NiceHash EQM Zcash NVIDIA optimized miner [Maxwell/Pascal] + CPU mining v1.0.4c

[scryptr]

GTX 960 GETS 120Sols/s --

At stock clocks my GTX 960 cards are mining Equihash at 120Sols/s with EQM v1.0.3a.  This is not so impressive, just look at the above post, but it is far better than the 30-40Sols/s they were earning a week or so ago.  For some reason, NiceHash miner auto-tunes EQM better than I was able to with a simple batch file.       --scryptr

[1713473099]

1713473099

[1713473099]

1713473099

[1713473099]

1713473099

[NiceHashSupport]

GTX 960 GETS 120Sols/s --

At stock clocks my GTX 960 cards are mining Equihash at 120Sols/s with EQM v1.0.3a.  This is not so impressive, just look at the above post, but it is far better than the 30-40Sols/s they were earning a week or so ago.  For some reason, NiceHash miner auto-tunes EQM better than I was able to with a simple batch file.       --scryptr

NHM runs 2 threads/instances by default. If you do the same, you will get the same speed as NHM.

i have 6xgtx 1070
my bat file -
eqm -u btcadress -w worker -t 0 -ca -ca -ca -ca -ca -ca
or right
eqm -u btcadress -w worker -t 0 -cd 0 0 0 0 0 0 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4 5 5 5 5 5 5

log:
[22:21:05][0x00000d68][info] device#0 | Initializing thread #35 (CUDA-DJEZO) GeForce GTX 1070 (#5) M=1
[22:21:29][0x00000d68][info] device#35 | Initializing thread #35 (CUDA-DJEZO) GeForce GTX 1070 (#5) M=1

the speed drops in 5 min
2665 ....2150

TPD in AfterBurner 80-85%

As said before, there is no sense to run so many instances... that may rather slow you down over a time due to massive context switching needed. Just do -ca -ca and nothing more (no -cd) and you will be fine.

[krnlx]

@NiceHashSupport

Can you check an issue on linux version, that I reported?

https://bitcointalk.org/index.php?topic=1677369.msg17163199#msg17163199

[NiceHashSupport]

We did not have an issue. You have overclocked cards? Try different driver? What happens if using multiple eqm instances?

[krnlx]

We did not have an issue. You have overclocked cards? Try different driver? What happens if using multiple eqm instances?

with multiple eqm instances(or wit 2 threads) speed is slower. With overclock, or without.

1 thread :

Code:

$ ./eqm -cd 0 -b

==================== www.nicehash.com ====================
     EQM v1.0.3a equihash CPU&GPU Miner for NiceHash.
  This miner is using proprietary base code by NiceHash.
    CPU solver: by xenoncat
    CUDA solver: by djeZo; heavily modified Tromp solver
==================== www.nicehash.com ====================

[00:19:57][0x00007ffb2c022740][info] Log started
[00:19:58][0x00007ffb2c022740][info] Starting up miner instances...
[00:19:58][0x00007ffb222b4700][info] device#0 | Initializing thread #0 (CUDA-DJEZO) GeForce GTX 1070 (#0) M=1
[00:20:00][0x00007ffb2c022740][info] Benchmarking, please wait about 20 seconds...
[00:20:20][0x00007ffb222b4700][info] device#0 | Thread #0 ended (CUDA-DJEZO)
[00:20:20][0x00007ffb2c022740][info] Benchmarking finished
[00:20:20][0x00007ffb2c022740][info] Total time : 20 s
[00:20:20][0x00007ffb2c022740][info] device#0: 178 I/s, 326.016 Sols/s
[00:20:20][0x00007ffb2c022740][info] Total measured: 178 I/s
[00:20:20][0x00007ffb2c022740][info] Total measured: 326.016 Sols/s
[00:20:20][0x00007ffb2c022740][info] Theoretical: 334.471 Sols/s

2 threads:

Code:

$ ./eqm -cd 0 0 -b

==================== www.nicehash.com ====================
     EQM v1.0.3a equihash CPU&GPU Miner for NiceHash.
  This miner is using proprietary base code by NiceHash.
    CPU solver: by xenoncat
    CUDA solver: by djeZo; heavily modified Tromp solver
==================== www.nicehash.com ====================

[00:20:28][0x00007f958269d740][info] Log started
[00:20:28][0x00007f958269d740][info] Starting up miner instances...
[00:20:28][0x00007f9577c9a700][info] device#1 | Initializing thread #1 (CUDA-DJEZO) GeForce GTX 1070 (#0) M=1
[00:20:28][0x00007f957849b700][info] device#0 | Initializing thread #0 (CUDA-DJEZO) GeForce GTX 1070 (#0) M=1
[00:20:30][0x00007f958269d740][info] Benchmarking, please wait about 20 seconds...
[00:20:50][0x00007f9577c9a700][info] device#1 | Thread #1 ended (CUDA-DJEZO)
[00:20:50][0x00007f957849b700][info] device#0 | Thread #0 ended (CUDA-DJEZO)
[00:20:50][0x00007f958269d740][info] Benchmarking finished
[00:20:50][0x00007f958269d740][info] Total time : 20 s
[00:20:50][0x00007f958269d740][info] device#0: 76.8692 I/s, 146.437 Sols/s
[00:20:50][0x00007f958269d740][info] device#1: 76.8615 I/s, 144.629 Sols/s
[00:20:50][0x00007f958269d740][info] Total measured: 153.731 I/s
[00:20:50][0x00007f958269d740][info] Total measured: 291.066 Sols/s
[00:20:50][0x00007f958269d740][info] Theoretical: 288.868 Sols/s

[NiceHashSupport]

Do you get ~330 on overclocked or stock 1070 ?

[krnlx]

Do you get ~330 on overclocked or stock 1070 ?

a bit overclocked.

Anyway, I ran the windows version under wine on same hardware

Code:

$ /opt/wine-staging/bin/wine eqm.exe -cd 0 -b
fixme:winediag:start_process Wine Staging 1.9.23 is a testing version containing experimental patches.
fixme:winediag:start_process Please mention your exact version when filing bug reports on winehq.org.
fixme:heap:RtlSetHeapInformation 0x3f0000 0 0x23fd20 4 stub

==================== www.nicehash.com ====================
     EQM v1.0.3a equihash CPU&GPU Miner for NiceHash.
  This miner is using proprietary base code by NiceHash.
    CPU solver: by xenoncat
    CUDA solver: by djeZo; heavily modified Tromp solver
==================== www.nicehash.com ====================

[00:39:49][0x00000026][info] Log started
[00:39:49][0x00000026][info] Starting up miner instances...
fixme:module:load_library unsupported flag(s) used (flags: 0x00000800)
fixme:ntdll:EtwRegisterTraceGuidsW (0x31ec34, (nil), {f7b697a3-4db5-4d3b-be71-c4d284e6592f}, 7, 0x3debb0, (null), (null), 0x3e2280): stub
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {72b14a7d-704c-423e-92f8-7e6d64bcb92a}
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {e2091f8a-1e0a-4731-84a2-0dd57c8a5261}
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {e8a3bf1f-a86b-4390-9c60-5390b969d22c}
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {5727a00f-50be-4519-8256-f7699871fecb}
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {7e854ec7-cdc4-405a-b5b2-aaf7c9e7d40c}
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {79a60dc6-5fc8-4952-a41c-1163aeec5eb8}
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {2718d25b-5bf5-4479-8e88-babc64bdbfca}
fixme:process:GetNumaHighestNodeNumber (0x23e7a0): semi-stub
fixme:ver:GetCurrentPackageId (0xfbe550 (nil)): stub
[00:39:49][0x00000033][info] device#0 | Initializing thread #0 (CUDA-DJEZO) GeForce GTX 1070 (#0) M=1
[00:39:51][0x00000026][info] Benchmarking, please wait about 20 seconds...
[00:40:11][0x00000033][info] device#0 | Thread #0 ended (CUDA-DJEZO)
[00:40:11][0x00000026][info] Benchmarking finished
[00:40:11][0x00000026][info] Total time : 20 s
[00:40:11][0x00000026][info] device#0: 186.828 I/s, 342.001 Sols/s
[00:40:11][0x00000026][info] Total measured: 186.828 I/s
[00:40:11][0x00000026][info] Total measured: 342.001 Sols/s
[00:40:11][0x00000026][info] Theoretical: 351.059 Sols/s

350 vs 330 , I reported same issue on 1.0.2 version, and you fixed it (1.0.2d, 1.0.2e, 1.0.2f) in last release

and memory controller usage 76% vs 73%

https://bitcointalk.org/index.php?topic=1677369.msg17068170#msg17068170

https://bitcointalk.org/index.php?topic=1677369.msg17064206#msg17064206

[wacko]

Do you get ~330 on overclocked or stock 1070 ?

Since most miners use non-reference 1070s (Founder's edition is noisy), there's really no such thing as "stock 1070". Cards from different brands all come with quite different "stock" clocks; what's considered "stock" by someone will be a noticeable overclock to someone else.

[NiceHashSupport]

Do you get ~330 on overclocked or stock 1070 ?

a bit overclocked.

Anyway, I ran the windows version under wine on same hardware

Code:

$ /opt/wine-staging/bin/wine eqm.exe -cd 0 -b
fixme:winediag:start_process Wine Staging 1.9.23 is a testing version containing experimental patches.
fixme:winediag:start_process Please mention your exact version when filing bug reports on winehq.org.
fixme:heap:RtlSetHeapInformation 0x3f0000 0 0x23fd20 4 stub

==================== www.nicehash.com ====================
     EQM v1.0.3a equihash CPU&GPU Miner for NiceHash.
  This miner is using proprietary base code by NiceHash.
    CPU solver: by xenoncat
    CUDA solver: by djeZo; heavily modified Tromp solver
==================== www.nicehash.com ====================

[00:39:49][0x00000026][info] Log started
[00:39:49][0x00000026][info] Starting up miner instances...
fixme:module:load_library unsupported flag(s) used (flags: 0x00000800)
fixme:ntdll:EtwRegisterTraceGuidsW (0x31ec34, (nil), {f7b697a3-4db5-4d3b-be71-c4d284e6592f}, 7, 0x3debb0, (null), (null), 0x3e2280): stub
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {72b14a7d-704c-423e-92f8-7e6d64bcb92a}
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {e2091f8a-1e0a-4731-84a2-0dd57c8a5261}
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {e8a3bf1f-a86b-4390-9c60-5390b969d22c}
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {5727a00f-50be-4519-8256-f7699871fecb}
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {7e854ec7-cdc4-405a-b5b2-aaf7c9e7d40c}
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {79a60dc6-5fc8-4952-a41c-1163aeec5eb8}
fixme:ntdll:EtwRegisterTraceGuidsW   register trace class {2718d25b-5bf5-4479-8e88-babc64bdbfca}
fixme:process:GetNumaHighestNodeNumber (0x23e7a0): semi-stub
fixme:ver:GetCurrentPackageId (0xfbe550 (nil)): stub
[00:39:49][0x00000033][info] device#0 | Initializing thread #0 (CUDA-DJEZO) GeForce GTX 1070 (#0) M=1
[00:39:51][0x00000026][info] Benchmarking, please wait about 20 seconds...
[00:40:11][0x00000033][info] device#0 | Thread #0 ended (CUDA-DJEZO)
[00:40:11][0x00000026][info] Benchmarking finished
[00:40:11][0x00000026][info] Total time : 20 s
[00:40:11][0x00000026][info] device#0: 186.828 I/s, 342.001 Sols/s
[00:40:11][0x00000026][info] Total measured: 186.828 I/s
[00:40:11][0x00000026][info] Total measured: 342.001 Sols/s
[00:40:11][0x00000026][info] Theoretical: 351.059 Sols/s

350 vs 330 , I reported same issue on 1.0.2 version, and you fixed it (1.0.2d, 1.0.2e, 1.0.2f) in last release

and memory controller usage 76% vs 73%

https://bitcointalk.org/index.php?topic=1677369.msg17068170#msg17068170

https://bitcointalk.org/index.php?topic=1677369.msg17064206#msg17064206

We will take a look into this...

[Kompik]

Did you change the payment threshold from 0.01 to 0.1 BTC? I really dont to wait a week for a payout.

[TR8888]

Which drivers should I use for this miner for a 1070? I'm currently using 372.54 and it seems a little slow with 290 sols on stock 1070, W10

[peonv]

346 per 1070 with overclocked memory.

[NiceHashSupport]

Which drivers should I use for this miner for a 1070? I'm currently using 372.54 and it seems a little slow with 290 sols on stock 1070, W10

Use -cd 0 0 or -ca -ca to get better speed.

[TR8888]

Which drivers should I use for this miner for a 1070? I'm currently using 372.54 and it seems a little slow with 290 sols on stock 1070, W10

Use -cd 0 0 or -ca -ca to get better speed.

I tried both, -cd 0 0 is the best achieving 299 sols.

Which drivers are recommended? The most recent from nVidia website?

[_javi_]

i cant pass 300 sols either..

Gigabyte g1 1070 (6 pin, does it matter??), win 7 64, 4 gb... latest drivers.

[dakky]

with single 980Ti and (single) -ca  @ 340 sols/s

but Off Topic... why next payout time at @ 0.03 ?? something changed, not anymore at 0.01?

[ioglnx]

Yeah good question was the payout frame changed again?
It was usually always each night when reaching 0.01

[reb0rn21]

I just got paid for two days, just a bit over 0.1, might be new limit,.. BTC blockchain is so overloaded I see sometimes even with fee BTC do not pass for hours

[ioglnx]

China wake up

[TR8888]

When wantring to CPU mine only, It asks for cuda.dll , even the newes ver of nh miner cpu mining only for equi asks for cuda dll's..