wchar_t... yes
checking for unsigned long long int... yes
checking for long long int... yes
checking whether stdint.h conforms to C99... yes
checking whether memmem is declared without a macro... yes
checking whether mempcpy is declared without a macro... yes
checking whether memrchr is declared without a macro... yes
checking whether rawmemchr is declared without a macro... yes
checking whether stpcpy is declared without a macro... yes
checking whether stpncpy is declared without a macro... yes
checking whether strchrnul is declared without a macro... yes
checking whether strdup is declared without a macro... yes
checking whether strncat is declared without a macro... yes
checking whether strndup is declared without a macro... yes
checking whether strnlen is declared without a macro... yes
checking whether strpbrk is declared without a macro... yes
checking whether strsep is declared without a macro... yes
checking whether strcasestr is declared without a macro... yes
checking whether strtok_r is declared without a macro... yes
checking whether strerror_r is declared without a macro... yes
checking whether strsignal is declared without a macro... yes
checking whether strverscmp is declared without a macro... yes
checking whether memmem works in linear time... yes
checking for memmem... (cached) yes
checking whether memmem works... (cached) yes
checking for struct sigaction.sa_sigaction... yes
checking for volatile sig_atomic_t... yes
checking for sighandler_t... yes
checking whether sigaction is declared without a macro... yes
checking whether sigaddset is declared without a macro... yes
checking whether sigdelset is declared without a macro... yes
checking whether sigemptyset is declared without a macro... yes
checking whether sigfillset is declared without a macro... yes
checking whether sigismember is declared without a macro... yes
checking whether sigpending is declared without a macro... yes
checking whether sigprocmask is declared without a macro... yes
checking for sigprocmask... yes
checking whether NULL can be used in arbitrary expressions... yes
checking for ANSI C header files... (cached) yes
checking syslog.h usability... yes
checking syslog.h presence... yes
checking for syslog.h... yes
checking for size_t... yes
checking for working alloca.h... yes
checking for alloca... yes
checking for OpenCL... no
checking for pthread_create in -lpthread... yes
checking for json_loads in -ljansson... no
checking for library containing addstr... -lncurses
checking for pkg-config... /usr/bin/pkg-config
checking pkg-config is at least version 0.9.0... yes
checking for LIBCURL... no
checking for LIBCURL... yes
configure: creating ./config.status
config.status: creating Makefile
config.status: creating compat/Makefile
config.status: creating compat/jansson/Makefile
config.status: creating ccan/Makefile
config.status: creating lib/Makefile
config.status: creating config.h
config.status: config.h is unchanged
config.status: executing depfiles commands



------------------------------------------------------------------------
yacminer 3.5.0-yac2
------------------------------------------------------------------------


Configuration Options Summary:

  curses.TUI...........: FOUND: -lncurses
  OpenCL...............: NOT FOUND. GPU mining support DISABLED
configure: error: No mining configured in




I am using g2.2xlarge instance in AWS and previously i have installed cuda toolkit and all the necessarily stuff for it to work. But still this problem surface it. Could anyone tell me what should i do next to solve the error above? If this error solved then what to do next to if there is another error? And thanks to Thirtybird for help to reply my post but this time i really need your help to get this miner works. I will be very thankful to you if you could help me.

[Thirtybird]

So....

to understand...

you're trying to compile the GPU miner to run on Amazon virtual servers?

[meodows92]

So....

to understand...

you're trying to compile the GPU miner to run on Amazon virtual servers?

Yes, and may in the further i will use it other graphic card that will be run in Linux operating system. For now i will run it in AWS first.

[Thirtybird]

So....

to understand...

you're trying to compile the GPU miner to run on Amazon virtual servers?

Yes, and may in the further i will use it other graphic card that will be run in Linux operating system. For now i will run it in AWS first.

You should contact Amazon then and tell them you can't access the OpenCL GPU on their virtual servers.

OR

Understand that those servers don't offer OpenCL mining.  You're probably looking for the CPU miner (poke around my github, it's there too)

[meodows92]

Hi finally it able to work as it show below:


------------------------------------------------------------------------
yacminer 3.5.0-yac2
------------------------------------------------------------------------


Configuration Options Summary:

  curses.TUI...........: FOUND: -lncurses
  OpenCL...............: FOUND. GPU mining support enabled
  scrypt...............: Enabled
  ADL..................: SDK NOT found, GPU monitoring support DISABLED

  Avalon.ASICs.........: Disabled
  BFL.ASICs............: Disabled
  BitForce.FPGAs.......: Disabled
  Icarus.FPGAs.........: Disabled
  ModMiner.FPGAs.......: Disabled
  Ztex.FPGAs...........: Disabled

Compilation............: make (or gmake)
  CPPFLAGS.............:
  CFLAGS...............: -g -O2
  LDFLAGS..............:  -lpthread
  LDADD................:  -lcurl   compat/jansson/libjansson.a -lpthread -lOpenCL    -lm

Installation...........: make install (as root if needed, with 'su' or 'sudo')
  prefix...............: /usr/local




But something is wrong which i tried to make it and it does not appeal "yacminer" file. All the files show below:



01-cgminer.rules           ccan              driver-avalon.c    libztex.c          scrypt140202.cl
aclocal.m4                 c++defs.h         driver-avalon.h    libztex.h          scrypt.c
adl.c                      ChangeLog         driver-bflsc.c     linux-usb-cgminer  scrypt-chacha.cl
adl_functions.h            code_flow.txt     driver-bitforce.c  logging.c          scrypt-chacha.cl.14
adl.h                      compat            driver-icarus.c    logging.h          scrypt.h
ADL_SDK                    compat.h          driver-modminer.c  ltmain.sh          scrypt-jane.c
api.c                      compile           driver-opencl.c    m4                 scrypt-jane.h
API.class                  config.guess      driver-opencl.h    Makefile           SCRYPT-README
api-example.c              config.h          driver-ztex.c      Makefile.am        sha2.c
api-example.php            config.h.in       elist.h            Makefile.in        sha2.h
API.java                   config.h.in~      example.conf       miner.h            stamp-h1
API-README                 config.log        findnonce.c        miner.php          usbutils.c
arg-nonnull.h              config.status     findnonce.h        missing            usbutils.h
ASIC-README                config.sub        FPGA-README        NEWS               uthash.h
AUTHORS                    configure         fpgautils.c        ocl.c              util.c
autogen.sh                 configure.ac      fpgautils.h        ocl.h              util.h
autom4te.cache             COPYING           GPU-README         phatk121016.cl     warn-on-use.h
bench_block.h              depcomp           hexdump.c          poclbm130302.cl    windows-build.txt
bitforce-firmware-flash.c  diablo130302.cl   install-sh         README.md          yacminer.c
bitstreams                 diakgcn121016.cl  lib                scrypt130511.cl    yacoin-schedule.txt



But when i use the yacminer 3.3.1 it appealed anybody knows why? And i will be very thankful to you. Another thing i use the yacminer 3.3.1 to mine and i found it very strange with the hash rate show below:




   yacminer version 3.3.1-yac2 - Started: [2014-04-30 13:30:24]
--------------------------------------------------------------------------------
 (5s):80.63M (avg):80.73Mh/s | A:0  R:0  HW:5  U:0.0/m  WU:1.8/m
 ST: 2  SS: 0  NB: 5  LW: 30  GF: 0  RF: 0
 Connected to yac.coinmine.pl diff 1 with stratum as user meodows92.meodows92
 Block: ea4cbc304a599bc8...  Diff:1  Started: [13:32:50]  Best share: 12
--------------------------------------------------------------------------------
 [P]ool management [G]PU management ettings [D]isplay options [Q]uit
 GPU 0:                | 80.90M/81.72Mh/s | A:0 R:0 HW:5 U:0.00/m I: 4
--------------------------------------------------------------------------------

 [2014-04-30 13:30:23] Started yacminer 3.3.1-yac2
 [2014-04-30 13:30:23] Probing for an alive pool
 [2014-04-30 13:30:25] Disabling extra threads due to dynamic mode.
 [2014-04-30 13:30:25] Tune dynamic intensity with --gpu-dyninterval
 [2014-04-30 13:30:25] Network diff set to 1
 [2014-04-30 13:30:26] Network diff set to 1
 [2014-04-30 13:30:26] Stratum from pool 0 detected new block
 [2014-04-30 13:30:26] Thread 1 being disabled
 [2014-04-30 13:31:30] Found block for pool 0!
 [2014-04-30 13:32:27] Found block for pool 0!
 [2014-04-30 13:32:27] Network diff set to 1
 [2014-04-30 13:32:27] Stratum from pool 0 detected new block
 [2014-04-30 13:32:44] Network diff set to 1
 [2014-04-30 13:32:44] Stratum from pool 0 detected new block
 [2014-04-30 13:32:50] Network diff set to 1
 [2014-04-30 13:32:50] Stratum from pool 0 detected new block
 [2014-04-30 13:32:55] Found block for pool 0!


It appeal to be the hash rate in a very large amount is in mega hash, it is true or i have to calculate my formula my own?
Anyways thanks to Thirtybird for your reply previously, although it doesn't help much and i found another way to solve it.

[Thirtybird]

But something is wrong which i tried to make it and it does not appeal "yacminer" file. All the files show below:



01-cgminer.rules           ccan              driver-avalon.c    libztex.c          scrypt140202.cl
aclocal.m4                 c++defs.h         driver-avalon.h    libztex.h          scrypt.c
adl.c                      ChangeLog         driver-bflsc.c     linux-usb-cgminer  scrypt-chacha.cl
adl_functions.h            code_flow.txt     driver-bitforce.c  logging.c          scrypt-chacha.cl.14
adl.h                      compat            driver-icarus.c    logging.h          scrypt.h
ADL_SDK                    compat.h          driver-modminer.c  ltmain.sh          scrypt-jane.c
api.c                      compile           driver-opencl.c    m4                 scrypt-jane.h
API.class                  config.guess      driver-opencl.h    Makefile           SCRYPT-README
api-example.c              config.h          driver-ztex.c      Makefile.am        sha2.c
api-example.php            config.h.in       elist.h            Makefile.in        sha2.h
API.java                   config.h.in~      example.conf       miner.h            stamp-h1
API-README                 config.log        findnonce.c        miner.php          usbutils.c
arg-nonnull.h              config.status     findnonce.h        missing            usbutils.h
ASIC-README                config.sub        FPGA-README        NEWS               uthash.h
AUTHORS                    configure         fpgautils.c        ocl.c              util.c
autogen.sh                 configure.ac      fpgautils.h        ocl.h              util.h
autom4te.cache             COPYING           GPU-README         phatk121016.cl     warn-on-use.h
bench_block.h              depcomp           hexdump.c          poclbm130302.cl    windows-build.txt
bitforce-firmware-flash.c  diablo130302.cl   install-sh         README.md          yacminer.c
bitstreams                 diakgcn121016.cl  lib                scrypt130511.cl    yacoin-schedule.txt

You didn't actually run "make" - there would be a bunch of .o files and a yacminer if you had.

But when i use the yacminer 3.3.1 it appealed anybody knows why? And i will be very thankful to you. Another thing i use the yacminer 3.3.1 to mine and i found it very strange with the hash rate show below:

It appeal to be the hash rate in a very large amount is in mega hash, it is true or i have to calculate my formula my own?
Anyways thanks to Thirtybird for your reply previously, although it doesn't help much and i found another way to solve it.

yacminer 3.3.1 is not supported here - I had no hand in that software.  The parameters are different, you looked like you were mining SHA256

[echo00114]

hello
 sorry disturb again but i cant solved the problem i working many hours the conf and not working

i try my old card radeon 4870 1 gb version set up yacmining but i cant

here my conf
yacminer --scrypt-chacha -o stratum+tcp://yac.coinmine.pl:9088 --worksize 128 -R 3200 --buffer-size 440 -g 1 --lookup-gap 4 --gpu-engine 750 --gpu-memclock 900 --gpu-powertune 0 -u xx -p xx

and many hardvare error i try lower raw intensity bit yacminer disabling the card only work R 1920 but also many hardware error and i try normal intensity -I 9 also hw errors  but speed 670 hs
the error message:  Error -54: Enqueueing kernel onto command queue.     (clEnqueueNDRangeKernel)
i read the readme but i cant understand how correct settings raw intensity
i use ati 13.9 driver last support this card x64 win
the card use 800 shaders
please help me.

thanx again

[Trokhon]

I use a R9 290 Saphire card and get only HW errors. Could anyone post a working config for this card please. I have 32GB RAM in my machine

[masat1404]

Hello
I use a 7870 Saphire card and get only 1kh/s per card in my case.Is this normal?
here my conf
yacminer.exe --scrypt-chacha -o stratum+tcp://yac.coinmine.pl:9088 -u X -p 1 --worksize 256 --lookup-gap 4 -g 1 -R 800 --buffer-size 1760

I have 8GB RAM in my machine.

[Thirtybird]

Hello
I use a 7870 Saphire card and get only 1kh/s per card in my case.Is this normal?
here my conf
yacminer.exe --scrypt-chacha -o stratum+tcp://yac.coinmine.pl:9088 -u X -p 1 --worksize 256 --lookup-gap 4 -g 1 -R 800 --buffer-size 1760

I have 8GB RAM in my machine.

That actually sounds about right for the configuration you posted. 

[masat1404]

Hello
I use a 7870 Saphire card and get only 1kh/s per card in my case.Is this normal?
here my conf
yacminer.exe --scrypt-chacha -o stratum+tcp://yac.coinmine.pl:9088 -u X -p 1 --worksize 256 --lookup-gap 4 -g 1 -R 800 --buffer-size 1760

I have 8GB RAM in my machine.

That actually sounds about right for the configuration you posted. 

As I understood, I can't get more from this card. Am I right?

[Thirtybird]

Hello
I use a 7870 Saphire card and get only 1kh/s per card in my case.Is this normal?
here my conf
yacminer.exe --scrypt-chacha -o stratum+tcp://yac.coinmine.pl:9088 -u X -p 1 --worksize 256 --lookup-gap 4 -g 1 -R 800 --buffer-size 1760

I have 8GB RAM in my machine.

That actually sounds about right for the configuration you posted. 

As I understood, I can't get more from this card. Am I right?

You can try increasing the LG and the -R, but with only 2GB on the card, you really aren't going to be able to get too much more out of it.

[ivanlabrie]

Wonder how these will fare:

Overclockers UK are one of the leading UK retailers for computer hardware, and one of their managers, Gibbo has posted a few very interesting comments on the OCUK forums.

“980 / 970 4GB reference cards with 8GB coming at a later date with custom coolers,” Gibbo began, confirming earlier rumors hinting that we’ll be seeing 8GB Maxwell models after launch. “Can’t say any more but don’t expect a huge leap in performance over current single GPU stuff.” he added.

Regarding the GPU’s performance, “They replace 780, not 780Ti,” said Gibbo, likely to the disappointment of many. “Performance wise, my lips are sealed on how it compares to 780Ti, but of course 980 is quicker than 780 for sure!”

source:
http://www.redgamingtech.com/nvidias-gtx-980-maxwell-slower-gtx-780-ti-8gb-model-coming-later/

UPDATE

At geizhals.at, an EU based price-watch engine started listing a GeForce GTX 980.

The card would be dual-slot in configuration and has two 6-pin power connectors. Other then that noting is listed.
There is a mistake in the listing though as it states GM104 "Maxwell" which would be GM204.

Chip: GM104 "Maxwell"
Memory: 4 GB GDDR5, 256bit
Shader units / TMU / ROP: 2560/160/64
Fab: 28nm
Shader model 5.0
Interface: PCIe 3.0 x16
Slot: dual-slot
Power: 2x 6-pins PCIe
Anyway, nothing is certain with this information.

nothing is certain with this information.



source:
http://www.guru3d.com/news-story/geforce-gtx-980-starts-listing-in-pricewatch-engines.html

[flippo]

Anyone mining YAC with 280x? What's your settings and how much are you getting out of it?

[Kracko]

I built Yacminer against the NVidia implementation of OpenCL just fooling around but I wasn't able to get more than two-thirds of what I could with Cudaminer- but at the same time, I'm not really familiar with tuning yacminer.   Has anyone tried this and achieved results worth mentioning?  It appears that NVidia's implementation of OpenCL limits the memory available to 1/4 of the total.  And also does not support OpenCL 1.2.   Anyone have recommended settings for the GTX 970?  Do I run multiple threads to take advantage of the memory limitations (something like -g 4)?

Gigabyte GTX 970
4GB RAM (1GB accessible by OpenCL)
1644 shading units

Or is this even worth pursuing?

[Thirtybird]

I built Yacminer against the NVidia implementation of OpenCL just fooling around but I wasn't able to get more than two-thirds of what I could with Cudaminer- but at the same time, I'm not really familiar with tuning yacminer.   Has anyone tried this and achieved results worth mentioning?  It appears that NVidia's implementation of OpenCL limits the memory available to 1/4 of the total.  And also does not support OpenCL 1.2.   Anyone have recommended settings for the GTX 970?  Do I run multiple threads to take advantage of the memory limitations (something like -g 4)?

Gigabyte GTX 970
4GB RAM (1GB accessible by OpenCL)
1644 shading units

Or is this even worth pursuing?

Not likely worth pursuing.  CUDA is supposed to be significantly faster on nvidia cards than straight out openCL.  With that said, you could try to run -g 4 or you could launch 4 separate instances of YACMiner. 

You might want to ask specifics about cudaminer/cuda or nvidia settings in the Cudaminer thread - likely more people with nvidia cards there

[Kracko]

That's what I was afraid of.  I tried running 4 separate threads and didn't really improve the results much.  I suppose the only real benefit of NVidia OpenCL would be in the case where there is not a CUDA implementation of the algo, as with Neoscrypt.

I'm good on Cudaminer tweaking and settings.  Translating NVidia settings over to OpenCL is different story.   I was more curious to see how close I could get an NVidia OpenCL implementation of your Yacminer running against Cudaminer.  And also possibly running a NVidia compile of sgminer for algo switching support.  Seems like it'd be more productive to move the API from sgminer to cudaminer.

I built Yacminer against the NVidia implementation of OpenCL just fooling around but I wasn't able to get more than two-thirds of what I could with Cudaminer- but at the same time, I'm not really familiar with tuning yacminer.   Has anyone tried this and achieved results worth mentioning?  It appears that NVidia's implementation of OpenCL limits the memory available to 1/4 of the total.  And also does not support OpenCL 1.2.   Anyone have recommended settings for the GTX 970?  Do I run multiple threads to take advantage of the memory limitations (something like -g 4)?

Gigabyte GTX 970
4GB RAM (1GB accessible by OpenCL)
1644 shading units

Or is this even worth pursuing?

Not likely worth pursuing.  CUDA is supposed to be significantly faster on nvidia cards than straight out openCL.  With that said, you could try to run -g 4 or you could launch 4 separate instances of YACMiner. 

You might want to ask specifics about cudaminer/cuda or nvidia settings in the Cudaminer thread - likely more people with nvidia cards there