SILENTARMY v5: Zcash miner, 115 sol/s on R9 Nano, 70 sol/s on GTX 1070

[jiggytom]

what processor? my celeron can only do about 490. gonna try the cpu fix now.

You can find my working  cpu fix in this thread

yup, that's what I'm using. Getting about 500-510 now. What memory clocks? Does core clock help at all, or do you decrease? I'm running +1550 memory.

[krnlx]

what processor? my celeron can only do about 490. gonna try the cpu fix now.

You can find my working  cpu fix in this thread

yup, that's what I'm using. Getting about 500-510 now. What memory clocks? Does core clock help at all, or do you decrease? I'm running +1550 memory.

My overclock and fans=100% script, X must be running, and set up well (due to shitty nvidia-settings)

And you must run silentarmy with --instances=1 on nvidia

Code:

#samsung mem

CLOCK=0
MEM=1600
CMD='/usr/bin/nvidia-settings'

export DISPLAY=:0

for i in {0..5}
  do
#    nvidia-smi -i ${i} -pm 0
#    nvidia-smi -i ${i} -pl 170
#    nvidia-smi -i ${i} -ac 4004,1911
#    ${CMD} -a [gpu:${i}]/GPUPowerMizerMode=1
#    ${CMD} -a [gpu:${i}]/GPUFanControlState=1
#    ${CMD} -a [fan:${i}]/GPUTargetFanSpeed=100

    for x in {3..3}
      do
echo
        ${CMD} -a [gpu:${i}]/GPUGraphicsClockOffset[${x}]=${CLOCK}
        ${CMD} -a [gpu:${i}]/GPUMemoryTransferRateOffset[${x}]=${MEM}
    done

done


#gpu 0 suck

${CMD} -a [gpu:0]/GPUGraphicsClockOffset[3]=0
${CMD} -a [gpu:0]/GPUMemoryTransferRateOffset[3]=1200

[zawawa]

I'm still working on an Windows version. I tried to remove Cygwin dependency, but, alas, I was unsuccessful. All you need is to install the Cygwin version of Python 3.4, however, so it shouldn't be that bad. I will keep you guys updated.

[jiggytom]

what processor? my celeron can only do about 490. gonna try the cpu fix now.

You can find my working  cpu fix in this thread

yup, that's what I'm using. Getting about 500-510 now. What memory clocks? Does core clock help at all, or do you decrease? I'm running +1550 memory.

My overclock and fans=100% script, X must be running, and set up well (due to shitty nvidia-settings)

And you must run silentarmy with --instances=1 on nvidia

Code:

#samsung mem

CLOCK=0
MEM=1600
CMD='/usr/bin/nvidia-settings'

export DISPLAY=:0

for i in {0..5}
  do
#    nvidia-smi -i ${i} -pm 0
#    nvidia-smi -i ${i} -pl 170
#    nvidia-smi -i ${i} -ac 4004,1911
#    ${CMD} -a [gpu:${i}]/GPUPowerMizerMode=1
#    ${CMD} -a [gpu:${i}]/GPUFanControlState=1
#    ${CMD} -a [fan:${i}]/GPUTargetFanSpeed=100

    for x in {3..3}
      do
echo
        ${CMD} -a [gpu:${i}]/GPUGraphicsClockOffset[${x}]=${CLOCK}
        ${CMD} -a [gpu:${i}]/GPUMemoryTransferRateOffset[${x}]=${MEM}
    done

done


#gpu 0 suck

${CMD} -a [gpu:0]/GPUGraphicsClockOffset[3]=0
${CMD} -a [gpu:0]/GPUMemoryTransferRateOffset[3]=1200

thanks! yea I haven't noticed a big difference running 1 instance or more - get about the same results.

edit - actually i lied, its a tad bit faster. getting 520 now. the rest all chalk up to the oc settings!

[jiggytom]

So is the 1080 just going to be slower because of the GDDR5x, or is there some optimizations possible?

[krnlx]

So is the 1080 just going to be slower because of the GDDR5x, or is there some optimizations possible?

I don't have 1080, can you share your results ?

[jiggytom]

getting about 60-65 sols/s with same exact setup.  I have 1080s mixed in with 1070s. 

[snakey]

4 RX 470 stock with 59~65

RX470 Nitro+ 4GB Elpida (straps 1500, GPU @1260, Mem @2000):

Code:

Total 80.7 sol/s [dev0 78.8] 156 shares
Total 81.4 sol/s [dev0 80.4] 156 shares
Total 81.3 sol/s [dev0 81.0] 156 shares
Total 80.8 sol/s [dev0 80.1] 156 shares
Total 80.3 sol/s [dev0 80.3] 156 shares
Total 79.7 sol/s [dev0 80.4] 156 shares
Total 80.5 sol/s [dev0 80.1] 156 shares

WooHoo!!!

Now the 4GB cards are getting better results then the 8GB cards?

can I use RX 470 boysie for my 8GB card?

[zawawa]

Here comes the Windows port of SILENTARMY v5! Enjoy!

https://mega.nz/#F!iwYQhI6S!2vl0D3jz3fH3-bZxHUurBg

[Relaqs]

Yes Thx.... linux.... noob

[mo35]

Here comes the Windows port of SILENTARMY v5! Enjoy!

https://mega.nz/#F!iwYQhI6S!2vl0D3jz3fH3-bZxHUurBg

by "Make sure to select the "python3" package under "Python"." you mean "Python3: Py3K language interpreter" ? anything else ?

[Relaqs]

Ready for R9 280x?

[jiggytom]

Here comes the Windows port of SILENTARMY v5! Enjoy!

https://mega.nz/#F!iwYQhI6S!2vl0D3jz3fH3-bZxHUurBg

didn't work for me. connected to stratum, but nothing starts mining. 

[zawawa]

Here comes the Windows port of SILENTARMY v5! Enjoy!

https://mega.nz/#F!iwYQhI6S!2vl0D3jz3fH3-bZxHUurBg

by "Make sure to select the "python3" package under "Python"." you mean "Python3: Py3K language interpreter" ? anything else ?

That should be enough.

[ylpkm]

Here comes the Windows port of SILENTARMY v5! Enjoy!

https://mega.nz/#F!iwYQhI6S!2vl0D3jz3fH3-bZxHUurBg

didn't work for me. connected to stratum, but nothing starts mining.  

same here, tried on nvidia and amd card. Amd card works though.

[zawawa]

Here comes the Windows port of SILENTARMY v5! Enjoy!

https://mega.nz/#F!iwYQhI6S!2vl0D3jz3fH3-bZxHUurBg

didn't work for me. connected to stratum, but nothing starts mining. 

Did you install the Cygwin version of Python? The official windows version would not work.

[zawawa]

Here comes the Windows port of SILENTARMY v5! Enjoy!

https://mega.nz/#F!iwYQhI6S!2vl0D3jz3fH3-bZxHUurBg

didn't work for me. connected to stratum, but nothing starts mining. 

same here, tried on nvidia and amd card.

Does sa-solver.exe work? If it does, the problem is definitely with the Python installation.

[mikhan]

testing https://github.com/dagf2101/silentarmy on windows, requires python of course
getting 90+ on 1070 https://forum.z.cash/t/silentarmy-v5-115-sol-s-on-r9-nano-70-sol-s-on-gtx-1070/3580/1163

[elgi76]

2 x R9 Fury + 4 x RX470 :

Total 488.1 sol/s [dev0 106.0, dev1 71.4, dev2 67.0, dev3 63.9, dev4 65.5, dev5 105.4] 1278 shares
Total 490.7 sol/s [dev0 106.8, dev1 71.4, dev2 68.8, dev3 64.5, dev4 67.7, dev5 107.6] 1284 shares
Total 491.6 sol/s [dev0 105.1, dev1 69.7, dev2 69.2, dev3 64.7, dev4 66.7, dev5 106.8] 1287 shares
Total 489.8 sol/s [dev0 106.5, dev1 66.9, dev2 69.3, dev3 65.4, dev4 64.9, dev5 105.9] 1290 shares
Total 490.4 sol/s [dev0 108.3, dev1 64.1, dev2 68.8, dev3 66.5, dev4 67.5, dev5 106.1] 1292 shares
Total 490.7 sol/s [dev0 107.2, dev1 63.3, dev2 69.3, dev3 66.7, dev4 68.5, dev5 110.1] 1297 shares
Total 489.7 sol/s [dev0 104.4, dev1 65.6, dev2 68.3, dev3 63.2, dev4 65.8, dev5 111.7] 1302 shares
Total 490.6 sol/s [dev0 106.4, dev1 66.0, dev2 67.1, dev3 65.4, dev4 66.2, dev5 109.2] 1305 shares
Total 492.4 sol/s [dev0 107.5, dev1 68.9, dev2 65.1, dev3 65.3, dev4 66.1, dev5 107.9] 1309 shares
Total 493.3 sol/s [dev0 109.2, dev1 68.0, dev2 66.9, dev3 68.9, dev4 66.1, dev5 107.2] 1311 shares
Total 490.7 sol/s [dev0 109.3, dev1 64.7, dev2 68.0, dev3 69.5, dev4 67.3, dev5 108.4] 1313 shares

Total 539.9 sol/s [dev0 115.2, dev1 82.5, dev2 78.3, dev3 74.0, dev4 78.4, dev5 109.3] 942 shares
Total 540.2 sol/s [dev0 115.4, dev1 79.8, dev2 73.1, dev3 75.2, dev4 80.8, dev5 112.8] 946 shares
Total 538.9 sol/s [dev0 112.9, dev1 82.3, dev2 71.0, dev3 73.9, dev4 79.5, dev5 114.8] 954 shares
Total 538.9 sol/s [dev0 113.0, dev1 82.1, dev2 69.9, dev3 74.1, dev4 79.0, dev5 111.6] 955 shares
Total 539.3 sol/s [dev0 114.2, dev1 79.6, dev2 72.6, dev3 77.1, dev4 78.3, dev5 112.8] 955 shares
Total 540.9 sol/s [dev0 113.7, dev1 79.4, dev2 73.2, dev3 78.1, dev4 81.8, dev5 113.0] 957 shares
Total 540.0 sol/s [dev0 112.1, dev1 77.1, dev2 73.0, dev3 75.7, dev4 82.0, dev5 114.2] 959 shares
Total 539.9 sol/s [dev0 111.9, dev1 72.3, dev2 72.6, dev3 75.1, dev4 82.2, dev5 117.5] 962 shares
Total 542.5 sol/s [dev0 111.4, dev1 74.7, dev2 75.0, dev3 81.2, dev4 82.9, dev5 117.9] 964 shares

[ylpkm]

Does sa-solver.exe work? If it does, the problem is definitely with the Python installation.

sa solver works using amd card, but doesn't work using my nvidia card. gtx 1080 gives openCL build failed (-11)