[GUIDE] BitFury Miner Support/Tuning

[tom99]

Add me to the mix on those seeing incompatibility mixing caps and capless. I've done no modding.

I just added three new cards to my original. With the original in the first slot, it slowly died and dropped to 0, and all the cards seemed to get pretty hot. I moved it to the last slot, and everything seems to be running good (about 3 hours stable now). Although the original card is slower, it was always around that speed, so I don't think it is being negatively affected by the capless cards.

   I think you will need fan to cool down chips or add heatsink.

[nowywbitcoinowymswiecie]

You'll have to look at it for a complete line:
2   AIfDSo   57   1.718   2.040   120   1   0   0   193   [0:1]   0   7 7 7 7 7 7 8 8 8 8 8 8 7 8 8 7    0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
It's showing how the cores of a chip(7 7 7 7 7 7 8 8 8 8 8 8 7 8 8 7) are doing in relation to the sum(120).
120 = Nonces/round
1 = False nonce
(For this particular chip)
They are read only fields.

Thanks! Finally I know what it is ;-)

[dacman61]

I NEED AN IMAGE FOR THE SD-CARD FOR V1 MBOARD FROM AUGUST.

Every link on the forums says "account has been disabled due to high traffic". This applies to all "drop box" links. I NEED AN IMAGE FOR V1 MBOARD SDCARD, BOTH MINE DIED.


Please and thank you...

Here is the image I used for my Raspberry Pi from my August kit with the V1 M-board. This file was originally downloaded on 9-27-2013. I used Win32DiskImager v0.9 to do the imaging of the SD card.

I used 7-Zip to compress it to save space and time for downloading. It will be available at this link for 1-2 weeks from the time of this post before I take it down.

https://drive.google.com/file/d/0B-oH77oD9QLNa3FnTEVlYm9MOUU/edit?usp=sharing

[irritant]

everything was running smooth for a couple days, autotune was off, suddenly one h-board (October delivery) dies    Tried putting it in different slots, didn't help, nothing seems to look wrong with it. I've pulled it out now, should I put it back in and wait for self healing effect?

Code:

233     aIfDSo  54      0.000   0.634   0       243     12      0       26     $
234     aIfDSo  54      0.000   0.829   0       127     15      2       34     $
235     aIfDSo  54      0.000   0.756   0       168     15      0       31     $
236     aIfDSo  54      0.000   1.439   0       362     32      2       59     $
237     aIfDSo  54      0.000   1.293   0       387     28      2       53     $
238     aIfDSo  54      0.000   1.268   0       370     28      2       52     $
239     aIfDSo  54      0.000   1.268   0       422     27      3       52     $
240     aIfDSo  54      0.000   1.342   0       378     30      3       55     $
241     aIfDSo  54      0.000   1.049   0       373     23      3       43     $
242     aIfDSo  54      0.000   1.244   0       384     29      0       51     $
243     aIfDSo  54      0.000   1.415   0       392     32      1       58     $
244     aIfDSo  54      0.000   1.244   0       392     28      0       51     $
245     aIfDSo  54      0.000   1.244   0       375     28      3       51     $
246     aIfDSo  54      0.000   1.342   0       403     30      1       55     $
247     aIfDSo  54      0.000   1.024   0       400     21      2       42     $
248     aIfDSo  54      0.000   1.171   0       382     28      1       48     $

[BenTuras]

everything was running smooth for a couple days, autotune was off, suddenly one h-board (October delivery) dies    Tried putting it in different slots, didn't help, nothing seems to look wrong with it. I've pulled it out now, should I put it back in and wait for self healing effect?

Turn on auto tune again for this board?
Try pencil mod if not done yet?

The Bitfury boards are a challenge, I haven't figured out a placing/tuning algorithm (based on any information available) to get best hashrate.

[Keefe]

Here's a script I'm running (in screen) to display per-board stats every time chainminer updates the file (every 5 minutes), and copy the full stats to a log dir:

Code:

pi@bitfury ~ $ cd /run/shm ; if [ ! -e logs ] ; then mkdir logs ; fi ; laststatmod= ; while [ 0 ] ; do statmod=`stat -c %Y .stat.log` ; if [ "$statmod" != "$laststatmod" ] ; then laststatmod=$statmod ; echo ; stat -c %y .stat.log ; grep --color=never -m 1 -A 16 noncerate .stat.log ; cp .stat.log logs/$statmod ; fi ; sleep 5 ; done

2013-10-27 09:34:41.136382762 +0000
speed:8061 noncerate[GH/s]:321.507 (2.233/chip) hashrate[GH/s]:325.651 good:22457 errors:273 spi-errors:5 miso-errors:0 jobs:323 (record[GH/s]:327.649)
0:      900     34.646  36.381  2420    18      0       0
1:      897     35.591  36.064  2486    9       0       0
2:      900     35.620  35.535  2488    51      0       0
3:      901     35.849  35.747  2504    12      1       0
4:      910     35.978  36.169  2513    45      1       0
5:      894     36.035  36.085  2517    14      0       0
6:      891     36.035  36.888  2517    13      0       0
7:      873     34.947  35.842  2441    79      0       0
8:      895     36.808  36.941  2571    32      3       0

2013-10-27 09:39:41.054781143 +0000
speed:8067 noncerate[GH/s]:318.558 (2.212/chip) hashrate[GH/s]:326.380 good:22251 errors:307 spi-errors:4 miso-errors:0 jobs:322 (record[GH/s]:327.649)
0:      899     36.020  36.455  2516    14      0       0
1:      899     35.691  36.391  2493    25      0       0
2:      902     34.331  35.969  2398    138     0       0
3:      904     36.035  36.148  2517    11      0       0
4:      909     35.663  35.757  2491    8       0       0
5:      896     34.417  36.085  2404    12      1       0
6:      891     35.047  36.930  2448    16      2       0
7:      873     33.730  35.673  2356    63      1       0
8:      894     37.624  36.973  2628    20      0       0

Watch out for running out of ramdisk space. It will fill up in less than 2 weeks with a full rig. I'll need to think of a way to send the files to another location for storage if I want, or just prune the old data occasionally. I'm doing this logging so I can notice if performance changes over time and which chips are the culprit.

[tom99]

everything was running smooth for a couple days, autotune was off, suddenly one h-board (October delivery) dies    Tried putting it in different slots, didn't help, nothing seems to look wrong with it. I've pulled it out now, should I put it back in and wait for self healing effect?

   I got 2 Hboards same as your hboard died with 30 minutes and board got very hot melt down solder around chips.  I got 3 newer hboards but need AC cool down it or hboards will shut off every 10 minutes without AC on it and too later for other 2 died by overheat.
  pictures of chips http://picts.imgur.com/all/

[jddebug]

everything was running smooth for a couple days, autotune was off, suddenly one h-board (October delivery) dies    Tried putting it in different slots, didn't help, nothing seems to look wrong with it. I've pulled it out now, should I put it back in and wait for self healing effect?

   I got 2 Hboards same as your hboard died with 30 minutes and board got very hot melt down solder around chips.  I got 3 newer hboards but need AC cool down it or hboards will shut off every 10 minutes without AC on it and too later for other 2 died by overheat.
  pictures of chips http://picts.imgur.com/all/

You suck at posting links... Try again.

The link works for me dacman61.  I'm not sure how it works but it comes up with 15 pictures of a burnt chip.

[CMaxx22]

Hi! Can anybody explain me how to connect to bitfury's stratum proxy (located on raspberry pi) with my erupter blades? They are plugged into the same switch port as bitfury starter kit. Is it possible to use the starter kit's stratum for blades? Thank you!

[spiccioli]

Hi! Can anybody explain me how to connect to bitfury's stratum proxy (located on raspberry pi) with my erupter blades? They are plugged into the same switch port as bitfury starter kit. Is it possible to use the starter kit's stratum for blades? Thank you!

On your bitfury kit stratum-proxy listens on three ports, one per each configured pool on its webgui.

8332,8333,8334

It should be enough to configure your blade to point to your bitfury kit's IP address:8332 to have your blade configured to mine on your first configured pool.

Mind you, I don't have blades available to test.

spiccioli

[CMaxx22]

Thank you! It was enough. Very helpful!

[spiccioli]

Thank you! It was enough. Very helpful!

You're welcome.

spiccioli

[darkfriend77]

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

[KNK]

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)

[karusloom]

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]

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

[GandalfG]

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.

[-Redacted-]

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]

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-]

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...