EWBF's CUDA Zcash miner

[mettalmag]

So far so good, just switched some of my cards to ZEC and this was the miner I was advised to use and here I am.
I have several questions I would to get answers to and find solutions.
I'm running  8xGigabyte 1080s on one rig and they are working like a charm, no specific config, nothing, except OC, all cards running about 565-575 Sol/s - mems 120/600. Uptime 16 days.
Second one is with 8x EVGA FTW 1080s, first try was 125/680 which was working fine for more then 12 hours, then I got "Looks like GPU0 are stopped. Restart attempt." after what I lowered the mems on same as Gigabyte 120/600, but I constantly getting the same error.
So the question is next: (I will solve the OC problem with EVGA but if you have any suggestions I'd be glad to hear that) how can I configure the miner that if it stops hashing to restart itself?
With Claymore's miner I'm using script which force stops the terminal itself and launches it again, but in case of Claymore I have a triger to run the bat file to do this (in case anyone interested here's the script)

Code:

@echo off
call :sub >log-%DATE:~-4%-%DATE:~4,2%-%DATE:~7,2%_%time:~0,2%%time:~3,2%%time:~6,2%.txt
exit /b

:sub
echo. #############################################
echo. %time% Check if old miner process exists and kill it
taskkill /f /im EthDcrMiner64.exe
echo. %time% Starting new Miner process one...
start /d "E:\Claymore's Miner" start.bat
echo. %time% Done!
timeout 5 >nul
exit 0

So, is there any solution with EWBF miner to make it more autonomous?

The next and less important for now question is how can I combine different rigs in one monitoring also like in Claymore's monitoring? For ether 100+ rigs are monitored with one monitoring instance and emailing and sms monitoring gives me ability to quickly respond if something happens, how is this possible in EWBF case?

This might help you to restart EWBF miner in case of any error:

Code:

:restart
TIMEOUT 8
miner --server zec-eu1.nanopool.org --port 6666 --user [USER] --pass x --eexit 3 --log 2 --logfile ZCASH.%date:~6,4%%date:~3,2%%date:~0,2%.ewbf.log --pec
goto :restart

I personally use 127.0.0.1:42000 to monitor a rig on localhost and if you add firewall exception in windows plus port forwarding in router and a DDNS service, this could be accessible from everywhere, I also use an uptime monitor service to ping this address for each rig and send me notification for any down time.
Hope that help you.

Thanks for this tip, what dows this command actually do --eexit 3? All others I'm familiar with. And btw are there any live commands in live terminal like s or r in Claymore's miner?
I'm using remote monitoring and it's way nicer and comfortable than any others out there, but so far it has been good only for live preview

[moeinbiz]

So far so good, just switched some of my cards to ZEC and this was the miner I was advised to use and here I am.
I have several questions I would to get answers to and find solutions.
I'm running  8xGigabyte 1080s on one rig and they are working like a charm, no specific config, nothing, except OC, all cards running about 565-575 Sol/s - mems 120/600. Uptime 16 days.
Second one is with 8x EVGA FTW 1080s, first try was 125/680 which was working fine for more then 12 hours, then I got "Looks like GPU0 are stopped. Restart attempt." after what I lowered the mems on same as Gigabyte 120/600, but I constantly getting the same error.
So the question is next: (I will solve the OC problem with EVGA but if you have any suggestions I'd be glad to hear that) how can I configure the miner that if it stops hashing to restart itself?
With Claymore's miner I'm using script which force stops the terminal itself and launches it again, but in case of Claymore I have a triger to run the bat file to do this (in case anyone interested here's the script)

Code:

@echo off
call :sub >log-%DATE:~-4%-%DATE:~4,2%-%DATE:~7,2%_%time:~0,2%%time:~3,2%%time:~6,2%.txt
exit /b

:sub
echo. #############################################
echo. %time% Check if old miner process exists and kill it
taskkill /f /im EthDcrMiner64.exe
echo. %time% Starting new Miner process one...
start /d "E:\Claymore's Miner" start.bat
echo. %time% Done!
timeout 5 >nul
exit 0

So, is there any solution with EWBF miner to make it more autonomous?

The next and less important for now question is how can I combine different rigs in one monitoring also like in Claymore's monitoring? For ether 100+ rigs are monitored with one monitoring instance and emailing and sms monitoring gives me ability to quickly respond if something happens, how is this possible in EWBF case?

This might help you to restart EWBF miner in case of any error:

Code:

:restart
TIMEOUT 8
miner --server zec-eu1.nanopool.org --port 6666 --user [USER] --pass x --eexit 3 --log 2 --logfile ZCASH.%date:~6,4%%date:~3,2%%date:~0,2%.ewbf.log --pec
goto :restart

I personally use 127.0.0.1:42000 to monitor a rig on localhost and if you add firewall exception in windows plus port forwarding in router and a DDNS service, this could be accessible from everywhere, I also use an uptime monitor service to ping this address for each rig and send me notification for any down time.
Hope that help you.

Thanks for this tip, what does this command actually do --eexit 3? All others I'm familiar with. And btw are there any live commands in live terminal like s or r in Claymore's miner?
I'm using remote monitoring and it's way nicer and comfortable than any others out there, but so far it has been good only for live preview

--eexit             Exit in case of error. Value 1 exit if miner cannot restart workers.
                    Value 2 if lost connection with the pool. 3 both cases.

--eexit 3 would close miner if couldn't restart miner or lost connection so batch file can start miner again.
Unfortunately EWBF's commands are limited and I don't think there is any runtime command.

[douirc]

Guys, do you know how to automatically restart the EWBF miner when the card is not response and produce 0 sols? I ran 7 1080ti on Win 10, about 5-10 hours, the 7 card stop working and produce 0 sols although the miner tried to restart the card several times, so the result is I need to close the miner and open in again in order making it to work. I run in bios stock and 80% power limit

Does it close the miner if you use --eexit 3?

If so you could just run the miner in a loop:

:loop
miner.exe ...
goto loop


If it doesn't quit and just get stuck then you could restart the miner periodically (like every couple hours) by running it in a loop like above and have another loop running which terminates the miner. Not ideal but better than having the cards not working for hours:

:restartloop
timeout -t 3600
taskkill -t -f /im miner.exe
goto restartloop

brilliant.  thank you.

[chiwalfrm]

I have two Nvidia 1080 GTX Ti's and I only want to mine on my second GPU.  When I ran the miner with "--cuda_devices 1", I checked nvidia-smi command and it says the miner is running on both cards.  Why is it running on the first GPU?

Screenshot: http://imgur.com/a/cCvT4

[Chipychop]

hi all!

do you consider obtaining 440-450 sol/s for a gtx 1070 strix is nice??? my temps are 78º celsius with fan at 75%

I've applied this parameters:

+4% core voltage
103% power limit
+99 core
+261 mem

can I oc it more? do you consider safely?

thanks!!!!!!

[Mikanoshi]

do you consider obtaining 440-450 sol/s for a gtx 1070 strix is nice???

It's normal.

my temps are 78º celsius with fan at 75%

Did you put it in the oven or something?) You should limit PL to downvolt/downclock it, 10-20 sols don't worth it anyway.
I'm running Palit JS 1080 @ 80% PL, 0.98-1 volt, core 1936, mem 10500 (stock for flashed Palit GRP BIOS), 67º, 67% fan and it gives 580 sols instead of about 600+ at full overclock 2050 MHz @ 1.093v.

[nmask]

Hey,

Do you find this ok or should I OC?
Trying to keep balance in power consumption and performance.

https://preview.ibb.co/kyAAVF/09_2.jpg

[luka1002]

I am using nvidia 1070 with :

+ 100 core
+ 800 mem
130w TDP limit and getting around 445 sol/s

Runing 8x1070 with i PSU 1200W.

Using linux mint.

Is there and option to add if miner got status exit 3 than reboot? Also in web stats is should appear some statistic like average sols for each VGA, average temps .... it could be useful..

Regards.

Luka

[mimimi2]

Hi. How can I monitor the farm remotely? Is there a prog analogue of EthMan?

[sublimus]

Hi. How can I monitor the farm remotely? Is there a prog analogue of EthMan?

awesomeminer.com

[TheHypnoToad]

I would get the new Galax 1070. I know they are a cheaper brand but I have 2 Galax 1060 6GB and they are great at mining ZEC or ETH.

Pascal chipset cards (10xx) use much less power for the same hash-rate as other cards deliver.

Also, if you decide to re-sell it, you are much more likely to get more of your money back with a Galax 1070

still funny how some people recommend nvidia for mining

nothing better than rx470-570's     , it s not for nothing you can not find any on sale

less power argument is not correct ever  the 570's undervolt pretty good

Hello, how many Sol/s would I get from an undervolted 570? How much power would it use undervolted? Thanks

check amd thread , not sure anymore , it s lot more profitable on eth  if my memory is correct  rx570 does some more than 280 on zec (up to 300 with OC)
power on eth is 100-112w   so for zec it s a bit less (or same but higher clock)

and so definetly cheaper and faster than an gtx1060 3G

i have a model with1 small fan, cannot oc higher because it hit 84° and clock down , it s only doing 250sol
i did not under volt yet

Hi, thanks for the info. For me the RX570 is around the same price as a GTX1070, the 1070 is tough to beat as it does 400Sol/s at ~100W

[sverox]

New variant of my start/restart/0 sols watch script.

Info:

EWBF Miner Auto Start/Restart Script by sverox

INFO:
This is autostart script for great EWBF's CUDA Zcash miner
https://bitcointalk.org/index.php?topic=1707546.0


FEATURES:
1. Start, autostart, watch and restart miner.
2. Creates new log for every start of miner with filename timestamps.
(for paranoids - do not worry doff.exe is safe to use tool).
3. Watch for error "0 Sol/s" state and force kill/restart miner.
4. Give some stats for loops/starts/restarts count.


USAGE:
1. Carefully edit _start_ZEC_flypool_xxx.bat with your datas.

2. Unpack Doff.zip doff.exe to path (is return timestamps for log filenames)
(c:\Windows\System32\doff.exe
c:\Windows\SysWOW64\doff.exe)

3. Put _start_ZEC_flypool_xxx.bat on your startup folder
(For the Current User Startup Folder, open the Run dialog and type shell:startup.)

Start your miner with _start_ZEC_flypool_xxx.bat

Donations welcome to ZEC: t1JW33AfzoeUN3EBnPcmxPM6KgFWcG2GoQC


DISCLAIMER:
NO WARRANTY OF ANY TYPE


COPYRIGHT:
No copyrights, no copylefts - use and modify as you want.

Download:

https://mega.nz/#!gTYxwSBb!09BmkgW6O3AePjqWX4Yr5W1FK2-VChZLtlCR4G3fdDg

[fr4nkthetank]

put pause in :

this is your bat file and settings.exe
pause

so you'll see what the problem is.  Edit:  its better to just launch .bat with miner.exe, and it will load the config file, which is nice to use.

[moeinbiz]

Hello my friends, im new to this EWBF, im trying to make a bat, or testing the ones existing and every time i open some bat, opens a windows very fast and closes. What is going wrong, i've tried everything , other miners work very well, but im only having issues with this one.
Any help would be appreciated, thanks a lot!

You can run miner.exe including all parameters in command prompt so you can still see output and error messages

[QuintLeo]

I have two Nvidia 1080 GTX Ti's and I only want to mine on my second GPU.  When I ran the miner with "--cuda_devices 1", I checked nvidia-smi command and it says the miner is running on both cards.  Why is it running on the first GPU?

Screenshot: http://imgur.com/a/cCvT4

 GPU Utilization zero on that card, not sure it's actually running ANYTHING.

 Or it's running the "post info to the display" code, since it seems to be the card that's running the display.

 Power consumption on your #0 card is also very low.

[Chipychop]

do you consider obtaining 440-450 sol/s for a gtx 1070 strix is nice???

It's normal.

my temps are 78º celsius with fan at 75%

Did you put it in the oven or something?) You should limit PL to downvolt/downclock it, 10-20 sols don't worth it anyway.
I'm running Palit JS 1080 @ 80% PL, 0.98-1 volt, core 1936, mem 10500 (stock for flashed Palit GRP BIOS), 67º, 67% fan and it gives 580 sols instead of about 600+ at full overclock 2050 MHz @ 1.093v.

thanks!!!

which core voltage do you use?

[The_Immortal]

Well guys, very need your help.

For unknown reasons my 1050Ti 4GB MSI GAMING X lost its hashrate - now it gives about 90-95 sol/sec in OC mode and about 85 sol/sec in stock settings. Its normal rate is 175-180 sol/sec in stock settings and 190 sol/sec in OC mode.
MSI GAMING X has a wonderful cooling system so 58º C is the maximum I saw on this card.

I've reinstalled drivers (and tried the recent one) but it didnt help...

What's the problem guys?.. 

[Mikanoshi]

which core voltage do you use?

Whichever gives 80% PL, it adjusts voltage automatically. No more than 1.0v usually.

[jenifive]

4x Palit 1070 Super Jetstream with 70 limit, 90 core, 350 memory

Nvidia driver 382.05

Version is 0.3.4b

Why my miner stopping every time?

I dont have such issues on Claymore or ccminer.

this miner is shit, its continues crashing...

[sublimus]

4x Palit 1070 Super Jetstream with 70 limit, 90 core, 350 memory

Nvidia driver 382.05

Version is 0.3.4b

Why my miner stopping every time?

I dont have such issues on Claymore or ccminer.

this miner is shit, its continues crashing...

maybe Teamviewer making it crash.
Does it still crash with stock settings?
http://www.tomshardware.com/forum/264823-31-troubleshooting#t1934282