Open Source ZEC (ZCash) GPU Miner AMD & NVidia (up to 45 sol/s on RX480)

[adaseb]

The 280X was a very good GPU, almost as fast as 290 in many algos.

When ETH launched, it was almost as fast as a 290, however since then its mostly been going downhill.

9.6 sols on 280x vaporx (win7, 15.12 drivers)

If the 290 gets like 18sol/s it should be at least 15sol/s or so.

Back in the Scrypt days it got 750Kh/s while the 290s got 900Kh/s.

When ETH came out it got 27MH/s while 290s got 31Mh/s

[1715365298]

1715365298

[1715365298]

1715365298

[1715365298]

1715365298

[1715365298]

1715365298

[ivcelmik]

The 280X was a very good GPU, almost as fast as 290 in many algos.

When ETH launched, it was almost as fast as a 290, however since then its mostly been going downhill.

9.6 sols on 280x vaporx (win7, 15.12 drivers)

The same hasrate like Nvidia 750Ti ) at 30 W

[antantti]

Testing on W7 and 4 x 970, looks like it needs some CPU with this setup. Maxes one core on i5 hashwell @4.4GH, give it two cores and it uses 40-50% on both of them.

sols 17-21 depending on settings, cards with elpida doing better than samsung. TDP in 60-70% range.

[Gaglam]

The 280X was a very good GPU, almost as fast as 290 in many algos.

When ETH launched, it was almost as fast as a 290, however since then its mostly been going downhill.

9.6 sols on 280x vaporx (win7, 15.12 drivers)

If the 290 gets like 18sol/s it should be at least 15sol/s or so.

Back in the Scrypt days it got 750Kh/s while the 290s got 900Kh/s.

When ETH came out it got 27MH/s while 290s got 31Mh/s

On ETH you get about 16MH/s on ETH today.
Zcash and 280x gives me about 9,6 sols too

[woodaxe]

just run your miner on my ether modded rig RX470 nitro+OC 8 gig

and i am getting between 19/20 sol per card

4 card rig pulling 410 watts at the wall for everything

really good  only problem i have is its not stratum so i will not be using it. As i like to choose
my own pool and tbh i dont know how good  http://coinsforall.io/#!/coin/ZCASH

[Amph]

if i set 0 for the devices to be used, it still use the device, there is a bug maybe?

[tZKtdvt6]

It is simply wonderful!.
Thank you very much.

[stanko10]

MSI R9 380 ??

[mogs]

When running the cmake - i get errors. Any help on this would be appreciated
Checking CMakeError.log I get this -


}

Determining if the function pthread_create exists in the pthreads failed with the following output:
Change Dir: /root/xpm/xpmclient/CMakeFiles/CMakeTmp

Run Build Command:"/usr/bin/make" "cmTryCompileExec1919454905/fast"
/usr/bin/make -f CMakeFiles/cmTryCompileExec1919454905.dir/build.make CMakeFiles/cmTryCompileExec1919454905.dir/build
make[1]: Entering directory `/root/xpm/xpmclient/CMakeFiles/CMakeTmp'
/usr/bin/cmake -E cmake_progress_report /root/xpm/xpmclient/CMakeFiles/CMakeTmp/CMakeFiles 1
Building C object CMakeFiles/cmTryCompileExec1919454905.dir/CheckFunctionExists.c.o
/usr/bin/cc   -DCHECK_FUNCTION_EXISTS=pthread_create   -o CMakeFiles/cmTryCompileExec1919454905.dir/CheckFunctionExists.c.o   -c$
Linking C executable cmTryCompileExec1919454905
/usr/bin/cmake -E cmake_link_script CMakeFiles/cmTryCompileExec1919454905.dir/link.txt --verbose=1
/usr/bin/cc   -DCHECK_FUNCTION_EXISTS=pthread_create    CMakeFiles/cmTryCompileExec1919454905.dir/CheckFunctionExists.c.o  -o cm$
/usr/bin/ld: cannot find -lpthreads
collect2: error: ld returned 1 exit status
make[1]: *** [cmTryCompileExec1919454905] Error 1
make[1]: Leaving directory `/root/xpm/xpmclient/CMakeFiles/CMakeTmp'
make: *** [cmTryCompileExec1919454905/fast] Error 2

[IOTUSA]

20-23 sols on Gigabyte 1070 G1

it seems ok, but power usage is at 50%

what are your OC numbers?

same results:

GPU 0 found share
Share accepted.
Work received: height=2369 diff=25916.523 latency=37ms
[GPU 0] T=-1C A=-1% sols=21.824
[GPU 1] T=-1C A=-1% sols=21.580
[GPU 2] T=-1C A=-1% sols=21.857
[GPU 3] T=-1C A=-1% sols=20.580
(ST/INV/DUP): 17x 0ch(0/0/0)
GPU 3 found share


root@kopiemtu:~# nvidia-smi | grep %
| 13%   57C    P2    64W / 170W |    623MiB /  8113MiB |     99%      Default |
| 14%   57C    P2    89W / 170W |    623MiB /  8113MiB |     94%      Default |
| 13%   56C    P2    70W / 170W |    623MiB /  8113MiB |     93%      Default |
| 11%   56C    P2    81W / 170W |    623MiB /  8113MiB |     95%      Default |
root@kopiemtu:~#


seems to be a lot of place for nVidia improvements...

can you share your config? i have a problem running six gpu with the cpu hanging at 99%

AMD are showing roughly the same, an entire rig of 380 cards run at 400w or so, meaning there is at least 100% improvement to make before they are even saturated or close to.

[t2yax]

there is problem with device choice settings.im laptop user.i wanna try my nvidia card.but it selects intel gpu everytime

i already played with cardplatform and use device settings

[adaseb]

Nicehash devs had a break-through with their OpenCL AMD miner.

R9 380


[2016-10-26 20:28:13.616267][0x00002b58]: Benchmark done!
[2016-10-26 20:28:13.620769][0x00002b58]: Total time : 13150 ms
[2016-10-26 20:28:13.623772][0x00002b58]: Total hashes: 200
[2016-10-26 20:28:13.626273][0x00002b58]: Total solutions found: 329
[2016-10-26 20:28:13.629275][0x00002b58]: Speed: 15.2091 I/s
[2016-10-26 20:28:13.632286][0x00002b58]: Speed: 25.019 Sols/s

[majestymage]

hi can someone help me? when i run windows miner i get this errors
OpenCL error: -5 at /build/xpmclient/src/zcash/zcashgpuclient.cpp:112
OpenCL error: -5 at /build/xpmclient/src/zcash/../opencl.h:87
OpenCL error: -4 at /build/xpmclient/src/zcash/zcashgpuclient.cpp:112
OpenCL error: -4 at /build/xpmclient/src/zcash/zcashgpuclient.cpp:379
what should i do? i have install latest drivers and cuda

i am getting the same error...so..solutions please?

[adaseb]

Anyone managed to get any older GPUs working like the Radeon 6990 ?

[tomothy]

Tried with a 6950; having difficulty.
Works on my laptop tho; booted right up hashes at 2.5/s lol

[mogs]

Getting there but just encountered this,

binsize = 1165300 bytes
prepare_adl success
GPU 0 iAdapterIndex 0 strUDID 512:26522:4098:9548:5208 iBusNumber 2 iDeviceNumber 0 iFunctionNumber 0 iVendorID 4098 strAdapterName AMD Radeon HD 7900 Series
GPU 1 iAdapterIndex 6 strUDID 1024:26522:4098:9548:5208 iBusNumber 4 iDeviceNumber 0 iFunctionNumber 0 iVendorID 4098 strAdapterName AMD Radeon HD 7900 Series
GPU 2 iAdapterIndex 12 strUDID 1280:26522:4098:9548:5208 iBusNumber 5 iDeviceNumber 0 iFunctionNumber 0 iVendorID 4098 strAdapterName AMD Radeon HD 7900 Series
GPU 0 AMD Radeon HD 7900 Series hardware monitoring enabled
GPU 1 AMD Radeon HD 7900 Series hardware monitoring enabled
GPU 2 AMD Radeon HD 7900 Series hardware monitoring enabled
N=12 SIZE=4096 STRIPES=420 WIDTH=20 PCOUNT=40960 TARGET=10
GPU 0: has 28 CUs
Existing OpenCL kernel (kernel.bin) incompatible with configuration
Please remove kernel.bin file and restart miner

Removed kernel.bin / restarted with same result.

Thanks

[Tmdz]

Nicehash devs had a break-through with their OpenCL AMD miner.

R9 380


[2016-10-26 20:28:13.616267][0x00002b58]: Benchmark done!
[2016-10-26 20:28:13.620769][0x00002b58]: Total time : 13150 ms
[2016-10-26 20:28:13.623772][0x00002b58]: Total hashes: 200
[2016-10-26 20:28:13.626273][0x00002b58]: Total solutions found: 329
[2016-10-26 20:28:13.629275][0x00002b58]: Speed: 15.2091 I/s
[2016-10-26 20:28:13.632286][0x00002b58]: Speed: 25.019 Sols/s

Woah where is the opencl version?  or is it not released yet?

[AltcoinScamfinder]

Nicehash devs had a break-through with their OpenCL AMD miner.

R9 380


[2016-10-26 20:28:13.616267][0x00002b58]: Benchmark done!
[2016-10-26 20:28:13.620769][0x00002b58]: Total time : 13150 ms
[2016-10-26 20:28:13.623772][0x00002b58]: Total hashes: 200
[2016-10-26 20:28:13.626273][0x00002b58]: Total solutions found: 329
[2016-10-26 20:28:13.629275][0x00002b58]: Speed: 15.2091 I/s
[2016-10-26 20:28:13.632286][0x00002b58]: Speed: 25.019 Sols/s

Woah where is the opencl version?  or is it not released yet?

I didn't think they were done with it yet? 0.2.x is just nVidia and CPU right?

[krnlx]

@eXtremal

have 100% cpu load on 6x1070 rig. (2 cores celeron) hashrate is 22-23 per card

Code:

 9951 krnl      20   0 53.309g 598484 479196 S 196.3 15.0  30:27.23 zcashgpuclient 

when running on one card hashrate is 26 on card

Code:

10674 krnl      20   0 29.165g 118208  93928 S  95.3  3.0   0:08.60 zcashgpuclient

so number of threads = number of cards and all of them loads 100% cpu and become a bottleneck on multigpu rig.

Can you fix it?

[jk_14]

can you share your config? i have a problem running six gpu with the cpu hanging at 99%

sure:

Code:

root@kopiemtu:~/TEST/zcashclient-0.1.1# cat config.txt
server = "coinsforall.io";
port = "6668";

# Your ZCASH payout address
address = "zthTb2ToG5kuAc11Ni6NxoYjoxQdEr2XvST7XEdLvmHG9k7kotEcwYCaAn1SH4zRdPoXn9kNeFm8CQok9SNtNgn8d9WcdTK";

# You can give this machine a custom name
# Computer name is used as default
# name = "worker_name";

# Platform:
#  "amd": AMD GCN cards
#  "nvidia": NVidia Maxwell cards
platform = "nvidia";

# GPU crash handling (properly untested feature):
# 0 - do nothing
# 1 - close miner application
# 2 - reboot system (need superuser rights)
onCrash = "0";

# - The following settings have one entry per GPU
# - If you have more than 4 just add more entries or leave it as is to use defaults
# - All entries must be separated by commas and set between double quotes

# 0 = don't use device
# 1 = use device
devices =        ["1", "1", "1", "1"];

# work sizes
worksizes = ["256", "256", "256", "256"];

# threads
threads = ["8192", "8192", "8192", "8192"];

# -1 means don't change
corefreq =       ["-1", "-1", "-1", "-1"];
memfreq =        ["-1", "-1", "-1", "-1"];
powertune =      ["-1", "-1", "-1", "-1"];
fanspeed =       ["-1", "-1", "-1", "-1"];
root@kopiemtu:~/TEST/zcashclient-0.1.1#

KopiemTu 3.0 image is ready-to-go with zcashgpuclient, just: boot, download zcash and run
so everyone with nVidia GPU problems may verify if there is also a problem under usb linux distro
(if problem is OS or HW related)