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

[martyroz]

What kind of efficiency are people getting with this miner and 1080ti's?

I'm typically getting 3.6-3.8 sols per Watt with EWBF.

 It's going to depend a LOT on where you set the card and if you are aiming for max efficiency (I can see over 4.1 sol/watt with EWBF on my Aorus or Windforce 1080ti cards at +100 core +150 memory 150 watt TDP for about 620 sol/s) or for hashrate (Aorus can hit 780 sol/s at 250 watts, and over 800 at 275 watts) or for something in between.

Yes, you're right. With EWBF, I typically achieve;
EVGA OC Black - 180W to get 670 sols (3.7 sols/W)
MSI Duke - 180W to get 670 sols (3.7 sols/W)
ASUS ROG STRIX - 180W to get 650 sols (3.6 sols/W)

[1713857089]

1713857089

[1713857089]

1713857089

[1713857089]

1713857089

[Nokia555]

EthControl ZM dstm updated to 0.5.3. (https://ethmonitoring.com)

[TheHas]

Excellent, version 0.5.3 is great. This miner keeps getting better and better!

Now is you could somehow make the difficulty of the coins reduce, I'd really appreciate it

[UrsaUrsa]

New Version 0.5.3
cpu: reduce cpu load
con: fix recon loop
con: network latency measurement
ui:  add accepted/rejected shares ratio
ui:  add network latency
ui:  add information about selected devices

I'm uploading this one earlier than planned since it contains two major improvements - slightly reduced cpu load and a fix for recon - thus it doesn't contain everything you requested/suggested yet.

I've added new stats to the terminal ui. There is more information for share transmission now - it looks now like this:

Code:

Sh: <shares per minute> <ratio accepted/rejected shares> <network latency in ms>

@dstm Thanks mate, I started using this miner 5 minutes ago, I'll leave the task scheduler to restart the app every 12 hours just in case but hoping that the recv problem was fixed as well. Suspiciously no mention of that fix?

[MostWanted]

Exelent work, dstm!

1060 looks fine :-)

How about AMD cards? :-) RX 570 and RX 580? Plan you make AMD miner for Windows?

[z0n0]

Exelent work, dstm!

1060 looks fine :-)

How about AMD cards? :-) RX 570 and RX 580? Plan you make AMD miner for Windows?

What do you get with 1060?

[cervajz]

Hi, first thanks for perfect miner.

I have problem with 0.5.3 (0.5.2 is OK)

On one PC (gaming) 1x1080+1x1070 i using two bat files for each gpu.
I can kill one if play game (second stil running), and start again after that.

But in 0.5.3 there is problem, that i cant run 2 bat files. Second always crashed.
Tried it on another rig and same problem.

If i run only one bat (doesnt matter if 1 or 2 gpu setup) is ok.

Windows 10 x64 fall update, i7-7700k, 16gb ram

[CarlOrff]

New version work fine here (0.5.3) with Ubuntu 16.04.
No any increase in hashrate, but work fine :-).

[Najszerszy]

Exelent work, dstm!

1060 looks fine :-)

How about AMD cards? :-) RX 570 and RX 580? Plan you make AMD miner for Windows?

What do you get with 1060?

With ASUS strix 1060 6GB OC i get 334 Sol with +150 core /+500 memory /65% Power

[dstm]

thanks dstm for this new version 0.5.3 - no recon error so far !
nice that you added --time flag
would it be possible to add also a running time info in the webinterface and maybe even manufacture of each card like other miners like ccminer do ?

Can someone explain (see screenshot) why my EVGA Cards show so different values for sols/w and watts than my Zotac Cards ?
They all running on 55% TDP and eg these 1070s are on nearly same results.
It seems to me that all my EVGAs are on wrong numbers cause even a powermeter displays values like the zotacs.

This was already requested, I'll include

poll informatin - server, port, latency
uptime
last reconnect time
                                                                                                                              
to the webui.

Anything else?

[dstm]

Thanks, I have figured it out now. Trying the latest version on a test rig. The readme file that came with it is comprehensive.
There is no intensity parameter or anything else that we can adjust ourselves for fine tuning? Not that it would be required, the windows interface is not lagging or anything like that.

No there is no possibility to throttle GPU-load currently.

[ap0stol]

Anything else?

for api - can you add card name too?

do you plan to add output to the log file?

[dstm]

Thank you for nice miner.

Is there any monitoring program for multiple rigs which is supporting your miner, like this one https://bitcointalk.org/index.php?topic=1448855.0

Any plans to add log file?

EthControl supports zm (I haven't tested it myself)

logfile: Yes, I'll add it.

[dstm]

Exelent work, dstm!

1060 looks fine :-)

How about AMD cards? :-) RX 570 and RX 580? Plan you make AMD miner for Windows?

No

[dstm]

Hi, first thanks for perfect miner.

I have problem with 0.5.3 (0.5.2 is OK)

On one PC (gaming) 1x1080+1x1070 i using two bat files for each gpu.
I can kill one if play game (second stil running), and start again after that.

But in 0.5.3 there is problem, that i cant run 2 bat files. Second always crashed.
Tried it on another rig and same problem.

If i run only one bat (doesnt matter if 1 or 2 gpu setup) is ok.

Windows 10 x64 fall update, i7-7700k, 16gb ram

Thx for reporting, will check.

[dstm]

Hi guys, i have a problem with your miner. Everybody says that it works better than EWBF´S.
So, my problem is that:

In my rig I can see a performance of 330-340 with my gtx 1060, but when I go to flypool, it only have a 280/285 sols. Also, in my other rig, i can see a performance of 180 and in flypool about 150. (gtx 970m)

If i use EWBF´S my gtx throw me about 320 in my computer and 315 in flypool with is normal, in the other rig about 190 in the computer and 185 in flypool.

So, i dont now what im doing wrong, here is the settings that I´m using:

" zm --server eu1-zcash.flypool.org --port 3333 --user t1L1tRYscLJuz36zpNGk9B4q3YmbtT3iHbj.test970 --pass x --telemetry=192.168.0.158:42002  "

Thanks.
(i write wrong my user for security)
(Posdata, sorry for my bad english)

There is probability involved. You have to run it for a somewhat longer time to get robust numbers, the longer it runs the more accurate it will get. Also check if you have rejected shares.

[Mursu]

New Version 0.5.3
cpu: reduce cpu load
con: fix recon loop
con: network latency measurement
ui:  add accepted/rejected shares ratio
ui:  add network latency
ui:  add information about selected devices

I'm uploading this one earlier than planned since it contains two major improvements - slightly reduced cpu load and a fix for recon - thus it doesn't contain everything you requested/suggested yet.

I've added new stats to the terminal ui. There is more information for share transmission now - it looks now like this:

Code:

Sh: <shares per minute> <ratio accepted/rejected shares> <network latency in ms>

Thanks for the update, I can confirm the reduction on CPU load.

However 0.5.2 crashed my display drivers with message "Display driver nvlddmkm stopped responding and has successfully recovered" (tried both newest ones and reverted back to 385.69) every 2-8 hours and after running 0.5.3 few hours it crashed Windows 10 with last message reading "recv failed" on miner window. I will make a one final try with reduced GPU/mem clocks, but if that fails I need to revert go back to EWBF. System specs is 7 x 1080 TI, i5-7500, 8 Gb RAM.

[dstm]

Anything else?

for api - can you add card name too?

do you plan to add output to the log file?

Are you talking about json data? Should it also include the devicename like on the webui?

[dstm]

New Version 0.5.3
cpu: reduce cpu load
con: fix recon loop
con: network latency measurement
ui:  add accepted/rejected shares ratio
ui:  add network latency
ui:  add information about selected devices

I'm uploading this one earlier than planned since it contains two major improvements - slightly reduced cpu load and a fix for recon - thus it doesn't contain everything you requested/suggested yet.

I've added new stats to the terminal ui. There is more information for share transmission now - it looks now like this:

Code:

Sh: <shares per minute> <ratio accepted/rejected shares> <network latency in ms>

Thanks for the update, I can confirm the reduction on CPU load.

However 0.5.2 crashed my display drivers with message "Display driver nvlddmkm stopped responding and has successfully recovered" (tried both newest ones and reverted back to 385.69) every 2-8 hours and after running 0.5.3 few hours it crashed Windows 10 with last message reading "recv failed" on miner window. I will make a one final try with reduced GPU/mem clocks, but if that fails I need to revert go back to EWBF. System specs is 7 x 1080 TI, i5-7500, 8 Gb RAM.

This is most likely due to overclocking, you have to readjust it, zm puts a different load to your GPUs.

[Andrey09]

v0.5.3  work good 3 hours past.

Need "Uptime time" in telemetry option. Please add in next release.