Pages:
Author

Topic: L3+/L3++ voltage tuning script - page 2. (Read 3509 times)

legendary
Activity: 1612
Merit: 1608
精神分析的爸
July 27, 2018, 04:12:28 AM
#70
Any chance you can mod this script to work with the BlissZ firmware?

AFAIK it should work with Blissz firmware too, somebody helped me earlier to adapt it for the json output of Blissz:

https://github.com/psycodad-bct/bitmain-tools/blob/master/scripts/l3plus_autotune.py#L102

If you have problems running it, please post the output that the script is spewing out before it terminates.
legendary
Activity: 3080
Merit: 1083
July 27, 2018, 02:14:42 AM
#69
Any chance you can mod this script to work with the BlissZ firmware?
legendary
Activity: 1612
Merit: 1608
精神分析的爸
July 23, 2018, 12:37:42 PM
#68
Since many asked how much you can save with down-volting, I took the opportunity to make some measurements with the new L3+ I received today with a Wattmeter.
The power supply used in this test is an EVGA 1600 platinum rated, power supply input was ~236VAC and the ambient temp was ~27-28°C:

L3+(+) Model Frequency Temperatures Wattage Voltage settings Errors/min (avg)
L3+ (07/18)38460/61/59/57878W0x80/0x80/0x80/0x800/0/0/0
L3+ (07/18)38456/56/55/53733W*0xfe/0xfe/0xfe/0xfe0.01/0.04/0/0

Code:
| Temp Chips | 56 C | 56 C | 55 C | 53 C |
| Err 5min   | 0.00 | 0.00 | 0.00 | 0.00 |
| Err 10min  | 0.00 | 0.00 | 0.00 | 0.00 |
| Err 15min  | 0.07 | 0.07 | 0.00 | 0.00 |
| Err All    | 0.01 | 0.04 | 0.00 | 0.00 |
**************************************************
| Start 18:14.52 | 0x80 | 0x80 | 0x80 | 0x80 |
| Start 18:14.52 | 60 C | 61 C | 59 C | 57 C |
| End   19:29.55 | 0xfe | 0xfe | 0xfe | 0xfe |
| End   19:29.55 | 56 C | 56 C | 55 C | 53 C |
**************************************************
Report written to /tmp/192.168.9.221-rxYPcO.rep
Finished tuning, miner stable AFAICS

*That's a whopping ~145W or ~16.5% savings, I did really expect more like 70-90W so I am quite (positively) surprised.

When I do further experiments with different frequencies I will update this post with new data, others are welcome to contribute their findings.

HTH
legendary
Activity: 1612
Merit: 1608
精神分析的爸
June 25, 2018, 02:02:01 PM
#67
That's a relief to hear! Still can't believe my own stupidity to oversee this.
Dont be sor hard on yourself
It works like a charm now
Thanks so so much
send me your BTC adress please for your time and help

The script shows a BTC and LTC address of mine when it finishes or gets aborted (CTRL-C), though it is really not necessary, your patience and help in getting this fixed were greatly appreciated and as you can see the topic now proudly claims "L3+/L3++..."  Cheesy


That's a relief to hear! Still can't believe my own stupidity to oversee this.

Hi,
I had a power outtage while running the script
now I get an error msg trying to run it again
The L3++ run and hash properly
I reboot them
remove the power for 5 mins and retried
still get that error since the power outtage

jhgosselin@Master-VRTJMB:~/bitmain-tools/scripts$ ./l3plus_autotune.py -i 10.0.100.135
/usr/lib/python2.7/dist-packages/Crypto/Cipher/blockalgo.py:141: FutureWarning: CTR mode needs counter parameter, not IV
  self._cipher = factory.new(key, *args, **kwargs)
Undefined errors occured fetching voltage settings from miner:
sh: /config/sv: Permission denied


Just for curiousity: Did that happen on all your L3++, multiples or just one?

Sounds like the sv binary on the Antminer got corrupted from the power outage, the easiest way to resolve this is by removing the file:

- ssh to the l3++ in question as root
- remove the file with "rm /config/sv"

Once it is removed (you can check this with "ls -la /config/") you just run the script again and it will install a fresh copy of it when it finds it missing.

HTH
jr. member
Activity: 71
Merit: 1
June 25, 2018, 10:10:18 AM
#66
That's a relief to hear! Still can't believe my own stupidity to oversee this.

Hi,
I had a power outtage while running the script
now I get an error msg trying to run it again
The L3++ run and hash properly
I reboot them
remove the power for 5 mins and retried
still get that error since the power outtage

jhgosselin@Master-VRTJMB:~/bitmain-tools/scripts$ ./l3plus_autotune.py -i 10.0.100.135
/usr/lib/python2.7/dist-packages/Crypto/Cipher/blockalgo.py:141: FutureWarning: CTR mode needs counter parameter, not IV
  self._cipher = factory.new(key, *args, **kwargs)
Undefined errors occured fetching voltage settings from miner:
sh: /config/sv: Permission denied
jr. member
Activity: 71
Merit: 1
June 24, 2018, 10:13:46 AM
#65
That's a relief to hear! Still can't believe my own stupidity to oversee this.
Dont be sor hard on yourself
It works like a charm now
Thanks so so much
send me your BTC adress please for your time and help
legendary
Activity: 1612
Merit: 1608
精神分析的爸
June 24, 2018, 09:25:11 AM
#64
That's a relief to hear! Still can't believe my own stupidity to oversee this.
jr. member
Activity: 71
Merit: 1
June 24, 2018, 03:40:23 AM
#63
I feared this wouldn't fix it, would have surprised me somehow anyways but everybody gets a lucky shot at times  Cool

I just don't seem to see the problem yet, I even have setup a "fake/simulated" L3++ for my script that hands out the json reply from you and guess what: works nicely.

Need some sleep first and will look over it tomorrow again, right now I am out of ideas and its hard to fix a problem I don't see and can't reproduce.

Though thanks for your support and input so far, that's motivating.



Thanks, I guess, but you are helping me out a lot more then I help you, I am the one with L3++ needing your help Smiley

I had another potentially uber brilliant idea for a fix  Grin Could you do 'git pull' and give it another try?

you are a genious lol
it seems to be working
I am running the script on a L3++ now no error, will let you know the outcome Smiley
legendary
Activity: 1612
Merit: 1608
精神分析的爸
June 23, 2018, 10:13:16 AM
#62
I feared this wouldn't fix it, would have surprised me somehow anyways but everybody gets a lucky shot at times  Cool

I just don't seem to see the problem yet, I even have setup a "fake/simulated" L3++ for my script that hands out the json reply from you and guess what: works nicely.

Need some sleep first and will look over it tomorrow again, right now I am out of ideas and its hard to fix a problem I don't see and can't reproduce.

Though thanks for your support and input so far, that's motivating.



Thanks, I guess, but you are helping me out a lot more then I help you, I am the one with L3++ needing your help Smiley

I had another potentially uber brilliant idea for a fix  Grin Could you do 'git pull' and give it another try?
jr. member
Activity: 71
Merit: 1
June 21, 2018, 07:46:02 PM
#61
I feared this wouldn't fix it, would have surprised me somehow anyways but everybody gets a lucky shot at times  Cool

I just don't seem to see the problem yet, I even have setup a "fake/simulated" L3++ for my script that hands out the json reply from you and guess what: works nicely.

Need some sleep first and will look over it tomorrow again, right now I am out of ideas and its hard to fix a problem I don't see and can't reproduce.

Though thanks for your support and input so far, that's motivating.



Thanks, I guess, but you are helping me out a lot more then I help you, I am the one with L3++ needing your help Smiley
legendary
Activity: 1612
Merit: 1608
精神分析的爸
June 21, 2018, 02:39:26 PM
#60
I feared this wouldn't fix it, would have surprised me somehow anyways but everybody gets a lucky shot at times  Cool

I just don't seem to see the problem yet, I even have setup a "fake/simulated" L3++ for my script that hands out the json reply from you and guess what: works nicely.

Need some sleep first and will look over it tomorrow again, right now I am out of ideas and its hard to fix a problem I don't see and can't reproduce.

Though thanks for your support and input so far, that's motivating.

jr. member
Activity: 71
Merit: 1
June 21, 2018, 04:47:51 AM
#59
Sorry for not getting back earlier, RL got some tasks for me lately.

I can only speculate that there are extraneous whitespace characters at the end of the json output, though actually the json module should be able to deal with that. I've made an attempt at fixing this by striping all whitespace chars outside the json response before parsing it, though I have some doubts it will fix it. But it's the best I could come up with from remote.

Please try to git pull and see if it improves it for you.


What, you have a Real Life !!!!!
Doesnt work
This is the msg I get
What's next Smiley

jhgosselin@Master-VRTJMB:~/bitmain-tools/scripts$ sudo git pull
[sudo] password for jhgosselin:
remote: Counting objects: 4, done.
remote: Compressing objects: 100% (3/3), done.
remote: Total 4 (delta 1), reused 4 (delta 1), pack-reused 0
Unpacking objects: 100% (4/4), done.
From https://github.com/psycodad-bct/bitmain-tools
   6349d9d..34f00a4  master     -> origin/master
Updating 6349d9d..34f00a4
Fast-forward
 scripts/l3plus_autotune.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
jhgosselin@Master-VRTJMB:~/bitmain-tools/scripts$ ./l3plus_autotune.py -i 10.0.100.118
/usr/lib/python2.7/dist-packages/Crypto/Cipher/blockalgo.py:141: FutureWarning: CTR mode needs counter parameter, not IV
  self._cipher = factory.new(key, *args, **kwargs)
Failed to decode json reply:
Extra data: line 1 column 1305 - line 1 column 1306 (char 1304 - 1305)

{"STATUS":[{"STATUS":"S","When":1529574372,"Code":70,"Msg":"CGMiner stats","Description":"cgminer 4.9.0"}],"STATS":[{"CGMiner":"4.9.0","Miner":"1.0.1.3","CompileTime":"Thu Apr 12 15:55:21 CST 2018","Type":"Antminer L3++"},{"STATS":0,"ID":"L30","Elapsed":38043,"Calls":0,"Wait":0.000000,"Max":0.000000,"Min":99999999.000000,"GHS 5s":"594.83","GHS av":585.57,"miner_count":4,"frequency":"450","fan_num":2,"fan1":2640,"fan2":2670,"temp_num":4,"temp1":55,"temp2":52,"temp3":50,"temp4":48,"temp2_1":63,"temp2_2":61,"temp2_3":59,"temp2_4":56,"temp31":0,"temp32":0,"temp33":0,"temp34":0,"temp4_1":0,"temp4_2":0,"temp4_3":0,"temp4_4":0,"temp_max":55,"Device Hardware%":0.0000,"no_matching_work":2137,"chain_acn1":72,"chain_acn2":72,"chain_acn3":72,"chain_acn4":72,"chain_acs1":" oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo","chain_acs2":" oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo","chain_acs3":" oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo","chain_acs4":" oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo","chain_hw1":0,"chain_hw2":0,"chain_hw3":1,"chain_hw4":2136,"chain_rate1":"144.93","chain_rate2":"151.31","chain_rate3":"153.99","chain_rate4":"144.60"}],"id":1}

legendary
Activity: 1612
Merit: 1608
精神分析的爸
June 21, 2018, 03:36:51 AM
#58
Sorry for not getting back earlier, RL got some tasks for me lately.

I can only speculate that there are extraneous whitespace characters at the end of the json output, though actually the json module should be able to deal with that. I've made an attempt at fixing this by striping all whitespace chars outside the json response before parsing it, though I have some doubts it will fix it. But it's the best I could come up with from remote.

Please try to git pull and see if it improves it for you.

jr. member
Activity: 71
Merit: 1
June 17, 2018, 11:54:56 AM
#57
jhgosselin@Master-VRTJMB:~$ sudo apt-get install netcat-openbsd
[sudo] password for jhgosselin:
Reading package lists... Done
Building dependency tree
Reading state information... Done
netcat-openbsd is already the newest version (1.105-7ubuntu1).
The following package was automatically installed and is no longer required:
  libfreetype6
Use 'sudo apt autoremove' to remove it.
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
jhgosselin@Master-VRTJMB:~$

jhgosselin@Master-VRTJMB:~$ echo "{\"command\":\"stats\"}"| nc 10.0.100.141 4028
{"STATUS":[{"STATUS":"S","When":1529254514,"Code":70,"Msg":"CGMiner stats","Description":"cgminer 4.9.0"}],"STATS":[{"CGMiner":"4.9.0","Miner":"1.0.1.3","CompileTime":"Thu Apr 12 15:55:21 CST 2018","Type":"Antminer L3++"}{"STATS":0,"ID":"L30","Elapsed":11560,"Calls":0,"Wait":0.000000,"Max":0.000000,"Min":99999999.000000,"GHS 5s":"612.27","GHS av":589.15,"miner_count":4,"frequency":"456","fan_num":2,"fan1":3690,"fan2":3750,"temp_num":4,"temp1":61,"temp2":59,"temp3":58,"temp4":55,"temp2_1":70,"temp2_2":68,"temp2_3":66,"temp2_4":62,"temp31":0,"temp32":0,"temp33":0,"temp34":0,"temp4_1":0,"temp4_2":0,"temp4_3":0,"temp4_4":0,"temp_max":61,"Device Hardware%":0.0000,"no_matching_work":3972,"chain_acn1":72,"chain_acn2":72,"chain_acn3":72,"chain_acn4":72,"chain_acs1":" oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo","chain_acs2":" oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo","chain_acs3":" oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo","chain_acs4":" oooooooo oooooooo oooooooo oooooooo oooxoooo oooooooo oooooooo oooooooo oooooooo","chain_hw1":866,"chain_hw2":28,"chain_hw3":108,"chain_hw4":2970,"chain_rate1":"152.65","chain_rate2":"146.27","chain_rate3":"163.72","chain_rate4":"149.63"}],"id":1} jhgosselin@Master-VRTJMB:~$
jr. member
Activity: 71
Merit: 1
June 17, 2018, 11:13:10 AM
#56
Interesting, the json output you posted gets parsed ok here and obviously it complains about an error at char 1309, though the string you posted is only 1308 chars long.

It sounds like the json output from cgminer has some extra whitespace or linefeed but then again the json module should be able to work with that just fine.

I am a bit lost here, could you please pm me or post the complete output of the following command(s):
Code:
sudo apt-get install netcat-openbsd
echo "{\"command\":\"stats\"}"| nc 10.0.100.141 4028

The above installs the networking swiss army knife 'netcat' and uses netcat to fetch the json output directly from cgminer api port of your miner so that I can see what could failing here when reading it from my script.

will do as soon as its done running on the L3+ Smiley
How long does it run its test before stopping ?
legendary
Activity: 1612
Merit: 1608
精神分析的爸
June 17, 2018, 10:47:43 AM
#55
Interesting, the json output you posted gets parsed ok here and obviously it complains about an error at char 1309, though the string you posted is only 1308 chars long.

It sounds like the json output from cgminer has some extra whitespace or linefeed but then again the json module should be able to work with that just fine.

I am a bit lost here, could you please pm me or post the complete output of the following command(s):
Code:
sudo apt-get install netcat-openbsd
echo "{\"command\":\"stats\"}"| nc 10.0.100.141 4028

The above installs the networking swiss army knife 'netcat' and uses netcat to fetch the json output directly from cgminer api port of your miner so that I can see what could failing here when reading it from my script.
jr. member
Activity: 71
Merit: 1
June 17, 2018, 09:57:12 AM
#54
jhgosselin@Master-VRTJMB:~/bitmain-tools/scripts$ ./l3plus_autotune.py -i 10.0.100.141
/usr/lib/python2.7/dist-packages/Crypto/Cipher/blockalgo.py:141: FutureWarning: CTR mode needs counter parameter, not IV
  self._cipher = factory.new(key, *args, **kwargs)
Failed to decode json reply:
Extra data: line 1 column 1308 - line 1 column 1309 (char 1307 - 1308)

{"STATUS":[{"STATUS":"S","When":1529247355,"Code":70,"Msg":"CGMiner stats","Description":"cgminer 4.9.0"}],"STATS":[{"CGMiner":"4.9.0","Miner":"1.0.1.3","CompileTime":"Thu Apr 12 15:55:21 CST 2018","Type":"Antminer L3++"},{"STATS":0,"ID":"L30","Elapsed":4401,"Calls":0,"Wait":0.000000,"Max":0.000000,"Min":99999999.000000,"GHS 5s":"600.56","GHS av":589.55,"miner_count":4,"frequency":"456","fan_num":2,"fan1":3690,"fan2":3750,"temp_num":4,"temp1":58,"temp2":57,"temp3":55,"temp4":52,"temp2_1":67,"temp2_2":65,"temp2_3":63,"temp2_4":60,"temp31":0,"temp32":0,"temp33":0,"temp34":0,"temp4_1":0,"temp4_2":0,"temp4_3":0,"temp4_4":0,"temp_max":58,"Device Hardware%":0.0000,"no_matching_work":1457,"chain_acn1":72,"chain_acn2":72,"chain_acn3":72,"chain_acn4":72,"chain_acs1":" oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo","chain_acs2":" oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo","chain_acs3":" oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo oooooooo","chain_acs4":" oooooooo oooooooo oooooooo oooooooo oooxoooo oooooooo oooooooo oooooooo oooooooo","chain_hw1":313,"chain_hw2":10,"chain_hw3":36,"chain_hw4":1098,"chain_rate1":"155.34","chain_rate2":"157.02","chain_rate3":"141.25","chain_rate4":"146.95"}],"id":1}
jhgosselin@Master-VRTJMB:~/bitmain-tools/scripts$
jr. member
Activity: 71
Merit: 1
June 17, 2018, 09:45:04 AM
#53
I can give you access to my L3++
just pm me

I feel honoured by your trust, but you really should never let others access any of your gear (even if they have more comfy sounding pseudonyms than 'psycodad'  Grin )
We'll get you there soon, have some patience and enjoy your first steps into the linux platform.

Oh well, another new cgminer output with json glitches, this time it's the L3++.
Don't know why every version of cgminer I have met has invalid json output that has to be fixed first.

Anyway, thanks to your pasted output I could (hopefully) fix that as I have no L3++ to test against.

Please change into your bitmain-tools directory and do:

Code:
git pull
cd scripts
./l3plus_autotune.py

Let's see if that gets us any further.

jhgosselin@Master-VRTJMB:~/bitmain-tools/scripts$ git pull
error: cannot open .git/FETCH_HEAD: Permission denied

jhgosselin@Master-VRTJMB:~/bitmain-tools/scripts$

Ah, should have seen this in your output in a previous post: You did checkout as user root and your user is now denied to modify these files, no problem just use sudo:
Code:
sudo git pull
cd scripts
./l3plus_autotune.py




./l3plus_autotune.py
no ip no -i ?
legendary
Activity: 1612
Merit: 1608
精神分析的爸
June 17, 2018, 09:41:23 AM
#52
I can give you access to my L3++
just pm me

I feel honoured by your trust, but you really should never let others access any of your gear (even if they have more comfy sounding pseudonyms than 'psycodad'  Grin )
We'll get you there soon, have some patience and enjoy your first steps into the linux platform.

Oh well, another new cgminer output with json glitches, this time it's the L3++.
Don't know why every version of cgminer I have met has invalid json output that has to be fixed first.

Anyway, thanks to your pasted output I could (hopefully) fix that as I have no L3++ to test against.

Please change into your bitmain-tools directory and do:

Code:
git pull
cd scripts
./l3plus_autotune.py

Let's see if that gets us any further.

jhgosselin@Master-VRTJMB:~/bitmain-tools/scripts$ git pull
error: cannot open .git/FETCH_HEAD: Permission denied

jhgosselin@Master-VRTJMB:~/bitmain-tools/scripts$

Ah, should have seen this in your output in a previous post: You did checkout as user root and your user is now denied to modify these files, no problem just use sudo:
Code:
sudo git pull
cd scripts
./l3plus_autotune.py


jr. member
Activity: 71
Merit: 1
June 17, 2018, 09:32:31 AM
#51
Oh well, another new cgminer output with json glitches, this time it's the L3++.
Don't know why every version of cgminer I have met has invalid json output that has to be fixed first.

Anyway, thanks to your pasted output I could (hopefully) fix that as I have no L3++ to test against.

Please change into your bitmain-tools directory and do:

Code:
git pull
cd scripts
./l3plus_autotune.py

Let's see if that gets us any further.

jhgosselin@Master-VRTJMB:~/bitmain-tools/scripts$ git pull
error: cannot open .git/FETCH_HEAD: Permission denied

jhgosselin@Master-VRTJMB:~/bitmain-tools/scripts$
Pages:
Jump to: