[OS] nvOC easy-to-use Linux Nvidia Mining

[fullzero]

I like nanopool and slushpool.

https://zec.nanopool.org/help

https://slushpool.com/help/get-started/getting_started_zcash

Nanopool you don't need an account for and their stats page is useful.

Slush Pool you need an account; but you get a lot of advanced features without any special effort on your part.

I noticed Nanopool charges 1% and Slush Pool 2%. Does using EWBF's CUDA through nvOC add an additional 2% to these pools?

You can choose the EWBF donation percent; it defaults to 0.

[1714295535]

1714295535

[1714295535]

1714295535

[1714295535]

1714295535

[1714295535]

1714295535

[slackert]

Thanks for the reply. Is the SOL/s rate the same on both pools or does this fluctuate? I guess I'm asking if there is an advantage besides stats for one pool over the other.

Also, I had a question about using Nvidia X Server settings window to change GPU core and mem. It seems to work for everything but the fan. Not sure this is the way to adjust settings on the fly. Maybe SSH is a better solution?

[fullzero]

Thanks for the reply. Is the SOL/s rate the same on both pools or does this fluctuate? I guess I'm asking if there is an advantage besides stats for one pool over the other.

Also, I had a question about using Nvidia X Server settings window to change GPU core and mem. It seems to work for everything but the fan. Not sure this is the way to adjust settings on the fly. Maybe SSH is a better solution?

This may seem strange at first but the reported hashrate at a pool is less important than the pools reward type. 

Most of the time a pool gives higher reported hashrates; it is only because of how they are calculating your shares (ultimately their reward type).

There are many ways of choosing a pool.

IMO most miners don't have a logical explanation for why they use the pool they do.  They tried it and were satisfied. 

There is a lot you need to learn before you can make an informed decision. 

To start I would look at this webpage on mining pool reward types:https://en.bitcoin.it/wiki/Comparison_of_mining_pools

I know it is for Bitcoin and most of these pools no longer exist. 

Pool reward types vary significantly; you want to know how you are being paid.  There is no substitute for learning all this yourself.


To change fan speed without restarting the oneBash process:

press f12 to open the guake terminal and enter:

nvidia-settings -a [fan:0]/GPUTargetFanSpeed=75

for each gpu

if I had 3 cards and wanted a fan speed of 75% I would enter:

nvidia-settings -a [fan:0]/GPUTargetFanSpeed=75

then:

nvidia-settings -a [fan:1]/GPUTargetFanSpeed=75

then:

nvidia-settings -a [fan:2]/GPUTargetFanSpeed=75

[fullzero]

I always try to test every card type; so I got 2x 1030 SCs:








I am posting about these to warn members DO NOT BUY ANY 1030s!!.

You are much better off getting 1x 1050ti SC than 2x 1030 SCs; and you will save ~$30

[Maxximus007]

You should run a half percent full percent commision for your nvOC.  Every other man and his dog takes a cut dor their hard work and you provide better customer service support than most very expensive software packages!

That has been suggested before, but as fullzero explained he wants to have it further developed (easy update, remote management) before taking donations. But IMHO it's quite simple to send some appreciation: keep the CPUminer on the default account, and/or mine a couple of hours on his accounts with your rig.

[cvitanihc0011]

Have 2 miners on this and they seem to work fine for a few hours and then crash :

09:29:41:752   bf3e6740   watchdog - fminer thread hangs
09:29:41:752   bf3e6740   em hbt: 0, fm hbt: 130832,
09:29:41:752   bf3e6740   Miner thread hangs, need to restart miner!

Too much OC ?
Also ... the USB seems hot ... doesn't it get damaged in time ?

What GPUs and OC settings are you using; also what is the ambient tempature near the rig?

Also what kind of USB key are you using?

Would it help to use a ssd instead of usb ? For me it lags each time i open something..click something ..
Even if the miner is closed. I see there are links for ssd connection

You can use an ssd if you want; see

https://bitcointalk.org/index.php?topic=1854250.msg19300078#msg19300078

https://bitcointalk.org/index.php?topic=1854250.msg19328385#msg19328385


In my experience a fast USB is reasonably responsive.  The lexar ones I recommend are fast and cheap.

If they aren't fast enough for you: these are faster HQ very high speed USB keys are around $22:

https://www.amazon.com/SanDisk-Extreme-Flash-Drive-SDCZ80-032G-GAM46/dp/B00KT7DXIU

https://www.amazon.com/dp/B00S6DBGLK

GPUs        - GTX 1070 - MSI Gaming X - Micron Memory
Core         - +150
Mem         - +1100
PowerLimit - +125
Risers        - v07s - prob gona change them for ver 06c
USB           - Kingston Adata Traveler - 32gb USB 3.0  = > but I doubt they are because it takes ages to write the image .. like 7mbs
Reported mh on a 7GPU rig - 211mh - no idea about consumption. Cheap electric bill here

I have a bunch of SSDs prob gona use those instead

I see somebody pushed  Core OC at 200 and Mem at 1600 waw .

[citronick]

Added on to the rig a bit today.  Went from 5 GPUs to 7.  All Gigabyte Founders Ed. cards, on an Asus Prime Z270-A, a Celeron 3930, and a Corsair HX1200i.

220 MH/sec ETH (31.42 MH/sec avg per GPU)

Core OC: 200
Mem OC: 1600

power: 120W cap per GPU, pulling 945W at the wall.



I'll begin building a 2nd rig in a couple of days.  First parts rolled in today (4 EVGA 1070s).  Mobo will be another Z270-A, and I have another HX1200i.  The idea will be to have another 7-GPU rig that pulls 950W.  

200/1600/120 PL ?
at 950watts with 7 x GPU = fantastic!

[cvitanihc0011]

What do you mean by "(only 6x OC)" for the z170/z270 boards? Can this OS support 7 cards with those boards?

and you say usb key only....does that mean an M.2 drive will not work? what size ssd is required?

edit: removed lazy question

I can now fully OC all cards; just need to rebuild and update the images. 

Also I just confirmed that if an ssd is imaged while connected to the build pc with a USB-to-sata adapter and then attached to a rig via the usb-to-sata adapter that it will also work fully with all the current nvOC images. 

Here are some usb-to-sata adapters:

https://www.amazon.com/dp/B00OJ3UJ2S

https://www.amazon.com/Sabrent-2-5-Inch-Adapter-Optimized-EC-SSHD/dp/B011M8YACM

https://www.amazon.com/dp/B004A98YTG

https://www.amazon.com/dp/B06WVDK9QD

https://www.amazon.com/Inateck-SATA-Adapter-Cable-Drives/dp/B01E4VHSYG

https://www.amazon.com/StarTech-SATA-Drive-Adapter-Cable/dp/B00HJZJI84


Here are some usb-to-m2 adapters:

https://www.amazon.com/dp/B015C5GDY4

https://www.amazon.com/Portable-External-Enclosure-Controller-Connector/dp/B01G51SATK

https://www.amazon.com/MEITK-Portable-Enclosure-Controller-Connector/dp/B01J3KJ04E

I'll image a SSD using the USB to SATA adapter and it should work

One Q I have guys if you can explain :
- i know there are m2  - > Pci adapters for extra pci slots if the board has a slot to attach this m2 adapter
This USB - > 2m gives me the option to add a m2 - > PIC adapter without a slot on MB ?

[Maxximus007]

I'll image a SSD using the USB to SATA adapter and it should work

One Q I have guys if you can explain :
- i know there are m2  - > Pci adapters for extra pci slots if the board has a slot to attach this m2 adapter
This USB - > 2m gives me the option to add a m2 - > PIC adapter without a slot on MB ?

Probably not, an M.2 slot is essentially a PCI-E slot. USB is not. I think the USB to M.2 emulates a PCI-E 1x (M.2) to be able to address the drive. The system recognises an USB drive.

[Maxximus007]

Just did a small test with the new EWBF 0.3.4B miner and it is indeed 2% faster. Not sure if dev fee is fixed now, that would make the performance gain useless. This version does have a web statuspage to view some info about the miner.

[gustavdp]

Hi,

Thanks for all your effort. Once I get everything up and running I hope there is somewhere I can donate.

I need some advice though. I'm trying to flash the image to a 16gb ADATA S102 flash drive (Capacity shows 15.81GB in HDD Raw Copy), but it fails at 99%. is there something I can do, or should I get a bigger flash?

[Maxximus007]

Hi,

Thanks for all your effort. Once I get everything up and running I hope there is somewhere I can donate.

I need some advice though. I'm trying to flash the image to a 16gb ADATA S102 flash drive (Capacity shows 15.81GB in HDD Raw Copy), but it fails at 99%. is there something I can do, or should I get a bigger flash?

You should get a bigger one. But at times it will still work, so you can test it. If the system boots and starts mining, all is ok. Be patient, booting (depending on number of cards) can take a couple of minutes. Make sure you've read the OP, connect a monitor to GPU0 (not the internal video card) to see the booting process. All is automated, so just wait and see ..

[cvitanihc0011]

Hi,

Thanks for all your effort. Once I get everything up and running I hope there is somewhere I can donate.

I need some advice though. I'm trying to flash the image to a 16gb ADATA S102 flash drive (Capacity shows 15.81GB in HDD Raw Copy), but it fails at 99%. is there something I can do, or should I get a bigger flash?

You should get a bigger one. But at times it will still work, so you can test it. If the system boots and starts mining, all is ok. Be patient, booting (depending on number of cards) can take a couple of minutes. Make sure you've read the OP, connect a monitor to GPU0 (not the internal video card) to see the booting process. All is automated, so just wait and see ..

The image still works if capacity is over 15.8 as shown in tool. Less..it doesn't work. Fullzero explained it a bit better a few pages ago to me

[gustavdp]

Hi,

Thanks for all your effort. Once I get everything up and running I hope there is somewhere I can donate.

I need some advice though. I'm trying to flash the image to a 16gb ADATA S102 flash drive (Capacity shows 15.81GB in HDD Raw Copy), but it fails at 99%. is there something I can do, or should I get a bigger flash?

You should get a bigger one. But at times it will still work, so you can test it. If the system boots and starts mining, all is ok. Be patient, booting (depending on number of cards) can take a couple of minutes. Make sure you've read the OP, connect a monitor to GPU0 (not the internal video card) to see the booting process. All is automated, so just wait and see ..

The image still works if capacity is over 15.8 as shown in tool. Less..it doesn't work. Fullzero explained it a bit better a few pages ago to me

Yea it doesn't complete the transfer and there's 1gb unallocated space. I'll just get a bigger flash.

[citronick]

Hi fullzero

After last night's BTC bloodbath... I want to adjust my mining strategy  

I am positioning my Zotac 1070 minis and some of my entry level 1080ti reference/blower types for ETH mining.... and keep the big guns GA-Aorus and Zotac AMP Extremes for ZEC mining..... how do I modify the onebash for the following?

I copied this cmd string right out off smOS  

-dbg -1  -wd 1 -r 1  -epool  stratum+tcp://eth-sg.dwarfpool.com:8008  -ewal $walletETH/$rigName -esm 0 -epsw x -allpools 1 -mport -3333

Thanks!

ensure you have set your:

ETH_ADDRESS

and

ETH_WORKER

then:

around line 925

Code:

if [ $COIN == "ETH" ]
then
HCD='/home/m1/9.0/ethdcrminer64'
ETHADDR="$ETH_ADDRESS/$ETH_WORKER"

if [ $ETHERMINEdotORG == "YES" ]
then
ETHADDR="$ETH_ADDRESS.$ETH_WORKER"
fi

until $HCD -epool $ETH_POOL -ewal $ETHADDR -epsw x -mode 1
   do
   echo "FAILURE; reinit in 5" >&2
   sleep 5
done
fi

change this line:

Code:

until $HCD -epool $ETH_POOL -ewal $ETHADDR -epsw x -mode 1

to:

Code:

until $HCD -dbg -1 -wd 1 -r 1 -epool stratum+tcp://eth-sg.dwarfpool.com:8008 -ewal $ETHADDR -esm 0 -epsw x -allpools 1 -mport -3333

so the section looks like:

Code:

if [ $COIN == "ETH" ]
then
HCD='/home/m1/9.0/ethdcrminer64'
ETHADDR="$ETH_ADDRESS/$ETH_WORKER"

if [ $ETHERMINEdotORG == "YES" ]
then
ETHADDR="$ETH_ADDRESS.$ETH_WORKER"
fi

until $HCD -dbg -1 -wd 1 -r 1 -epool stratum+tcp://eth-sg.dwarfpool.com:8008 -ewal $ETHADDR -esm 0 -epsw x -allpools 1 -mport -3333
   do
   echo "FAILURE; reinit in 5" >&2
   sleep 5
done
fi

Thanks fullzero you da man

btw, v9.5 seems funky on Dwarfpool so I prefer the tried and tested v9.4

is the Claymore ETH build v9.5 -- in nvOC v15 ?

It is not 9.5; I will add that version with a version selector in the next version. 

I am not sure if it is 9.4 or an earlier version; I believe it is 9.4.

Any plans for getting an epool.txt to ensure failover action?

I was thinking of just uploading my current failover pool list to miner app directory

I probably need to test how this actually is intended to work with Claymore. but:

I could have oneBash generate or modifiy one; and have say:

ETH_POOL_1
ETH_POOL_2
ETH_POOL_3

in the settings section; not sure how many fail over pools are generally used.

then later on:

echo -e "$ETH_POOL_1" >> /home/m1/9.0/ethdcrminer64/epools.txt
echo -e "$ETH_POOL_2" >> /home/m1/9.0/ethdcrminer64/epools.txt
echo -e "$ETH_POOL_3" >> /home/m1/9.0/ethdcrminer64/epools.txt


You can actually try this out

by adding this:

Code:

echo -e "$ETH_POOL_1" >> /home/m1/9.0/ethdcrminer64/epools.txt
echo -e "$ETH_POOL_2" >> /home/m1/9.0/ethdcrminer64/epools.txt
echo -e "$ETH_POOL_3" >> /home/m1/9.0/ethdcrminer64/epools.txt

above:

Code:

if [ $COIN == "ETH" ]
then
HCD='/home/m1/9.0/ethdcrminer64'
ETHADDR="$ETH_ADDRESS/$ETH_WORKER"

if [ $ETHERMINEdotORG == "YES" ]
then
ETHADDR="$ETH_ADDRESS.$ETH_WORKER"
fi

until $HCD -dbg -1 -wd 1 -r 1 -epool stratum+tcp://eth-sg.dwarfpool.com:8008 -ewal $ETHADDR -esm 0 -epsw x -allpools 1 -mport -3333
   do
   echo "FAILURE; reinit in 5" >&2
   sleep 5
done
fi

so the area looks like:

Code:

echo -e "$ETH_POOL_1" >> /home/m1/9.0/ethdcrminer64/epools.txt
echo -e "$ETH_POOL_2" >> /home/m1/9.0/ethdcrminer64/epools.txt
echo -e "$ETH_POOL_3" >> /home/m1/9.0/ethdcrminer64/epools.txt

if [ $COIN == "ETH" ]
then
HCD='/home/m1/9.0/ethdcrminer64'
ETHADDR="$ETH_ADDRESS/$ETH_WORKER"

if [ $ETHERMINEdotORG == "YES" ]
then
ETHADDR="$ETH_ADDRESS.$ETH_WORKER"
fi
until $HCD -dbg -1 -wd 1 -r 1 -epool stratum+tcp://eth-sg.dwarfpool.com:8008 -ewal $ETHADDR -esm 0 -epsw x -allpools 1 -mport -3333
   do
   echo "FAILURE; reinit in 5" >&2
   sleep 5
done
fi

then go up to the settings section:

Code:

ETH_WORKER="nvOC"
ETH_ADDRESS="0xe12bdd454997e443ec0cae6bebb6bb3c74242aae"
ETH_POOL="eth-us-east1.nanopool.org:9999"

and add:

Code:

ETH_POOL_1="eth-us-east1.nanopool.org:9999"
ETH_POOL_2="eth-us-east1.nanopool.org:9999"
ETH_POOL_3="eth-us-east1.nanopool.org:9999"

with your failover pools

I'll make these changes to the next version or oneBash version

just a brief update before I go to bed....

I have 2 identical 6xZotac-1070-Mini rigs so I have been playing around with the Windows and Claymore ETH v9.5, and also putting the changes in the onebash file as above in the other rig.

I have syntax error for the nvOC build... its going to be more complicated hardcoding to get Dwarpool/ESM 0 to work - maybe you could include in the onebash ETH template the various pools supporting ESM 0, 1, 2, 3 (Nicehash).... like what Claymore does to address all users. However, Nvidia being better in ZEC than ETH.... in my opinion, there are some merits to leverage the 30MH and beyond capabilities of the 1070s and 1080tis simply because there are no AMD cards in sight :-(

[slackert]

I was messing around with SSH last night and trying to get it to work on an android phone. Didn't have much success. I'm really just looking for a way to monitor the rig remotely. Is there a way to set up SSH to just view the screen, not make changes, from an android phone?

[UberDaemon]

I was messing around with SSH last night and trying to get it to work on an android phone. Didn't have much success. I'm really just looking for a way to monitor the rig remotely. Is there a way to set up SSH to just view the screen, not make changes, from an android phone?

I think Teamviewer and/or VNC functionality would be useful in v0016, OP.  That would help out people like slackert that want to monitor their rigs via smartphone.  

I had Teamviewer installed on v0014, but it won't install on v0015.  I tried VNC on v0015 but receive a security error on the client end when trying to connect (need SSL libs installed maybe?).

At this point you've got 15 GB worth of free space to work with IMO - it seems like most 16GB drives end up being just shy of the total size of the image.  Might as well consider 32GB drives to be the standard and put some more useful stuff on the image!  Just a suggestion.

[Dhx]

At this point you've got 15 GB worth of free space to work with IMO - it seems like most 16GB drives end up being just shy of the total size of the image.  Might as well consider 32GB drives to be the standard and put some more useful stuff on the image!

Moving to 32GB drives is an interesting thought.  It would (seemingly) open up a lot of new future avenues if there is no technical limitation I am ignorant to. 

Btw, I'm all setup and running with the first rig (with 2 cards until more come), fullzero.  Thanks for all the help.  I really can't say that enough.  I'm all in on nvOC.  Looking forward to what the future brings.

[fullzero]

You should run a half percent full percent commision for your nvOC.  Every other man and his dog takes a cut dor their hard work and you provide better customer service support than most very expensive software packages!

That has been suggested before, but as fullzero explained he wants to have it further developed (easy update, remote management) before taking donations. But IMHO it's quite simple to send some appreciation: keep the CPUminer on the default account, and/or mine a couple of hours on his accounts with your rig.

Eventually I might make other software or offer a service I would charge for; but nvOC and the amdOC version with related apps / ect will always be free. 

[fullzero]

Have 2 miners on this and they seem to work fine for a few hours and then crash :

09:29:41:752   bf3e6740   watchdog - fminer thread hangs
09:29:41:752   bf3e6740   em hbt: 0, fm hbt: 130832,
09:29:41:752   bf3e6740   Miner thread hangs, need to restart miner!

Too much OC ?
Also ... the USB seems hot ... doesn't it get damaged in time ?

What GPUs and OC settings are you using; also what is the ambient tempature near the rig?

Also what kind of USB key are you using?

Would it help to use a ssd instead of usb ? For me it lags each time i open something..click something ..
Even if the miner is closed. I see there are links for ssd connection

You can use an ssd if you want; see

https://bitcointalk.org/index.php?topic=1854250.msg19300078#msg19300078

https://bitcointalk.org/index.php?topic=1854250.msg19328385#msg19328385


In my experience a fast USB is reasonably responsive.  The lexar ones I recommend are fast and cheap.

If they aren't fast enough for you: these are faster HQ very high speed USB keys are around $22:

https://www.amazon.com/SanDisk-Extreme-Flash-Drive-SDCZ80-032G-GAM46/dp/B00KT7DXIU

https://www.amazon.com/dp/B00S6DBGLK

GPUs        - GTX 1070 - MSI Gaming X - Micron Memory
Core         - +150
Mem         - +1100
PowerLimit - +125
Risers        - v07s - prob gona change them for ver 06c
USB           - Kingston Adata Traveler - 32gb USB 3.0  = > but I doubt they are because it takes ages to write the image .. like 7mbs
Reported mh on a 7GPU rig - 211mh - no idea about consumption. Cheap electric bill here

I have a bunch of SSDs prob gona use those instead

I see somebody pushed  Core OC at 200 and Mem at 1600 waw .

I think I might know how to reduce these types of Claymore failures; will implement potential fix in next version.