Bitcoin Forum
April 27, 2024, 12:17:29 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 [23] 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 »
  Print  
Author Topic: [GUIDE] BitFury Miner Support/Tuning  (Read 147973 times)
This is a self-moderated topic. If you do not want to be moderated by the person who started this topic, create a new topic.
KNK
Hero Member
*****
Offline Offline

Activity: 692
Merit: 502


View Profile
October 29, 2013, 05:06:50 PM
 #441

1) just pull out the pcie ... one by one on the miner? dangerous? also is reinserting to have it on again ... a problem when connecting one pcie after the other when psu is on?
Shouldn't be a problem at all for the few seconds it will run with a single PCIe even if it keeps hashing, but you can stop the miner and power-off the RPi first (recommended in order not to damage the file system)

Mega Crypto Polis - www.MegaCryptoPolis.com
BTC tips: 1KNK1akhpethhtcyhKTF2d3PWTQDUWUzHE
1714220249
Hero Member
*
Offline Offline

Posts: 1714220249

View Profile Personal Message (Offline)

Ignore
1714220249
Reply with quote  #2

1714220249
Report to moderator
1714220249
Hero Member
*
Offline Offline

Posts: 1714220249

View Profile Personal Message (Offline)

Ignore
1714220249
Reply with quote  #2

1714220249
Report to moderator
The forum was founded in 2009 by Satoshi and Sirius. It replaced a SourceForge forum.
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
1714220249
Hero Member
*
Offline Offline

Posts: 1714220249

View Profile Personal Message (Offline)

Ignore
1714220249
Reply with quote  #2

1714220249
Report to moderator
1714220249
Hero Member
*
Offline Offline

Posts: 1714220249

View Profile Personal Message (Offline)

Ignore
1714220249
Reply with quote  #2

1714220249
Report to moderator
1714220249
Hero Member
*
Offline Offline

Posts: 1714220249

View Profile Personal Message (Offline)

Ignore
1714220249
Reply with quote  #2

1714220249
Report to moderator
karusloom
Newbie
*
Offline Offline

Activity: 27
Merit: 0


View Profile
October 29, 2013, 09:51:48 PM
 #442

have it on again ... a problem when connecting one pcie after the other when psu is on?
2) is there a switch for a pcie cable? ...

thanx or your help

As the 12+ of the cable comes from one output u could just add a kill switch to it, cut the cable and solder the switch. Any 220V 10A rocker switch would do the trick, dont forget to insulate the soldered terminals. Or you can make an patch cable out of these (link) and the switch, that way you dont have to hack you psu cables.
Keefe
Hero Member
*****
Offline Offline

Activity: 681
Merit: 500


View Profile
October 30, 2013, 06:30:44 AM
 #443

Even without overclocking, a full rig pulls something like 25A from the 12V input. You'll need a much heftier switch than that.

GandalfG
Sr. Member
****
Offline Offline

Activity: 259
Merit: 250


Dig your freedom


View Profile
October 30, 2013, 11:18:59 AM
 #444

Even without overclocking, a full rig pulls something like 25A from the 12V input. You'll need a much heftier switch than that.
Little overclocked (2k2 resistor instead 1k8)  H-Board with RPi  take 2.4-2.6A 12V
Thats means 2A 12V are minimum for one board.

Want to say thanks? 16ragydppe9QFRVhrdwEUjgfMS7KCfEFGY
-Redacted-
Hero Member
*****
Offline Offline

Activity: 574
Merit: 501


View Profile
October 30, 2013, 12:01:22 PM
 #445

have it on again ... a problem when connecting one pcie after the other when psu is on?
2) is there a switch for a pcie cable? ...

thanx or your help

As the 12+ of the cable comes from one output u could just add a kill switch to it, cut the cable and solder the switch. Any 220V 10A rocker switch would do the trick, dont forget to insulate the soldered terminals. Or you can make an patch cable out of these (link) and the switch, that way you dont have to hack you psu cables.

Rather than trying to switch all that high amperage, why not just use the green wire on the 24 pin connector the way it's meant to be used - as an on/off switch?  You don't normally turn off your computer by yanking out the PCIe power connectors from the motherboard, or by yanking the plug out of the wall, do you?  It's bad for the attached electronics to do it that way.

 http://www.robotshop.com/en/cytron-atx-power-supply-breakout-board-right-angle.html
Keefe
Hero Member
*****
Offline Offline

Activity: 681
Merit: 500


View Profile
October 30, 2013, 12:03:08 PM
 #446

have it on again ... a problem when connecting one pcie after the other when psu is on?
2) is there a switch for a pcie cable? ...

thanx or your help

As the 12+ of the cable comes from one output u could just add a kill switch to it, cut the cable and solder the switch. Any 220V 10A rocker switch would do the trick, dont forget to insulate the soldered terminals. Or you can make an patch cable out of these (link) and the switch, that way you dont have to hack you psu cables.

Rather than trying to switch all that high amperage, why not just use the green wire on the 24 pin connector the way it's meant to be used - as an on/off switch?   http://www.robotshop.com/en/cytron-atx-power-supply-breakout-board-right-angle.html


He wants to power multiple rigs from one PSU but be able to turn off just one of them.

-Redacted-
Hero Member
*****
Offline Offline

Activity: 574
Merit: 501


View Profile
October 30, 2013, 12:07:18 PM
 #447

So he's going to yank power connectors off - and risk damaging - multiple, multi-thousand dollar pieces of hardware just to try and save himself $90 on a PSU?  

He deserves whatever happens to his rigs...
darkfriend77 (OP)
Sr. Member
****
Offline Offline

Activity: 434
Merit: 265


View Profile WWW
October 30, 2013, 01:54:31 PM
 #448

So he's going to yank power connectors off - and risk damaging - multiple, multi-thousand dollar pieces of hardware just to try and save himself $90 on a PSU?  

He deserves whatever happens to his rigs...

Hmmm ... I don't know why you think I would deserve something ... but anyway ...

I'm working with 2 PSU right now until stability isn't an issue anymore and then move onto one PSU.

Thx for all the constructive comments especially yours Redacted.

Peace and happy mining.

-:| www.DOTMog.com |:-
-Redacted-
Hero Member
*****
Offline Offline

Activity: 574
Merit: 501


View Profile
October 30, 2013, 04:37:26 PM
Last edit: October 30, 2013, 06:27:17 PM by -Redacted-
 #449

Just doesn't seem to be a good idea skimping on fairly inexpensive PSU's, or pulling out power cables to turn off an expensive rig - for the same reason you wouldn't expect a motherboard to last for very long if that was how you shut your computer down - by pulling power cables out of the connectors.

PSU's have bleeder circuitry to prevent big amp surges from taking place when you shut them off.  If you yank a connector out of the socket, or just pull the plug out of the wall, there are lots of electrons that suddenly find themselves needing to go home -RIGHT NOW- and they aren't too fussy about the path they take getting there.  They're perfectly willing to make themselves a new path of least resistance, and will happily travel through - and burn up - a part that wasn't designed to handle that kind of surge, if they can.

Just sayin'...

I only mention this having learned a great deal about electron return, flyback currents, and paths-of-least resistance working in an environment that sometimes required us to switch 100Kv @ 10K Amps on and off.  (Think LHC accelerator-beam superconducting magnet type power.)  People are better conductors than air-gaps, and electrons were more than happy to oblige if you did something incredibly stupid and offered them a shorter return path than the one you expected them to take through air gaps and enormous relays and pools of mercury. 



Jumpsu
Newbie
*
Offline Offline

Activity: 29
Merit: 0


View Profile
October 30, 2013, 06:02:21 PM
 #450

I made step by step, but i cant acces, 6. How to start tuning chips manually?

What is wrong...? I get:
pi@bitfury ~ $ nano /run/shm/.stat.log
g++ -g -Wall -pthread -O2 -march=armv6 -mfloat-abi=hard -mfpu=vfp -Wall -I . -c spidevc.cpp -o spidevc.o
g++ -g -Wall -pthread -O2 -march=armv6 -mfloat-abi=hard -mfpu=vfp -Wall -I . -c miner.cpp -o miner.o
make -C json
make[1]: Entering directory `/opt/bitfury/chainminer/json'
g++ -pthread -I .. -O2 -g -c json_reader.cpp -o json_reader.o
g++ -pthread -I .. -O2 -g -c json_value.cpp -o json_value.o

Also:

pi@bitfury ~ $ nano /run/shm/.stat.log
g++ -g -Wall -pthread -O2 -march=armv6 -mfloat-abi=hard -mfpu=vfp -Wall -I . -c handylib.cpp -o handylib.o
In file included from /usr/include/c++/4.6/backward/strstream:52:0,
                 from handylib.cpp:5:
/usr/include/c++/4.6/backward/backward_warning.h:33:2: warning: #warning This file includes at least one deprecated or antiquated header which may be removed without further notice at a future date. Please use a non-deprecated interface with equivalent functionality instead. For a listing of replacement headers and interfaces, consult the file backward_warning.h. To disable this warning use -Wno-deprecated. [-Wcpp]
handylib.cpp: In member function ‘void handylib::HTTPServer::loadMultipart(std::istream&)’:
handylib.cpp:653:18: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
handylib.cpp: In member function ‘void handylib::Time::scache() const’:
handylib.cpp:2850:25: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
handylib.cpp: In function ‘void handylib::tm_loadtzi()’:
handylib.cpp:3048:13: warning: variable ‘tm’ set but not used [-Wunused-but-set-variable]

Also

pi@bitfury ~ $ nano /run/shm/.stat.log

g++ -g -Wall -pthread -O2 -march=armv6 -mfloat-abi=hard -mfpu=vfp -Wall -I . -c spidevc.cpp -o spidevc.o
g++ -g -Wall -pthread -O2 -march=armv6 -mfloat-abi=hard -mfpu=vfp -Wall -I . -c miner.cpp -o miner.o
make -C json
make[1]: Entering directory `/opt/bitfury/chainminer/json'
g++ -pthread -I .. -O2 -g -c json_reader.cpp -o json_reader.o

/rung++ -pthread -I .. -O2 -g -c json_value.cpp -o json_value.o
g++ -pthread -I .. -O2 -g -c json_writer.cpp -o json_writer.o
rm -f libjson.a
ar cruv libjson.a json_reader.o json_value.o json_writer.o
a - json_reader.o
a - json_value.o
a - json_writer.o
make[1]: Leaving directory `/opt/bitfury/chainminer/json'
g++ -g -Wall -pthread -O2 -march=armv6 -mfloat-abi=hard -mfpu=vfp -Wall -I . jobconnect.o handylib.o spidevc.o miner.o json/libjson.a -o miner
pi@bitfury /opt/bitfury/chainminer $ sudo killall screen
pi@bitfury /opt/bitfury/chainminer $ sudo /etc/rc.local
miner: no process found
pi@bitfury /opt/bitfury/chainminer $ pi@bitfury ~ $ nano /run/shm/.stat.log
-bash: pi@bitfury: command not found
pi@bitfury /opt/bitfury/chainminer $ pi@bitfury ~ $ nano /run/shm/.stat.log
-bash: pi@bitfury: command not found
pi@bitfury /opt/bitfury/chainminer $
pi@bitfury /opt/bitfury/chainminer $
pi@bitfury /opt/bitfury/chainminer $ /run
Keefe
Hero Member
*****
Offline Offline

Activity: 681
Merit: 500


View Profile
October 30, 2013, 06:36:24 PM
 #451

Why are you compiling chainminer? If all you're trying to do is change clock speeds, just edit /opt/bitfury/best.cnf and reboot. If that file doesn't exist yet, copy /run/shm/.stat.log to /opt/bitfury/best.cnf first.

Jumpsu
Newbie
*
Offline Offline

Activity: 29
Merit: 0


View Profile
October 30, 2013, 07:12:16 PM
Last edit: October 30, 2013, 07:58:08 PM by Jumpsu
 #452



Why are you compiling chainminer? If all you're trying to do is change clock speeds, just edit /opt/bitfury/best.cnf and reboot. If that file doesn't exist yet, copy /run/shm/.stat.log to /opt/bitfury/best.cnf first.

1. How i copy /run/shm/.stat.log to /opt/bitfury/best.cnf ?
2. How i get /run/shm/.stat.log ?
tom99
Sr. Member
****
Offline Offline

Activity: 420
Merit: 250


View Profile
October 30, 2013, 08:11:24 PM
 #453



Why are you compiling chainminer? If all you're trying to do is change clock speeds, just edit /opt/bitfury/best.cnf and reboot. If that file doesn't exist yet, copy /run/shm/.stat.log to /opt/bitfury/best.cnf first.

1. How i copy /run/shm/.stat.log to /opt/bitfury/best.cnf ?
2. How i get /run/shm/.stat.log ?

use cp  /run/shm/.stat.log  /opt/bitfury/best.cnf

you can read more about cp command http://www.computerhope.com/unix/ucp.htm
Keefe
Hero Member
*****
Offline Offline

Activity: 681
Merit: 500


View Profile
October 30, 2013, 08:11:56 PM
 #454



Why are you compiling chainminer? If all you're trying to do is change clock speeds, just edit /opt/bitfury/best.cnf and reboot. If that file doesn't exist yet, copy /run/shm/.stat.log to /opt/bitfury/best.cnf first.

1. How i copy /run/shm/.stat.log to /opt/bitfury/best.cnf ?
2. How i get /run/shm/.stat.log ?

1. cp /run/shm/.stat.log /opt/bitfury/best.cnf
2. If /run/shm/.stat.log doesn't exist, let the miner run for at least 5 minutes and it should be there.

Meizirkki
Hero Member
*****
Offline Offline

Activity: 616
Merit: 500



View Profile
October 30, 2013, 08:33:12 PM
 #455

Even without overclocking, a full rig pulls something like 25A from the 12V input. You'll need a much heftier switch than that.
10A 250V switch should do fine. Low voltage DC is much easier on the switch.

I've used a 10A 250V switch on my electric kick scooter for years. Peak current >100A and it's still working fine after probably 2000 on/off cycles.
goxed
Legendary
*
Offline Offline

Activity: 1946
Merit: 1006


Bitcoin / Crypto mining Hardware.


View Profile
October 31, 2013, 01:50:32 AM
 #456

I've some questions ... maybee some one has a good solution ...

I'm using a multi rail atx psu ... 1200 Watt ... at which 2 miners are connected.

So if I want to do some overclocking at the moment I've to shut down both miners ...

What solution do I have to only shut down the miner Im modding ....

1) just pull out the pcie ... one by one on the miner? dangerous? also is reinserting to have it on again ... a problem when connecting one pcie after the other when psu is on?
2) is there a switch for a pcie cable? ...
3) a second psu ... yep i got one but i would be happy to only use ... one ...

thanx or your help
It may trigger the TVS diode on your raspberryPi and destroy it. I would suggest turning off the PSU instead of yanking out PCIE power cables.

Revewing Bitcoin / Crypto mining Hardware.
CMaxx22
Newbie
*
Offline Offline

Activity: 56
Merit: 0


View Profile
November 01, 2013, 07:46:56 AM
Last edit: November 01, 2013, 08:11:32 AM by CMaxx22
 #457

Another question:
What means if one chip shows aifdso 240?
Like the first one
http://s020.radikal.ru/i716/1311/ea/46685811bb64.jpg

And it seems that autotune is not working, on the screenshot I manually worked with 2,5,12 chips. The first one is setted up to AIfDSo 55 but automatically falls down to aifdso 240. The second chip was turned off by me because of very low hashrate (200Mhs).
Can you tell something about the first chip? And maybe you know what should I do for 2,5,12 chips to increase their hashrate?
Thank you!

ps: overclocking to 0.95v and oil cooling change nothing...
-Redacted-
Hero Member
*****
Offline Offline

Activity: 574
Merit: 501


View Profile
November 01, 2013, 01:04:34 PM
 #458

A lowercase 'a' starting the string means you have turned off autotune for that chips and manually set the values...
CMaxx22
Newbie
*
Offline Offline

Activity: 56
Merit: 0


View Profile
November 01, 2013, 01:13:31 PM
 #459

But why the other chips with capital "A" is not changed their frequency from 55? Day after day frequency is not changing...
Keefe
Hero Member
*****
Offline Offline

Activity: 681
Merit: 500


View Profile
November 01, 2013, 01:21:48 PM
 #460

The column after the [4:x] shows how many of the 756 cores have not produced good results. It looks like chips 5 and 12 are half (or more) dead, so you'll need to leave autotune off and just accept that the noncerate on those will be less than half their hashrate. You can try different speeds to see what produces the best noncerate. You'll probably want to clock them as high as the rest, and just expect 50%+ error rate. Chainminer autotune can't handle chips with alot of dead cores and assumes they need to be downclocked or shut off. After running for long enough to get most of the bad-core counts near 0, look at /tmp/.core.log and see what 5 and 12 look like. It'll probably show that one side or the other is dead. Here's one of my half-dead chips:
Code:
pi@bitfury ~ $ grep ^41 /tmp/.core.log
41        35  36  30  22  31  31  35  35  33  39  21  31  34   0   0   0   0   0   0   0   0
41        31  35  36  38  29  18  42  27  28  26  25  31  25   0   0   0   0   0   0   0   0
41        34  34  36  34  25  28  27  29  26  31  28  26  33   0   0   0   0   0   0   0   0
41        32  31  29  40  33  29  28  26  35  40  31  35  42   0   0   0   0   0   0   0   0
41        19  23  27  29  40  31  40  18  32  22  35  30  38   0   0   0   0   0   0   0   0
41        31  18  26  34  21  25  31  38  39  24  25  34  39   0   0   0   0   0   0   0   0
41        30  24  35  34  32  19  33  26  36  38  30  34  34   0   0   0   0   0   0   0   0
41        43  28  26  36  24  33  36  26  33  44  23  29  40   0   0   0   0   0   0   0   0
41        31  37  32  24  28  31  39  31  22  42  28  37  30   0   0   0   0   0   0   0   0
41        36  20  25  35  28  29  30  25  27  25  35  29  33   0   0   0   0   0   0   0   0
41        28  25  36  38  24  33  32  27  36  25  28  34  34   0   0   0   0   0   0   0   0
41        37  41  36  40  37  24  27  20  39  32  30  34  26   0   0   0   0   0   0   0   0
41        32  26  34  28  32  32  26  33  31  36  35  31  23   0   0   0   0   0   0   0   0
41        27  31  33  26  24  37  30  32  29  28  26  32  31   0   0   0   0   0   0   0   0
41        30  23  25  30  25  23  30  24  30  19  26  39  30   0   0   0   0   0   0   0   0
41        29  38  39  33  30  32  32  37  27  25  41  29  32   0   0   0   0   0   0   0   0
41        29  31  33  30  30  32  39  37  25  35  19  30  51   0   0   0   0   0   0   0   0
41        25  20  22  33  29  39  31  29  30  36  36  34  38   0   0   0   0   0   0   0   0
41        19  30  33  37  23  26  29  37  34  28  35  27  24   0   0   0   0   0   0   0   0
41        41  23  40  27  32  33  28  36  26  28  45  32  30   0   0   0   0   0   0   0   0
41        33  36  29  34  23  36  31  25  27  25  30  37  20   0   0   0   0   0   0   0   0
41        24  33  26  38  23  29  23  30  28  41  36  35  23   0   0   0   0   0   0   0   0
41        35  31  34  34  28  29  38  18  27  33  31  37  37   0   0   0   0   0   0   0   0
41        36  41  30  20  30  35  45  29  37  38  25  26  44   0   0   0   0   0   0   0   0
41        43  33  33  27  41  19  27  33  25  30  23  36  31   0   0   0   0   0   0   0   0
41        26  26  32  30  30  25  39  33  32  32  35  29  37   0   0   0   0   0   0   0   0
41        45  29  34  38  36  25  29  37  34  37  36  26  38   0   0   0   0   0   0   0   0
41        30  43  34  25  35  29  28  30  33  27  27  39  28   0   0   0   0   0   0   0   0
41        32  27  39  36  38  28  34  44  31  32  37  30  35   0   0   0   0   0   0   0   0
41        28  30  37  35  39  16  39  30  25  27  30  46  28   0   0   0   0   0   0   0   0
41        35  27  28  30  39  35  19  23  33  39  35  43  30   0   0   0   0   0   0   0   0
41        30  31  29  34  28  37  33  22  34  43  37  32  45   0   0   0   0   0   0   0   0
41        30  27  39  23  25  24  39  38  28  28  28  29  30   0   0   0   0   0   0   0   0
41        27  28  26  29  37   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0
41        27  25  21  32  31   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0
41        35  31  40  35  22   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0
41      312

Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 [23] 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 »
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!