[ANN] dstm's ZCash / Equihash Nvidia Miner v0.6.2 (Linux / Windows)

[NameTaken]

I am trying to test zm on debian
BUT:

$./zm
./zm: /usr/lib/x86_64-linux-gnu/libcrypto.so.1.0.0: version `OPENSSL_1.0.0' not found (required by ./zm)
./zm: /usr/lib/x86_64-linux-gnu/libssl.so.1.0.0: version `OPENSSL_1.0.0' not found (required by ./zm)

what I should to fix?

BTW
I have already made links
#ln -n libcrypto.so.1.1 libcrypto.so.1.0.0
#ln -n libssl.so.1.1 libssl.so.1.0.0

https://bitcointalk.org/index.php?topic=2021765.msg22078213#msg22078213

[grchina]

Can someone post results for 1080ti with dual core cpu?Would be nice to know how cpu heavy it is

yes, sure, I posted before, the rig with 4x1080ti and G4560 is working on windows 10
it shows like 730-740 sol/s per card,  a little low comparing to ewbf, but I guess that's cause of the cpu, in fact it should push around 770-790  if pared with properly overclocked and good 4core cpu
and all is working stable, just speed is a little low

thx man was hoping i could use it on my g3260 rigs but guess i am going to stay on cc miner algos

[papampi]

I think in 0.5.1 The network time out is too high and only gives slow network.
I removed the network cable and the miner was running for more than 5 minutes without network and was just giving Slow network warning.

Is thhis fixed in 0.5.2 ?

[papampi]

It's still a bit CPU heavy since I use my CPUs to mine VRM or XMG at the same time, so DSTM might consider fixing that, or minimizing the CPU usage atleast.

I noticed high cpu too
(7x1070 cards)

Code:

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND                              
28958 m1        20   0 90.606g 728128 607436 S  48.7  8.9  69:40.31 zm_miner                             
10549 root      20   0       0      0      0 S   8.3  0.0   0:58.25 kworker/3:3                          
 1488 root     -51   0       0      0      0 S   6.7  0.0  56:58.30 irq/132-nvidia                       
 1641 root     -51   0       0      0      0 S   6.3  0.0  52:35.57 irq/135-nvidia                       
  989 root     -51   0       0      0      0 S   6.0  0.0  47:25.49 irq/131-nvidia   

$ lscpu | grep name

Code:

Model name:            Intel(R) Core(TM) i3-6100 CPU @ 3.70GHz

[raven1322]

It's still a bit CPU heavy since I use my CPUs to mine VRM or XMG at the same time, so DSTM might consider fixing that, or minimizing the CPU usage atleast.

Hello fellow VRM miner! The cpu usage is very apparent, h/s went down I hope this gets fixed in the next release..

[induktor]

induktor,

card: 1 x1080 Ti
...
Hash rate at top efficiency: 600 Sols average  (15 Sols more than EWBF miner)
...

Is it a joke or a mistake? 1080Ti should give u from 750 to 790 sol (in OC mode) and from 695 to 750 (in non-OC mode) on EWFB. What 600 Sol/s are u talking about, man?

Top eficiency means, maximum hashrate at the lowest possible power usage (hash/Watts)
unless you don't pay for electricity, i pay a fortune for power, so i keep my miners at top efficiency all the time to get maximum profit.

At least so far is what i got, still have to test other drivers/miners, but everything so far points to 600 Sols at 140 watts (155 at the wall)

I agree with efficiency being a huge factor specially when power is expensive. That being said, how’d you manage a 1080Ti to pull only 140W? Can’t go below 200W on my strix oc’s..

Driver 384 will do it, 381 and earlier has some weird issues with the power target limit.

[dstm]

I think in 0.5.1 The network time out is too high and only gives slow network.
I removed the network cable and the miner was running for more than 5 minutes without network and was just giving Slow network warning.

Is thhis fixed in 0.5.2 ?

This is how TCP works. If you remove the cable it will take some time for the connection to timeout. Zm isn't able to detect this until the OS signals a socket timeout.

[induktor]

Hi
I have been testing this miner for 8 hours and then go back to EWBF miner, and test another 8 hours.

in my particular case, the results are like this:  (1x1080Ti STRIX)

ZM: 604 Sol     Reported at the pool as average: 581 Sol
EWBF: 585 Sol  Reported at the pool as average: 580 Sol

power usage (the same 155W)

even when ZM reports more hash rate, the average is the same for both miners.

I don't see any instabilities with ZM like others reported, it is perfectly stable
I don't see difference in hashrate at the pool, only at the miner side.

again this is at this lower power levels, so at a higher power the situation could be different.
would be interesting someone else to test and report, i will live both miners installed and switch back and forth to test before decide if it worth it to move the new miner to the farms.

I also tested the TPRUVOT's CCMINER 2.2.2 and I probably doing something wrong because I only getting 322 Sols, and that's impossible with a 1080Ti hehe.

my specs:
OS: Kopiemtu 2.0 Mod #5 (lubuntu 14.04) driver 384.90
target power limit at 140, overclock +100 core -1200 mem
cpu is AM1 5350 APU SOC w/8gb ram (25% load with 1 card)

indkt.

[dstm]

Hi
I have been testing this miner for 8 hours and then go back to EWBF miner, and test another 8 hours.

in my particular case, the results are like this:  (1x1080Ti STRIX)

ZM: 604 Sol     Reported at the pool as average: 581 Sol
EWBF: 585 Sol  Reported at the pool as average: 580 Sol

power usage (the same 155W)

even when ZM reports more hash rate, the average is the same for both miners.

I don't see any instabilities with ZM like others reported, it is perfectly stable
I don't see difference in hashrate at the pool, only at the miner side.

again this is at this lower power levels, so at a higher power the situation could be different.
would be interesting someone else to test and report, i will live both miners installed and switch back and forth to test before decide if it worth it to move the new miner to the farms.

I also tested the TPRUVOT's CCMINER 2.2.2 and I probably doing something wrong because I only getting 322 Sols, and that's impossible with a 1080Ti hehe.

my specs:
OS: Kopiemtu 2.0 Mod #5 (lubuntu 14.04) driver 384.90
target power limit at 140, overclock +100 core -1200 mem
cpu is AM1 5350 APU SOC w/8gb ram (25% load with 1 card)

indkt.

This thread contains pool side comparisons
https://forum.z.cash/t/dstm-s-zcash-cuda-miner

There are also several on this thread.

[thesavoyard]

So this miner has no .bat files and every time I try to start the exe it closes instantly. What can I do?


nevermind, mad a .bat file to start it.

[dstm]

Segmentation fault  gpu_id 1 3 0x10830010600 unspecified launch failure
gpu 1 unresponsive - check overclocking
cudaMemcpy 1 failed
./zm-zec.sh: line 1: 32357 Segmentation fault
((((

ps: in ewbf all ok

Try reduce your OC. I have been running 2 days now fully stable with my current OC settings used in EWBF.

dstm, as already mentioned by many you need to look into the cpu usage of your miner. That is my only problem with it right now... otherwise working great.

--ypsi

I'm working on it.

[yobigd20]

I can't get this miner to work on Win10 at all.  I've tried many different settings.  No luck.


zm_win_0.5.2\zm --server zen.suprnova.cc --port 3618 --user myuser.mywoker --pass x
(obviously i replace myuser and myworker with my own stuff)

and i get this:
#  GPU0 connected to: zen.suprnova.cc:3618
connect failed: 10061

same thing for zcash.

zm_win_0.5.2\zm>zm --server us1-zcash.flypool.org --port 3333 --user mywallet.myrigname --pass x
#  GPU0 connected to: us1-zcash.flypool.org:3333
connect failed: 10061


i just keep getting these 10061 errors.

Do you block some ports?

Do you mean block outbound connections or block opening local sockets? My outbound destinations are restricted but you can open as many local sockets as you want. Ewbf has no issue connecting to those same exact proto:ip:port as I have them whitelisted.    Is ZM trying to connect to some other unrelated site? (If so this is quite alarming).

[dstm]

I can't get this miner to work on Win10 at all.  I've tried many different settings.  No luck.


zm_win_0.5.2\zm --server zen.suprnova.cc --port 3618 --user myuser.mywoker --pass x
(obviously i replace myuser and myworker with my own stuff)

and i get this:
#  GPU0 connected to: zen.suprnova.cc:3618
connect failed: 10061

same thing for zcash.

zm_win_0.5.2\zm>zm --server us1-zcash.flypool.org --port 3333 --user mywallet.myrigname --pass x
#  GPU0 connected to: us1-zcash.flypool.org:3333
connect failed: 10061


i just keep getting these 10061 errors.

Do you block some ports?

Do you mean block outbound connections or block opening local sockets? My outbound destinations are restricted but you can open as many local sockets as you want. Ewbf has no issue connecting to those same exact proto:ip:port as I have them whitelisted.    Is ZM trying to connect to some other unrelated site? (If so this is quite alarming).

ZM uses tcp port 3443 to submit dev fee shares.

[yobigd20]

I can't get this miner to work on Win10 at all.  I've tried many different settings.  No luck.


zm_win_0.5.2\zm --server zen.suprnova.cc --port 3618 --user myuser.mywoker --pass x
(obviously i replace myuser and myworker with my own stuff)

and i get this:
#  GPU0 connected to: zen.suprnova.cc:3618
connect failed: 10061

same thing for zcash.

zm_win_0.5.2\zm>zm --server us1-zcash.flypool.org --port 3333 --user mywallet.myrigname --pass x
#  GPU0 connected to: us1-zcash.flypool.org:3333
connect failed: 10061


i just keep getting these 10061 errors.

Do you block some ports?

Do you mean block outbound connections or block opening local sockets? My outbound destinations are restricted but you can open as many local sockets as you want. Ewbf has no issue connecting to those same exact proto:ip:port as I have them whitelisted.    Is ZM trying to connect to some other unrelated site? (If so this is quite alarming).

ZM uses tcp port 3443 to submit dev fee shares.

it's not ports i block, its address ranges.  and zm.exe is trying to connect to 94.23.60.113.   what server is this?  reverse ip lookups is not returning any known domains.  if it is a known pool I can whitelist it.  but if its some server running in your basement I'd be a complete moron to allow this.

[dstm]

I can't get this miner to work on Win10 at all.  I've tried many different settings.  No luck.


zm_win_0.5.2\zm --server zen.suprnova.cc --port 3618 --user myuser.mywoker --pass x
(obviously i replace myuser and myworker with my own stuff)

and i get this:
#  GPU0 connected to: zen.suprnova.cc:3618
connect failed: 10061

same thing for zcash.

zm_win_0.5.2\zm>zm --server us1-zcash.flypool.org --port 3333 --user mywallet.myrigname --pass x
#  GPU0 connected to: us1-zcash.flypool.org:3333
connect failed: 10061


i just keep getting these 10061 errors.

Do you block some ports?

Do you mean block outbound connections or block opening local sockets? My outbound destinations are restricted but you can open as many local sockets as you want. Ewbf has no issue connecting to those same exact proto:ip:port as I have them whitelisted.    Is ZM trying to connect to some other unrelated site? (If so this is quite alarming).

ZM uses tcp port 3443 to submit dev fee shares.

it's not ports i block, its address ranges.  and zm.exe is trying to connect to 94.23.60.113.   what server is this?  reverse ip lookups is not returning any known domains.  if it is a known pool I can whitelist it.  but if its some server running in your basement I'd be a complete moron to allow this.

Yes, that's the right thing to do - you should check the connections a closed source program is making.

Code:

$ host eu1-zcash.flypool.org
eu1-zcash.flypool.org has address 94.23.20.210
eu1-zcash.flypool.org has address 94.23.12.63
eu1-zcash.flypool.org has address 94.23.60.113
eu1-zcash.flypool.org has address 188.165.195.21
eu1-zcash.flypool.org has address 94.23.199.191

So, nothing special in zm

[yobigd20]

Yes, that's the right thing to do - you should check the connections a closed source program is making.

Code:

$ host eu1-zcash.flypool.org
eu1-zcash.flypool.org has address 94.23.20.210
eu1-zcash.flypool.org has address 94.23.12.63
eu1-zcash.flypool.org has address 94.23.60.113
eu1-zcash.flypool.org has address 188.165.195.21
eu1-zcash.flypool.org has address 94.23.199.191

So, nothing special in zm

thanks, that was the issue.  i've whitelisted eu1-zcash.flypool.org and it works now.    

[UrsaUrsa]

@dstm: First of all thanks for your hard work.

Secondly, your miner started erroring out after a few hours please see below. Any ideas?

Thx for reporting, this is a bug - will be fixed in next release.

Damn, it's getting worse and I'm not sure why. Its crashing like every 2 hours... Once it crashed the whole system starts lagging like there is no tomorrow. Does this happen with the linux version as well?

[dstm]

@dstm: First of all thanks for your hard work.

Secondly, your miner started erroring out after a few hours please see below. Any ideas?

Thx for reporting, this is a bug - will be fixed in next release.

Damn, it's getting worse and I'm not sure why. Its crashing like every 2 hours... Once it crashed the whole system starts lagging like there is no tomorrow. Does this happen with the linux version as well?

Is it crashing or do you get an recv error?

[UrsaUrsa]

@dstm: First of all thanks for your hard work.

Secondly, your miner started erroring out after a few hours please see below. Any ideas?

Thx for reporting, this is a bug - will be fixed in next release.

Damn, it's getting worse and I'm not sure why. Its crashing like every 2 hours... Once it crashed the whole system starts lagging like there is no tomorrow. Does this happen with the linux version as well?

Is it crashing or do you get an recv error?

I get the recv error, zm does not exit just keeps running like there is no tomorrow. I assume after a while there is memory overflow or something and windows decides to die. It will pose a problem when I'm not home.

[Brzina]

Maybe, this could help someone.  This service took more than 50% of my CPU (microsoft software protection platform service)  and this was the cause of ZM crashing I think.

Now everything is stable and I get more SOLs after disabling it like in this manual >>> http://www.tomshardware.co.uk/forum/id-3259400/microsoft-software-protection-platform-service-high-cpu.html

"Just found out how to resolve this very annoying issue:
Open regedit,
go to HKLM\SYSTEM\CurrentControlSet\Services\sppsvc
Modify "Start" from "whatever number is here" to 4
This will disable the service after you restart.
No more CPU wastage"

Is this service something important, will I have problems after disabling it ? Thanks