EWBF's CUDA Zcash miner

[Edward_Teach]

Why am I getting hit with DEVFEES every other minute? I thought it was supposed to be every 10 minutes.

[1714877460]

1714877460

[1714877460]

1714877460

[1714877460]

1714877460

[1714877460]

1714877460

[1714877460]

1714877460

[Larz952002]

I am also getting hit with Dev fees pretty frequently.  I think it collects a DEV fee per card but I did see it hit GPU 2 twice in 30 seconds.  Maybe the DEV has some input on this

[Edward_Teach]

I am also getting hit with Dev fees pretty frequently.  I think it collects a DEV fee per card but I did see it hit GPU 2 twice in 30 seconds.  Maybe the DEV has some input on this

I got hit 10 times in 4 minutes. I thought that was a little excessive.

[siampumpkin]

Please add a watchdog or some other type of reset command if GPU hangs. I have some of my rigs at a remote co-location center and without a reboot command I cannot use this program for those rigs.

Thanks. 

[jackbox]

I am also getting hit with Dev fees pretty frequently.  I think it collects a DEV fee per card but I did see it hit GPU 2 twice in 30 seconds.  Maybe the DEV has some input on this

I got hit 10 times in 4 minutes. I thought that was a little excessive.

If the fees are that excessive add the"--fee 0" option to remove dev fees.

[Robcsa]

I am also getting hit with Dev fees pretty frequently.  I think it collects a DEV fee per card but I did see it hit GPU 2 twice in 30 seconds.  Maybe the DEV has some input on this

I got hit 10 times in 4 minutes. I thought that was a little excessive.

If the fees are that excessive add the"--fee 0" option to remove dev fees.

Yeah but if you turn off the dev fee, than your miner will mine a little bit slover.

[Robcsa]

Please add a watchdog or some other type of reset command if GPU hangs. I have some of my rigs at a remote co-location center and without a reboot command I cannot use this program for those rigs.

Thanks. 

If you are using Windows than you can try my monitorizing tool.
https://drive.google.com/file/d/0B4Yvkc957FWKVjloYTU2YWMxalU/view?usp=sharing

[sublimus]

Why am I getting hit with DEVFEES every other minute? I thought it was supposed to be every 10 minutes.

Unlike other devs EWBF collects fees every n...Cycles instead of every n....Minutes.

[sublimus]

Please add a watchdog or some other type of reset command if GPU hangs. I have some of my rigs at a remote co-location center and without a reboot command I cannot use this program for those rigs.

Thanks. 

If you are using Windows than you can try my monitorizing tool.
https://drive.google.com/file/d/0B4Yvkc957FWKVjloYTU2YWMxalU/view?usp=sharing

Does your watchdog works under WiFi or Ethernet only?

[farmtester]

Why am I getting hit with DEVFEES every other minute? I thought it was supposed to be every 10 minutes.

Unlike other devs EWBF collects fees every n...Cycles instead of every n....Minutes.

Every number of cycles, what does that mean?

[Robcsa]

Please add a watchdog or some other type of reset command if GPU hangs. I have some of my rigs at a remote co-location center and without a reboot command I cannot use this program for those rigs.

Thanks. 

If you are using Windows than you can try my monitorizing tool.
https://drive.google.com/file/d/0B4Yvkc957FWKVjloYTU2YWMxalU/view?usp=sharing

Does your watchdog works under WiFi or Ethernet only?

It is working under both networks. Actually network needs only for the e-mail notifications it can be work "offline". The application needs to be placed on that computer where your ewbf cuda miner is running. If you start my app it will starting to check the miner's status in every few sec ( if api enabled in ewbf cuda miner ). If the miner is fails (including hanging GPU) it will restart the miner automatically. An email notif will be sent if the miner is stopped. There is also an e-mail notification in every 30 min about the status of your rig like power usage, speed, temperature, accepted shares rejected shares for each of your cards.

[sublimus]

It is working under both networks. Actually network needs only for the e-mail notifications it can be work "offline". The application needs to be placed on that computer where your ewbf cuda miner is running. If you start my app it will starting to check the miner's status in every few sec ( if api enabled in ewbf cuda miner ). If the miner is fails (including hanging GPU) it will restart the miner automatically. An email notif will be sent if the miner is stopped. There is also an e-mail notification in every 30 min about the status of your rig like power usage, speed, temperature, accepted shares rejected shares for each of your cards.

1. status notification's every 30 minutes/rig means a LOT of emails, is there a switch to adjust it ? I do not see any in the readme file.


2. I also have reserves about the 12 minutes mining at once. Should be distributed more evenly. I do not like to set dev fees to 0 because I think every worker deserves his retribution.
    OK, 12min/day is only 30 sec per hour but 12 min at once makes me unconfortable. (just saying)

[Robcsa]

It is working under both networks. Actually network needs only for the e-mail notifications it can be work "offline". The application needs to be placed on that computer where your ewbf cuda miner is running. If you start my app it will starting to check the miner's status in every few sec ( if api enabled in ewbf cuda miner ). If the miner is fails (including hanging GPU) it will restart the miner automatically. An email notif will be sent if the miner is stopped. There is also an e-mail notification in every 30 min about the status of your rig like power usage, speed, temperature, accepted shares rejected shares for each of your cards.

1. status notification's every 30 minutes/rig means a LOT of emails, is there a switch to adjust it ? I do not see any in the readme file.


2. I also have reserves about the 12 minutes mining at once. Should be distributed more evenly. I do not like to set dev fees to 0 because I think every worker deserves his retribution.
    OK, 12min/day is only 30 sec per hour but 12 min at once makes me unconfortable. (just saying)

Thank you for your Feedback. Currently no option to set the frequency of e-mail sending, but according to your request i'm going to add a new parameter where you can set the time frequency of e-mail sending.

2 if you feel that unconfortable just turn off developer fee and if you want to support you can donate on the given addresses.

[NvidiaMining]

@NvidiaMining
Thanks dude.

IMO you need find stable undervolt + overclock settings.
Goal for 1070 stable 450-470 sols. 1080 stable is 730-760 sols.

Full restart PC (took minutes) lower your shares to the pool (mean lower profit). So avoid full restart PCs.
Restart miner.exe is not so important (of course i talk for seconds delays).

Hey sverox,
you are right of course, i looked into it again. So your script has prevented every error (there was no 0 sol incident during the time), but i have noticed something else, which happened on both rigs individually.

So it starts like this: I look into the pool page and see that 1 rig does not contribute as much as it should. I access the rig and notice that the mining program is doing absolutely nothing and the last share submitted was some time ago (5-10 minutes). There is no error, the power usage on the wall stays the same and the mining program appears to be running, but is not producing shares. If i open a 2nd miner instance it works properly, but with only half the regular speed (which is what you would expect).

However, if you click on the mining program that is not working properly and press "spacebar" the program wakes up and tells that a Devfee share was rejected, essentially blocking the entire miner without an error message, and resulting in many further rejected shares. I was able to capture it on screen: the yellow arrow shows where the miner stopped, after i pressed spacebar the following lines below the arrow were shown. Please also note how the Devfee share line starts with "Temp" instead of "Info" and how the "Temp" line has an earlier time stamp than the share before. The following lines have stamps +5 minutes, indicating the time when i pressed space.

edit: And after all the rejected shares are sent anyway the miner works properly again

So long story short Would it be possible for you to include some code in your script that "presses spacebar" everytime your script checks the miner? I would be thankful forever! Your support is very much appreciated!

http://imgur.com/a/W47Ei

edit2: it happened again
http://imgur.com/a/RJNMr

[djn053]

when will this 0.3.4b update？
0 Sol and stop is very annoyed.........
waste lot of time.....

[chronossc]

Hello folks!

So I have here two GPUs, one EVGA GTX 1060 6GB SC and one EVGA GTX 970 FTW+. The first is used exclusively to mining, the last is used to mining when I'm not using the machine.

In order to overclock and enable custom fan control of GTX 1060 I need to configure Xorg to have a ServerLayout entry, like that:

Code:

Section "ServerLayout"
    Identifier     "Layout0"
    Screen         "Screen0"
    Screen         "Screen1"
EndSection

Check full file in https://gist.github.com/anonymous/39706554209e5c4e1889319d265727ca (don't like to plot a lot of text in this post :]).

miner.cfg without pools:

Code:

[common]
cuda_devices 0 #1
intensity    64 #64
templimit    70
pec          1
boff         0
eexit        3
tempunits    c
log          1
logfile      miner.log
api          0.0.0.0:42000

miner output, usually got to 300 sol/s or more, but with ServerLayout configuration:

Code:

 ~/projects/mining/ewbf-nanopool-zec-miner 
❯ ./miner
+-------------------------------------------------+
|         EWBF's Zcash CUDA miner. 0.3.4b         |
+-------------------------------------------------+
INFO: Used config file: miner.cfg
INFO: Current pool: zec-us-east1.nanopool.org:6666
INFO: Selected pools: 5
INFO: Solver: Auto.
INFO: Devices: User defined.
INFO: Temperature limit: 70
INFO: Api: Listen on 0.0.0.0:42000
---------------------------------------------------
INFO: Target: 000369d000000000...
INFO: Detected new work: 1501405872
CUDA: Device: 0 GeForce GTX 1060 6GB, 6072 MB i:64
CUDA: Device: 1 GeForce GTX 970 Not selected.
INFO: Detected new work: 1501405873
CUDA: Device: 0 Selected solver: 0
Temp: GPU0: 39C 
GPU0: 0 Sol/s 
Total speed: 0 Sol/s
+-----+-------------+--------------+
| GPU | Power usage |  Efficiency  |
+-----+-------------+--------------+
|  0  |     30W     |  0.00 Sol/W  |
+-----+-------------+--------------+
CUDA: Device: 0 Thread exited with code: 6
INFO: Detected new work: 1501405874
Temp: GPU0: 37C 
GPU0: 0 Sol/s 
Total speed: 0 Sol/s
+-----+-------------+--------------+
| GPU | Power usage |  Efficiency  |
+-----+-------------+--------------+
|  0  |     22W     |  0.00 Sol/W  |
+-----+-------------+--------------+
Temp: GPU0: 37C 
GPU0: 0 Sol/s 
Total speed: 0 Sol/s
+-----+-------------+--------------+
| GPU | Power usage |  Efficiency  |
+-----+-------------+--------------+
|  0  |     22W     |  0.00 Sol/W  |
+-----+-------------+--------------+
INFO: Detected new work: 1501405875
ERROR: Looks like GPU0 are stopped. Restart attempt.
INFO: GPU0 are restarted.
CUDA: Device: 0 User selected solver: 0
CUDA: Device: 0 Thread exited with code: 46
^C

miner.log:

Code:

CUDA: Device: 0 Thread exited with code: 6
ERROR: Looks like GPU0 are stopped. Restart attempt.
CUDA: Device: 0 Thread exited with code: 46

So finally, if I don't use the ServerLayout with two screens in xorg.conf, miner runs very well reaching 300 sol/s for GTX 1060, but I can't overclock or control it's fans. I know that ServerLayout works well cuz I use it to mine Ethereum with Claymore miner and got GTX 1060 to 21.7 Mh/s.

Someone know how to solve this bug? Some word from DEV?

Have nice weekstart .

[Robcsa]

when will this 0.3.4b update？
0 Sol and stop is very annoyed.........
waste lot of time.....

Hello if you want. You can try my monitorizing tool to solve this issue:
https://bitcointalk.org/index.php?topic=2053739.0

[Robcsa]

Hello folks!

So I have here two GPUs, one EVGA GTX 1060 6GB SC and one EVGA GTX 970 FTW+. The first is used exclusively to mining, the last is used to mining when I'm not using the machine.

In order to overclock and enable custom fan control of GTX 1060 I need to configure Xorg to have a ServerLayout entry, like that:

Code:

Section "ServerLayout"
    Identifier     "Layout0"
    Screen         "Screen0"
    Screen         "Screen1"
EndSection

Check full file in https://gist.github.com/anonymous/39706554209e5c4e1889319d265727ca (don't like to plot a lot of text in this post :]).

miner.cfg without pools:

Code:

[common]
cuda_devices 0 #1
intensity    64 #64
templimit    70
pec          1
boff         0
eexit        3
tempunits    c
log          1
logfile      miner.log
api          0.0.0.0:42000

miner output, usually got to 300 sol/s or more, but with ServerLayout configuration:

Code:

 ~/projects/mining/ewbf-nanopool-zec-miner 
❯ ./miner
+-------------------------------------------------+
|         EWBF's Zcash CUDA miner. 0.3.4b         |
+-------------------------------------------------+
INFO: Used config file: miner.cfg
INFO: Current pool: zec-us-east1.nanopool.org:6666
INFO: Selected pools: 5
INFO: Solver: Auto.
INFO: Devices: User defined.
INFO: Temperature limit: 70
INFO: Api: Listen on 0.0.0.0:42000
---------------------------------------------------
INFO: Target: 000369d000000000...
INFO: Detected new work: 1501405872
CUDA: Device: 0 GeForce GTX 1060 6GB, 6072 MB i:64
CUDA: Device: 1 GeForce GTX 970 Not selected.
INFO: Detected new work: 1501405873
CUDA: Device: 0 Selected solver: 0
Temp: GPU0: 39C 
GPU0: 0 Sol/s 
Total speed: 0 Sol/s
+-----+-------------+--------------+
| GPU | Power usage |  Efficiency  |
+-----+-------------+--------------+
|  0  |     30W     |  0.00 Sol/W  |
+-----+-------------+--------------+
CUDA: Device: 0 Thread exited with code: 6
INFO: Detected new work: 1501405874
Temp: GPU0: 37C 
GPU0: 0 Sol/s 
Total speed: 0 Sol/s
+-----+-------------+--------------+
| GPU | Power usage |  Efficiency  |
+-----+-------------+--------------+
|  0  |     22W     |  0.00 Sol/W  |
+-----+-------------+--------------+
Temp: GPU0: 37C 
GPU0: 0 Sol/s 
Total speed: 0 Sol/s
+-----+-------------+--------------+
| GPU | Power usage |  Efficiency  |
+-----+-------------+--------------+
|  0  |     22W     |  0.00 Sol/W  |
+-----+-------------+--------------+
INFO: Detected new work: 1501405875
ERROR: Looks like GPU0 are stopped. Restart attempt.
INFO: GPU0 are restarted.
CUDA: Device: 0 User selected solver: 0
CUDA: Device: 0 Thread exited with code: 46
^C

miner.log:

Code:

CUDA: Device: 0 Thread exited with code: 6
ERROR: Looks like GPU0 are stopped. Restart attempt.
CUDA: Device: 0 Thread exited with code: 46

So finally, if I don't use the ServerLayout with two screens in xorg.conf, miner runs very well reaching 300 sol/s for GTX 1060, but I can't overclock or control it's fans. I know that ServerLayout works well cuz I use it to mine Ethereum with Claymore miner and got GTX 1060 to 21.7 Mh/s.

Someone know how to solve this bug? Some word from DEV?

Have nice weekstart .

See my other post  above.

[threeflappp]

when will this 0.3.4b update？
0 Sol and stop is very annoyed.........
waste lot of time.....

Usually this means that your OC is not stable...

But you can check out @sverox's script here: https://bitcointalk.org/index.php?topic=1707546.msg20430756#msg20430756

I OC as much as I can so my GPU crashes every 15 hours or so. His script hasn't failed to keep the miner running yet.

Although, this doesn't actually restart the computer, it only handles the miner.

[sverox]

@NvidiaMining

See your PM.