toptek
Legendary
Offline
Activity: 1274
Merit: 1000
|
|
November 14, 2016, 03:43:39 AM |
|
If open source mining software is going to work and keep up with other fee paying mining software we must show support,I suggest with next release everyone mines to the developers zcash address for first hour at lest and show support for the great work.
been dong it .
|
|
|
|
zawawa
Sr. Member
Offline
Activity: 728
Merit: 304
Miner Developer
|
|
November 14, 2016, 03:46:53 AM |
|
If open source mining software is going to work and keep up with other fee paying mining software we must show support,I suggest with next release everyone mines to the developers zcash address for first hour at lest and show support for the great work.
Yeah, I think it's totally worth it to put up a serious open source competition against Sir. Claymore We shall be back.
|
Gateless Gate Sharp, an open-source ETH/XMR miner: http://bit.ly/2rJ2x4VBTC: 1BHwDWVerUTiKxhHPf2ubqKKiBMiKQGomZ
|
|
|
adamvp
|
|
November 14, 2016, 04:24:34 AM |
|
I am always testing some time with built-in .bat
|
I am looking for signature campaign pm me
|
|
|
ivcelmik
|
|
November 14, 2016, 05:51:57 AM |
|
I am always testing some time with built-in .bat
Guys can someone tell me if i can mine with more tha 3+Nvidia cards per rig without hasrate decrease from the CPU usage problem thx in advance
|
|
|
|
oslak
|
|
November 14, 2016, 06:20:29 AM |
|
With the Linux binary releases I make I also included a "mine" script that will mine to my mph account. I just checked the balance, and it's .0035 ZEC, or ~42c. Who says open-source software work doesn't pay. :-)
I'm mining with single core to mrb's account for a few days, post your address and I'll put another core for yours 'nerdralph' is my miningpoolhub account, or with a different pool use: t1MaD62DS2bbzxCuZHnkB8g5Yc2wbeT1g9z Thanks in advance. Not that I need the money, but it's nice to know when someone likes the work enough to make a small donation. Usually when testing the miner, it is run on the default setting and mined to whoever account is pre-set. It will be 1 zec later on.
|
|
|
|
spyder09
Member
Offline
Activity: 60
Merit: 10
|
|
November 14, 2016, 07:09:30 AM |
|
does v5 have a windows version? I thought it was supposed to be released? if I missed it sorry, thanks! looks like great work!
|
|
|
|
Amph
Legendary
Offline
Activity: 3248
Merit: 1070
|
|
November 14, 2016, 07:37:58 AM Last edit: November 14, 2016, 08:42:49 AM by Amph |
|
only 30sols on a 1060, does not seem right. what are others getting?
it should be over 60 sol with the last build(reported early in this thread, so maybe now it's even more) but i don't know the setting because i don't own any zawawa you found a way to port the version on windows without all the other things that need to be installed?
|
|
|
|
Stealth8368
|
|
November 14, 2016, 09:03:25 AM |
|
does v5 have a windows version? I thought it was supposed to be released? if I missed it sorry, thanks! looks like great work!
Yes there is a port for windows,if you read developments here it's all there with instructions
|
|
|
|
banana771
Newbie
Offline
Activity: 6
Merit: 0
|
|
November 14, 2016, 09:24:33 AM |
|
Can anyone tell mee if I can use this miner for Zcash CLassic? ZCL if I'm not mistaken. I did try to run this and it seems it's mining, but address is still from ZEC, that I created on jaxx program.
Is there any way to get paid on ZCL address? How do I create ZCL adress if I havee only windows?
|
|
|
|
mgmk
Newbie
Offline
Activity: 3
Merit: 0
|
|
November 14, 2016, 09:26:42 AM |
|
Is there any way to get paid on ZCL address? How do I create ZCL adress if I havee only windows?
You can create ZCL address on the exchange account, for example c-cex
|
|
|
|
banana771
Newbie
Offline
Activity: 6
Merit: 0
|
|
November 14, 2016, 09:37:37 AM |
|
Is there any way to get paid on ZCL address? How do I create ZCL adress if I havee only windows?
You can create ZCL address on the exchange account, for example c-cex Okay, I changed my start.bat to mine on pool.mn with my pool.mn account credentials, and set in profile for payment address that one that I got on c-cex website after registering. I hope I'm doing that right. Kinda new to mining and all.
|
|
|
|
gross
|
|
November 14, 2016, 09:44:37 AM |
|
How to compiler silentarmy in cygwin ? I setup cygwin but don't know compiler it.
|
|
|
|
restless
Legendary
Offline
Activity: 1151
Merit: 1001
|
|
November 14, 2016, 10:31:23 AM |
|
does v5 have a windows version? I thought it was supposed to be released? if I missed it sorry, thanks! looks like great work!
Yes there is a port for windows,if you read developments here it's all there with instructions Where is that?! :/
|
|
|
|
boestin
Legendary
Offline
Activity: 1092
Merit: 1000
|
|
November 14, 2016, 10:57:37 AM |
|
I just copied the newer files over the old one's without compiling it. Basically no speed advantage there, so I think we really need a new compile for Windows.
|
|
|
|
mgmk
Newbie
Offline
Activity: 3
Merit: 0
|
|
November 14, 2016, 11:30:34 AM |
|
Build instructions for Windows:1. Install AMD APP SDK for Windows from here http://developer.amd.com/tools-and-sdks/opencl-zone/amd-accelerated-parallel-processing-app-sdk/2. Get cygwin from http://cygwin.com3. Install cygwin with the packages: - Devel/git - Devel/make - Devel/gcc-core - Python/python3 4. Start cygwin64 terminal 5. Clone repo: git clone https://github.com/mbevand/silentarmy.git6. cd silentarmy 7. Build: make OPENCL_HEADERS="/cygdrive/c/Program\ Files\ \(x86\)/AMD\ APP\ SDK/3.0/include/" LIBOPENCL="/cygdrive/c/Program\ Files\ \(x86\)/AMD\ APP\ SDK/3.0/lib/x86_64" 8. Copy cygwin1.dll: cp /cygdrive/c/cygwin64/bin/cygwin1.dll . 9. Get your build path by typing: cygpath . -w -a 10. Create start.bat with: c:\cygwin64\bin\python3.4m silentarmy --instances=1 -c stratum+tcp://address -u x.x -p x --use 1,2 pause
|
|
|
|
nrg_wolf
Legendary
Offline
Activity: 882
Merit: 1000
|
|
November 14, 2016, 11:31:25 AM |
|
only 30sols on a 1060, does not seem right. what are others getting?
it should be over 60 sol with the last build(reported early in this thread, so maybe now it's even more) but i don't know the setting because i don't own any zawawa you found a way to port the version on windows without all the other things that need to be installed? thats what iam useing, latest drivers with the correct python files, but wont go over 30 sols, got me beat atm. my 970 does 43 sols and my 750ti does 25 sols, so they are working fine. is there a prefered driver version to be useing with it?
|
|
|
|
Amph
Legendary
Offline
Activity: 3248
Merit: 1070
|
|
November 14, 2016, 11:35:12 AM |
|
only 30sols on a 1060, does not seem right. what are others getting?
it should be over 60 sol with the last build(reported early in this thread, so maybe now it's even more) but i don't know the setting because i don't own any zawawa you found a way to port the version on windows without all the other things that need to be installed? thats what iam useing, latest drivers with the correct python files, but wont go over 30 sols, got me beat atm. my 970 does 43 sols and my 750ti does 25 sols, so they are working fine. is there a prefered driver version to be useing with it? i don't know but i'm not using the last driver but the .63, you tried with different setting on msiafterburner like tweaking core and mem?
|
|
|
|
krnlx
|
|
November 14, 2016, 11:37:12 AM |
|
Try replacing ht_store function with this I'm getting little speed increase( 1070) uint ht_store(uint round, __global char *ht, uint i, ulong xi0, ulong xi1, ulong xi2, ulong xi3, __global uint *rowCounters) { uint row; __global char *p; uint cnt; uint tid = get_global_id(0); uint tlid = get_local_id(0); #if NR_ROWS_LOG == 16 if (!(round % 2)) row = (xi0 & 0xffff); else // if we have in hex: "ab cd ef..." (little endian xi0) then this // formula computes the row as 0xdebc. it skips the 'a' nibble as it // is part of the PREFIX. The Xi will be stored starting with "ef..."; // 'e' will be considered padding and 'f' is part of the current PREFIX row = ((xi0 & 0xf00) << 4) | ((xi0 & 0xf00000) >> 12) | ((xi0 & 0xf) << 4) | ((xi0 & 0xf000) >> 12); #elif NR_ROWS_LOG == 18 if (!(round % 2)) row = (xi0 & 0xffff) | ((xi0 & 0xc00000) >> 6); else row = ((xi0 & 0xc0000) >> 2) | ((xi0 & 0xf00) << 4) | ((xi0 & 0xf00000) >> 12) | ((xi0 & 0xf) << 4) | ((xi0 & 0xf000) >> 12); #elif NR_ROWS_LOG == 19 if (!(round % 2)) row = (xi0 & 0xffff) | ((xi0 & 0xe00000) >> 5); else row = ((xi0 & 0xe0000) >> 1) | ((xi0 & 0xf00) << 4) | ((xi0 & 0xf00000) >> 12) | ((xi0 & 0xf) << 4) | ((xi0 & 0xf000) >> 12); #elif NR_ROWS_LOG == 20 if (!(round % 2)) row = (xi0 & 0xffff) | ((xi0 & 0xf00000) >> 4); else row = ((xi0 & 0xf0000) >> 0) | ((xi0 & 0xf00) << 4) | ((xi0 & 0xf00000) >> 12) | ((xi0 & 0xf) << 4) | ((xi0 & 0xf000) >> 12); #else #error "unsupported NR_ROWS_LOG" #endif xi0 = (xi0 >> 16) | (xi1 << (64 - 16)); xi1 = (xi1 >> 16) | (xi2 << (64 - 16)); xi2 = (xi2 >> 16) | (xi3 << (64 - 16)); p = ht + row * NR_SLOTS * SLOT_LEN; uint rowIdx = row/ROWS_PER_UINT; uint rowOffset = BITS_PER_ROW*(row%ROWS_PER_UINT); uint xcnt = atomic_add(rowCounters + rowIdx, 1 << rowOffset); xcnt = (xcnt >> rowOffset) & ROW_MASK; cnt = xcnt; if (cnt >= NR_SLOTS) { // avoid overflows atomic_sub(rowCounters + rowIdx, 1 << rowOffset); return 1; } p += cnt * SLOT_LEN + xi_offset_for_round(round); // store "i" (always 4 bytes before Xi) // *(__global uint *)(p - 4) = i; if (round == 0 || round == 1) { //*(__global uint *)(p - 4) = i; // store 24 bytes ulong2 store; store.x=xi1; store.y=xi2; //*(__global ulong *)(p + 0) = xi0; *(__global uint *)(p - 4) = i; *(__global ulong *)(p + 0) = xi0; *(__global ulong2 *)(p + 8)=store;
} else if (round == 2) { // *(__global uint *)(p - 4) = i; // store 20 bytes
*(__global ulong *)(p - 4) = ((ulong)i) | (xi0 << 32); *(__global ulong *)(p + 4) = (xi0 >> 32) | (xi1 << 32); *(__global ulong *)(p + 12) = (xi1 >> 32) | (xi2 << 32);
} else if (round == 3) { // *(__global uint *)(p - 4) = i; // store 16 bytes //8 byte align *(__global ulong *)(p - 4) = ((ulong)i) | (xi0 << 32); *(__global ulong *)(p + 4) = (xi0 >> 32) | (xi1 << 32); *(__global uint *)(p + 12) = (xi1 >> 32); } else if (round == 4) { // *(__global uint *)(p - 4) = i; // store 16 bytes *(__global uint *)(p - 4) = i; *(__global ulong *)(p + 0) = xi0; *(__global ulong *)(p + 8) = xi1; } else if (round == 5) { //*(__global uint *)(p - 4) = i; // store 12 bytes // *(__global uint *)(p - 4) = i;
*(__global uint *)(p - 4) = i; *(__global ulong *)(p + 0) = xi0; *(__global uint *)(p + 8) = xi1; } else if (round == 6 || round == 7) { // *(__global uint *)(p - 4) = i; // store 8 bytes *(__global ulong *)(p - 4) = ((ulong)i) | (xi0 << 32); *(__global uint *)(p + 4) = (xi0 >> 32); } else if (round == 8) { //4 byte align *(__global uint *)(p - 4) = i; // store 4 bytes *(__global uint *)(p + 0) = xi0;
}
//*(__global uint *)(p - 4) = i; return 0; }
And part of xor_and_store else if (round == 3) { // xor 20 bytes uint one = *(__global uint *)a ^ *(__global uint *)b;
uint4 loada = *(__global uint4 *)((__global char *)a + 4); uint4 loadb = *(__global uint4 *)((__global char *)b + 4); uint4 stor = loada ^ loadb; xi0 = ((ulong)one ) | ((ulong) stor.x << 32); xi1 = ((ulong)stor.y << 32) | ((ulong)stor.z ); xi2 = stor.w;
//xi0 = half_aligned_long(a, 0) ^ half_aligned_long(b, 0); //xi1 = half_aligned_long(a, 8) ^ half_aligned_long(b, 8); //xi2 = well_aligned_int(a, 16) ^ well_aligned_int(b, 16); }
|
|
|
|
snk0752
Newbie
Offline
Activity: 39
Merit: 0
|
|
November 14, 2016, 11:46:45 AM |
|
Try replacing ht_store function with this I'm getting little speed increase( 1070)
krnlx, it works for AMD (R9 NANO) as well. thx! P.S. Curious how to use the HBM memory bandwidth capabilities on R9 NANO/FURY cards in order to adjust the equihash algo.
|
|
|
|
boestin
Legendary
Offline
Activity: 1092
Merit: 1000
|
|
November 14, 2016, 12:13:10 PM |
|
Build instructions for Windows:1. Install AMD APP SDK for Windows from here http://developer.amd.com/tools-and-sdks/opencl-zone/amd-accelerated-parallel-processing-app-sdk/2. Get cygwin from http://cygwin.com3. Install cygwin with the packages: - Devel/git - Devel/make - Devel/gcc-core - Python/python3 4. Start cygwin64 terminal 5. Clone repo: git clone https://github.com/mbevand/silentarmy.git6. cd silentarmy 7. Build: make OPENCL_HEADERS="/cygdrive/c/Program\ Files\ \(x86\)/AMD\ APP\ SDK/3.0/include/" LIBOPENCL="/cygdrive/c/Program\ Files\ \(x86\)/AMD\ APP\ SDK/3.0/lib/x86_64" 8. Copy cygwin1.dll: cp /cygdrive/c/cygwin64/bin/cygwin1.dll . 9. Get your build path by typing: cygpath . -w -a 10. Create start.bat with: c:\cygwin64\bin\python3.4m silentarmy --instances=1 -c stratum+tcp://address -u x.x -p x --use 1,2 pause Thank you Sir!!
|
|
|
|
|