Topic: [ANN] dstm's ZCash / Equihash Nvidia Miner v0.6.2 (Linux / Windows) - page 45. (Read 224961 times)

Hey guys i was switching back over to dstm to compare it vs bminer and i realized something odd.

For some reason it's slowing down my radeon cards mining and even my desktop?. claymore miner slows down from 49 down to 44~ and my desktop even feels laggy while using dstm miner.

When i was using Bminer this wouldn't happen and was wondering if anyone knew what may be the cause, my cpu is nowhere near overloaded so im genuinely curious.

Im using v 5.8 and im just preparing since there's a lot of mixes reviews about bminer and im not so sure about it.

Thanks

Paying for shit that's isnt yours is a drag right??

Hello guys!
Want to remind to all of you, that if somebody has any problems with stability of EWBF, DSTM, BMINER, CLAYMORE, CC miners or needs some extra control over your rigs please read carefully my message below.

I want to introduce you a script, which can solve most of your problems. It is a result of half a year hard work with my beta testers. Now it is clean of any bugs and ready to be introduced for the wide mining community.

1) https://bitcointalksearch.org/topic/cmd-farmwatchbot-ewbf-claymore-bminer-dstm-cc-eth-castxmr-phoenix-2071108
2) https://forum.z.cash/t/cmd-equihash-miner-autorun-autorestart-watchdog-for-ewbf-claymore-dstm/20640

It will definetely reduce your valuable time spent near your rigs, because this script will do all the work for you. Just start it and forget about your rigs, relax or do anything else. In case of any problem this script will restart your miner, or the rig, if necessary. It will notify you by Telegramm message about all problems and corrective actions taken. It is open source code (CMD/BAT, not an .exe or something like this), so anyone can check throug the code if needed.
Enjoy your life, let the script to monitor your rigs for you.

Make sure you have OC enabled in 2D mode. Miner does not trigger rendering anything. I am on MSI Afterburner and it works.


look for the connection it tries to open, perhaps it will help to know what is the miner connecting to. (My suspicion is that miner is trying to connect to something that is no longer responding)
In windows you can use install PSTools from sysinternals and use "Process Explorer" for convenience.
In linux netstat should be enough.


Correct me if I am wrong, but increasing lag between CPU and GPU would only increase the demand on memory on the GPU. The impact on average amount of solutions should not be impacted by it.
On a unrelated note - is it normal that the  miner takes so little GPU memory? (<500MB for me)
This based on luck. There is a guarantee you will find a share, but the amount of hashes it takes changes. Pools will report higher hash rate due to this. It is still ok to use pool hashrate.
Difficulty is used to balance workload. Higher difficulty will result in miner working in longer "cycles" which is beneficial to faster hardware.
Average solutions per second are the measurement of how your miner performs. Note that the reliance is always on the miner. That is why I am using pools as well.


Miner fee and Pool fee are unrelated as for the amounts it is 2% of my hashing power and 0.9% from the pool for my payment.

What I mean is that pools expect difficulty configuration as a number, so even I know where my performance sweet spot is - I cannot just paste it as a hint to the pool.
I am not sure if that is possible, but it would be appreciated.


The idea of logging change is to make it easier to spot problems. The log right now has the information you need, it is just clumped together.
For example I switched to mining pool hub and performance dropped 5-10% - I used python to split the information and found out they are seemingly randomly changing difficulty back to 0001....
I checked the shares and found out the coin switching is not nicely implemented by the pool.
Other card had performance problems due to fan not scaling (For some reason the fan speed was not scaled on temperature in 2D mode).
 - Here might be a good innovation. If you can read PerfCap Reason and accordingly to it would be better rather than temperature.
 - I suspect targeting a temperature will be a good workaround
Now I am trying to find sweet spots for Sol/W, because going as fast as possible is not always the best efficiency.


Since you took the time to respond, I took the time to replicate my original problem.
Unfortunately I was not able to. I am now running at 0.5.8

hi DSTM

please help me.

with reference to two rigs of mine that do over 3000 sols each.  

almost every morning when i wake up i notice no hashrate on pool monitor for these two rigs.  i then check miner and see  -  
connection closed by server   r 0
reconnecting


what do i do ?
i have done much troubleshooting with no luck.


regards



note:  i have other rigs under 3000sols on the same network that dont do this

I have it crashing within a day or so using the latest version. zm.exe has stopped working. Anyone figure a fix for this?

Awesome this is one of the main troubles I face, and it is usually fixed with the old turn it off and on again. I'll test it out and see how it works

I second that.... It's a great script

use this script ( completly open source and free) by Acrefawn. It will take care of your rigs automatically, restart at set intervals or if errors happen. It's pretty great.
If you use it and it's helpful remember to support the dev after you tested it thoroughly.

https://bitcointalksearch.org/topic/cmd-farmwatchbot-ewbf-claymore-bminer-dstm-cc-eth-castxmr-phoenix-2071108

So in Windows 10 this will keep it running?.   Perhaps dstm can add that to the compiled .exe so it won't fail overnight?  (I'll add it and check to see if it works)  thanks

It does not start, it does not work completely, or it will not apply your boost to Core clocks?

If it does not apply only core clock, try to run Google Chrome along with DSTM miner and check if it works now.

.....
You don't realize that article also applies to Windows 10, do you?

That does not make sense!
Mining ZEC requires calculating a lot of Blake2b hashes for the Equihash algorithm.

Normal windows software doesn't work on windows server, especially 2008.

No drivers or support for any pro hardware.

Open Solaris is a better choice.  It's very stable and supports anything out of the box.

I have a workstation for video editing w 2 Xeons and 4 GPUs running on opensolaris.

BeOS is decent too.  The fastest I've ever used.  Still insanely fast 22 years later on my raspberry pi.

Window pops up a box?

I have never seen error "zm is stop working".

When my rigs crash they eject the GPUs or "connection to GPU lost.  Few cases where the whole thing shuts down.

You can't auto restart.  I got tired of it and let it rain.

This is to prevent "... stopped working" window pop-up
https://www.raymond.cc/blog/disable-program-has-stopped-working-error-dialog-in-windows-server-2008/

After that the bat file would be

:here
zm ....................

goto :here

How to get the miner auto start after it crash and window pops up a box with " zm is stop working "  ?

How to OC ?
MSI afterburner does not work

SOLVED! Thanks DSTM, tried to get help from ethos chat didn't get a clear answer