[GUIDE] BitFury Miner Support/Tuning

[Gyrsur]

why do my hashrate diversify so much with "autotune off" and "55" for all chips. between 520 and 440.

You probably have some H-boards with overloaded VRMs. Check /run/shm/.stat.log every 5 minutes and see if the hashrate of some chips are changing significantly. If so, turn down the speeds on all the chips on that H-board until that H-board's VRM can handle it.

thank you very much! will try to do so. does this mean the "Voltage Regulator Module" on some boards is not able to supply the power which is needed for the chips to work with this clock settings?

[1714211588]

1714211588

[goxed]

Question to PCB designers:

I submitted a Bitfury h board for fab and got the following message

Code:

1. It appears that your design has 9-12 mil holes, we require a minimum of 15mil drill with drill+10 pads for vias and drill+15 pads for mounting holes.
2. We found over 5 thousand isntances of 5.8-5.9 mil spacing, we require a minimum of 6mil spacing and no less than 6.

Can someone comment if these changes can be easily done in Eagle?

[Keefe]

why do my hashrate diversify so much with "autotune off" and "55" for all chips. between 520 and 440.

You probably have some H-boards with overloaded VRMs. Check /run/shm/.stat.log every 5 minutes and see if the hashrate of some chips are changing significantly. If so, turn down the speeds on all the chips on that H-board until that H-board's VRM can handle it.

thank you very much! will try to do so. does this mean the "Voltage Regulator Module" on some boards is not able to supply the power which is needed for the chips to work with this clock settings?

The V3 rigs were shipped with the H-board trimpots set to a higher voltage than the original design, so when overclocking you'll usually overload the H-boards' VRMs before getting many nonce errors.

[IchibahnSLC]

So the v2.3 m-boards work with the v1.2 h-cards?  I'm just about to design these in PADs.

[ricolein]

Hi,

i havent since days no access to support chat at http://ircnet.eversible.com/. Does anybody knows why?

Ciao

Enrico

[Keefe]

So the v2.3 m-boards work with the v1.2 h-cards?  I'm just about to design these in PADs.

Yes, I believe so.

[Keefe]

Hi,

i havent since days no access to support chat at http://ircnet.eversible.com/. Does anybody knows why?

Ciao

Enrico

Try connecting with an irc client to ircnet.eversible.com and join #bfsbsupport

[spiccioli]

hmmm .. somehow after plugin back one of my old h-boards ... one just stopped getting recoginzed.

I tried to use different slots without success and remarked that having other boards in subsequent slots in the same bank makes them also unrecognized.

I have no idear what happend I can't find any traces of something bad ... does any one can help or has an idear?

https://dl.dropboxusercontent.com/u/43738391/1.JPG

https://dl.dropboxusercontent.com/u/43738391/2.JPG


What can I check? Anyone?

I have something similar,

a new board, one without capacitors, all of a sudden is not recognized anymore.

I've moved it into nearly every slot, it is not recognized.

No burn signs, at a visual inspection board seems like all the others.

Any idea?

Thanks.

spiccioli

[-Redacted-]

Ditto.   I RMA'ed mine.  Nothing obviously different about this board than any others.

[buzzdave]

Hi guys - since the latest round of H-cards was shipped at .865v, rigs pull significantly higher wattage.  I'm seeing signs of connector burn & overloaded power supplies with some customers.  Please be safe - if you are overclocking or running a full rig, a 1200 watt supply is probably a better choice. 

Keep your rigs away from flammables - remember these are high-wattage devices!

I've been a little out of touch, sorry if this comes late - we don't have these problems at the data center.  We use a 10g high strand count copper wire between 12v supply and the metal lugs on the M-board.  The length of our power cables is only about 12 inches.

Best,
Dave

[Micky25]

maybe this fits  http://www.youtube.com/watch?v=xNnAvTTaJjM

[goxed]

Hey guys, did you ever get a board which will not hash no matter what settings you try? I have obtained one such board. All it throws are loads of errors. I have tested the usual suspects voltage, etc.
It's a new PCIexpress style board, so I cannot bypass a chip.

I have tried disabling chips in software but the board will not hash. I tried it as a sole board connected to an Mboard.

[tom99]

I did change dead from 0 to 56 and get better hashrate.

speed:5272 noncerate[GH/s]:190.339 (1.983/chip) hashrate[GH/s]:201.838 good:13295 errors:556 spi-errors:2 miso-errors:0 jobs:293 (record[GH/s]:0.000)
0:   878   30.938   33.907   2161   173   0   0
1:   879   32.227   34.161   2251   76   0   0
2:   877   33.257   34.225   2323   28   1   0
3:   878   32.627   34.605   2279   129   0   0
4:   879   32.914   34.372   2299   77   0   0
5:   881   28.375   30.567   1982   73   1   0


last one from Aug. starter kit no OC.

[CA Coins]

Hi guys - since the latest round of H-cards was shipped at .865v, rigs pull significantly higher wattage.  I'm seeing signs of connector burn & overloaded power supplies with some customers.  Please be safe - if you are overclocking or running a full rig, a 1200 watt supply is probably a better choice. 

Keep your rigs away from flammables - remember these are high-wattage devices!

I've been a little out of touch, sorry if this comes late - we don't have these problems at the data center.  We use a 10g high strand count copper wire between 12v supply and the metal lugs on the M-board.  The length of our power cables is only about 12 inches.

Best,
Dave

Just to clarify after reading the fire hazard thread.  I am using 2 separate pci express power connectors to power a full rig right now.  Should I also power the 12V/GND power connectors on the other end of the M-board?  The rigs have been running stable for the past month.

V3M/V2.2H/Corsair 870 platinum, stock settings, about 650W at the wall

[tom99]

Hi guys - since the latest round of H-cards was shipped at .865v, rigs pull significantly higher wattage.  I'm seeing signs of connector burn & overloaded power supplies with some customers.  Please be safe - if you are overclocking or running a full rig, a 1200 watt supply is probably a better choice. 

Keep your rigs away from flammables - remember these are high-wattage devices!

I've been a little out of touch, sorry if this comes late - we don't have these problems at the data center.  We use a 10g high strand count copper wire between 12v supply and the metal lugs on the M-board.  The length of our power cables is only about 12 inches.

Best,
Dave

Just to clarify after reading the fire hazard thread.  I am using 2 separate pci express power connectors to power a full rig right now.  Should I also power the 12V/GND power connectors on the other end of the M-board?  The rigs have been running stable for the past month.

V3M/V2.2H/Corsair 870 platinum, stock settings, about 650W at the wall

  I think 2 cables might over load it but if me add 2 more cables to reduce heat up on pci-e cable.

[CA Coins]

I assume it's ok to run power to both the pci express power connectors and the 12V/GND screw terminals?

[Keefe]

I assume it's ok to run power to both the pci express power connectors and the 12V/GND screw terminals?

Yes, that's what I do. I bought some splitters on ebay, cut them up, and added ring terminals, like what came with the v1 M-boards. I don't like hot connectors.

[tom99]

  anyone mine ppc?  I am trying to mined ppc get alot of errors

speed:5272 noncerate[GH/s]:109.293 (1.138/chip) hashrate[GH/s]:29.563 good:7634 errors:3164 spi-errors:0 miso-errors:34 jobs:57 (record[GH/s]:0.000)
0:   878   17.194   4.947   1201   648   0   7
1:   878   18.898   4.957   1320   521   0   11
2:   877   19.370   5.052   1353   502   0   5
3:   878   17.495   4.883   1222   530   0   5
4:   880   19.127   5.031   1336   476   0   3
5:   881   17.209   4.693   1202   487   0   3

work fined for btc.

[spegelius]

  anyone mine ppc?  I am trying to mined ppc get alot of errors

speed:5272 noncerate[GH/s]:109.293 (1.138/chip) hashrate[GH/s]:29.563 good:7634 errors:3164 spi-errors:0 miso-errors:34 jobs:57 (record[GH/s]:0.000)
0:   878   17.194   4.947   1201   648   0   7
1:   878   18.898   4.957   1320   521   0   11
2:   877   19.370   5.052   1353   502   0   5
3:   878   17.495   4.883   1222   530   0   5
4:   880   19.127   5.031   1336   476   0   3
5:   881   17.209   4.693   1202   487   0   3

work fined for btc.

I mine PPC occasionally, i have automated the switching of coins based on profitability info from web. Pool i use is https://bitcointalk.org/index.php?topic=173272.0 which works fine.

With bitfury it seems that what pool you are using means alot; with some pools, you need to remove the -rt from stratum proxy's command line. Also when selecting the pool, it's a good idea to select pools where you can manually set the desired difficulty. With pools that select difficulty automatically i've seen that sometimes the difficulty drops too low and then .putstat shows a lot of requests queued, and performance tanks...

[tom99]

  anyone mine ppc?  I am trying to mined ppc get alot of errors

speed:5272 noncerate[GH/s]:109.293 (1.138/chip) hashrate[GH/s]:29.563 good:7634 errors:3164 spi-errors:0 miso-errors:34 jobs:57 (record[GH/s]:0.000)
0:   878   17.194   4.947   1201   648   0   7
1:   878   18.898   4.957   1320   521   0   11
2:   877   19.370   5.052   1353   502   0   5
3:   878   17.495   4.883   1222   530   0   5
4:   880   19.127   5.031   1336   476   0   3
5:   881   17.209   4.693   1202   487   0   3

work fined for btc.

I mine PPC occasionally, i have automated the switching of coins based on profitability info from web. Pool i use is https://bitcointalk.org/index.php?topic=173272.0 which works fine.

With bitfury it seems that what pool you are using means alot; with some pools, you need to remove the -rt from stratum proxy's command line. Also when selecting the pool, it's a good idea to select pools where you can manually set the desired difficulty. With pools that select difficulty automatically i've seen that sometimes the difficulty drops too low and then .putstat shows a lot of requests queued, and performance tanks...

I used ppc D7 pool and I thank you.  I am going to try it later on.