[GUIDE] BitFury Miner Support/Tuning

[Swimmer63]

I just received two new h-boards, V1.2 as I have a V1 M-board.
Both cards are running at 33-34 Gh/s out of the box.
I measured the R02F resistor and both boards read 2.1KOhm.
My other boards read 1.3KOhm.
Can I still attempt to mod the two new boards?  If so, I don't want to move them much because they are already running high.  But do have very good cooling, 4 fans on 6 cards.
Thoughts?  Input from the wise men?

[Meizirkki]

Lol, i am electronics technician, so i repair/change Ti-chip,I do not want to break the warranty...

you will lose the warranty. But so long as you know what you're doing you wont need it. *high five* I'm an electronics technician too.

Can I still attempt to mod the two new boards?  If so, I don't want to move them much because they are already running high.  But do have very good cooling, 4 fans on 6 cards.
Thoughts?  Input from the wise men?

Iirc some where reporting problems when mixing old pencil modded cards with modded new ones. I'd say do it you can always clean the grphite off if problems arise.

[Isokivi]

I just received two new h-boards, V1.2 as I have a V1 M-board.
Both cards are running at 33-34 Gh/s out of the box.
I measured the R02F resistor and both boards read 2.1KOhm.
My other boards read 1.3KOhm.
Can I still attempt to mod the two new boards?  If so, I don't want to move them much because they are already running high.  But do have very good cooling, 4 fans on 6 cards.
Thoughts?  Input from the wise men?

Yes, you can pencil mod them and they respond as well as the previous boards, do not go as low tough. I've found that any board can take 1.72 and ones with very good cooling (3 fans per bank and heatsinks) can go as low as 1.63KOhm. I'd try to lower the resistance by 0.05KOhm at a step and see how the cards respond.

[Swimmer63]

I just received two new h-boards, V1.2 as I have a V1 M-board.
Both cards are running at 33-34 Gh/s out of the box.
I measured the R02F resistor and both boards read 2.1KOhm.
My other boards read 1.3KOhm.
Can I still attempt to mod the two new boards?  If so, I don't want to move them much because they are already running high.  But do have very good cooling, 4 fans on 6 cards.
Thoughts?  Input from the wise men?

Yes, you can pencil mod them and they respond as well as the previous boards, do not go as low tough. I've found that any board can take 1.72 and ones with very good cooling (3 fans per bank and heatsinks) can go as low as 1.63KOhm. I'd try to lower the resistance by 0.05KOhm at a step and see how the cards respond.

Makes sense.  Thanks guys.

[GandalfG]

For all those who measure the voltage applied to chip on H-Boards, as a minus (GND) use for this GND on the H-Board instead power suply or M-boards. These differences arising from the resistance of sockets and contacts on a card. They range tenths of volts. It turns out that, instead of 0.865V is 0.825 or even 0.760
Therefore, the chip receives voltage measured preferably on next chip capacitors or capacitors next to coils (PULSE) of DC-DC converter labeled C12 to C16.

https://www.dropbox.com/s/rxhufkcy1tgtwx2/20131110_220257.jpg

[Keefe]

For all those who measure the voltage applied to chip on H-Boards, as a minus (GND) use for this GND on the H-Board instead power suply or M-boards. These differences arising from the resistance of sockets and contacts on a card. They range tenths of volts. It turns out that, instead of 0.865V is 0.825 or even 0.760
Therefore, the chip receives voltage measured preferably on next chip capacitors or capacitors next to coils (PULSE) of DC-DC converter labeled C12 to C16.

https://www.dropbox.com/s/rxhufkcy1tgtwx2/20131110_220257.jpg

Check voltage on top of the inductor (grey box with PULSE printed on it and any of the GND pin from the PCIE power supply)

Just FYI... There is a small problem with measuring voltage like that, but it's still the easiest way in a crowded rig. There's a significant and varying voltage drop on the ground line between the power supply and a card. Just between the M-board's GND terminal and a slot's ground pins, there's a drop of between 0.023v (slot 0) and 0.067v (slot F) on my v1 rig. So while it may look like you're running a card at 0.900v when measured the easy way, it could actually be as low as 0.833v. And I didn't even account for any further drop from the slot to the chips' ground plane.

[inform]

Yeah nice

[greaterninja]

I have a version 2.3 M board.  I updated my Chainminer and now it only detects 2/5 cards.

there is no best.cnf file

and instead of 5x16 = 80 chips, my bitfury shows 88 chips when I know there are only 80 chips.


My question is what do I do now?

[Keefe]

I have a version 2.3 M board.  I updated my Chainminer and now it only detects 2/5 cards.

there is no best.cnf file

and instead of 5x16 = 80 chips, my bitfury shows 88 chips when I know there are only 80 chips.


My question is what do I do now?

Reboot it until the problem goes away. Try reseating boards also.

[kano]

Guess I should mention it in here.

I've written a cgminer driver (it's in the current cgminer 3.8.0/3.8.1) for the BlackArrow Bitfury V1 boards.
I've no idea if that is the same as a standard Bitfury board - I never had one and the twits would never sell me one either.
... maybe I should have put in the copyright header that Punin and Buzzdave aren't allowed to use the driver

Anyway, as I posted about it the other day:

For anyone who has a V1 BlackArrow Bitfury board - the latest cgminer 3.8.1 (3.8.0 also) has the RPi source for a driver now - consider it an early release since there are certain improvements I'll make to it that I've already sorted out what to do - in future versions - but it already runs with great performance.

The configure option is --enable-bab

It runs at 39.9GH/s valid shares with a bit under 10% HW errors above the 39.9GH/s on a standard board
i.e. if you were to also count bad nonces returned (which I prefer not to but I will mention in case anyone wishes to compare that number) it's returning 44GH/s of good+bad nonces.

I've run it with a single board on Raspbian to write and test it - it's based on the V1 initialisation of a single board.
I based the I/O code on the chainminer GPIO/SPI code as should be obvious for anyone who looks at my driver-bab.c and chainminer (as I also mention that at the top of driver-bab.c)

It's a bit CPU hungry so soon I'll be spending more effort on reducing that - 20% CPU with a single board - I've only tested with a single board.

It also doesn't tune the chip at all - so there is also assured performance improvement there - it runs at default settings at the moment.
That's the next effort I'll expend on the driver.

If anyone has a standard V1 BitFury board I'd also be curious to know if it works with them.

[AMD FTW]

Whoever gets me up and running 1 Btc here.

https://asktom.cf/index.php?topic=332035.msg3561229#msg3561229

[spegelius]

Guess I should mention it in here.

I've written a cgminer driver (it's in the current cgminer 3.8.0/3.8.1) for the BlackArrow Bitfury V1 boards.
I've no idea if that is the same as a standard Bitfury board - I never had one and the twits would never sell me one either.
... maybe I should have put in the copyright header that Punin and Buzzdave aren't allowed to use the driver

Anyway, as I posted about it the other day:

For anyone who has a V1 BlackArrow Bitfury board - the latest cgminer 3.8.1 (3.8.0 also) has the RPi source for a driver now - consider it an early release since there are certain improvements I'll make to it that I've already sorted out what to do - in future versions - but it already runs with great performance.

The configure option is --enable-bab

It runs at 39.9GH/s valid shares with a bit under 10% HW errors above the 39.9GH/s on a standard board
i.e. if you were to also count bad nonces returned (which I prefer not to but I will mention in case anyone wishes to compare that number) it's returning 44GH/s of good+bad nonces.

I've run it with a single board on Raspbian to write and test it - it's based on the V1 initialisation of a single board.
I based the I/O code on the chainminer GPIO/SPI code as should be obvious for anyone who looks at my driver-bab.c and chainminer (as I also mention that at the top of driver-bab.c)

It's a bit CPU hungry so soon I'll be spending more effort on reducing that - 20% CPU with a single board - I've only tested with a single board.

It also doesn't tune the chip at all - so there is also assured performance improvement there - it runs at default settings at the moment.
That's the next effort I'll expend on the driver.

If anyone has a standard V1 BitFury board I'd also be curious to know if it works with them.

Tried it, built with --enable-bab, but it didn't find any chips.

Edit: i have 5 v.12 boards, but i suppose that the version shouldn't matter. Had to run with sudo as cgminer wanted access to /dev/mem.

I'll be happy to try more builds/do debug runs, if there will be any development regarding these boards (although it seems that the motivation might be lacking ). Would love to replace the chainminer/stratumproxy combo with more sophisticated system.

[goxed]

Guys I have a question about the H-board.

Where can I download the H-board layout file? I am trying to get one manufactured.

Thanks

 

[GandalfG]

Guys I have a question about the H-board.

Where can I download the H-board layout file? I am trying to get one manufactured.

Thanks

 

Check OP man

[BTClobsta]

Okay I got it working. The only problem I got now is that I can only get bank number 1-11 working. I have 12 plugged in. but only 11 looks to be working.
I’m have a problem running the rig when I have all 16 h boards plugged in. it would just power up and shut off. But when I remove 4 (left with 12) it would power on and I would be able to hash. However it would only hash with 11 boards. i've also notice that after hashing for about 20min or so, some of the h boards stop hashing. only 4 would be hashing still and the rest would stop. why is that?

Any idea on how to fix this problem.

i doubt that its the psu. but i could be wrong. my psu is 700w and 570w towards 12v. i have 2 pci-e cables on the psu.
if you think its the psu. please direct me to the right place to buy the right psu.
psu model:
coolmax
cu700b

[GandalfG]

Okay I got it working. The only problem I got now is that I can only get bank number 1-11 working. I have 12 plugged in. but only 11 looks to be working.
I’m have a problem running the rig when I have all 16 h boards plugged in. it would just power up and shut off. But when I remove 4 (left with 12) it would power on and I would be able to hash. However it would only hash with 11 boards. i've also notice that after hashing for about 20min or so, some of the h boards stop hashing. only 4 would be hashing still and the rest would stop. why is that?

Any idea on how to fix this problem.

i doubt that its the psu. but i could be wrong. my psu is 700w and 570w towards 12v. i have 2 pci-e cables on the psu.
if you think its the psu. please direct me to the right place to buy the right psu.
psu model:
coolmax
cu700b

Mboard version ?

[Keefe]

Okay I got it working. The only problem I got now is that I can only get bank number 1-11 working. I have 12 plugged in. but only 11 looks to be working.
I’m have a problem running the rig when I have all 16 h boards plugged in. it would just power up and shut off. But when I remove 4 (left with 12) it would power on and I would be able to hash. However it would only hash with 11 boards. i've also notice that after hashing for about 20min or so, some of the h boards stop hashing. only 4 would be hashing still and the rest would stop. why is that?

Any idea on how to fix this problem.

i doubt that its the psu. but i could be wrong. my psu is 700w and 570w towards 12v. i have 2 pci-e cables on the psu.
if you think its the psu. please direct me to the right place to buy the right psu.
psu model:
coolmax
cu700b

That PSU is definitely insufficient, because it is multi-rail and has low per-rail limits. Even if the two cables are on separate rails, you can only pull ~400W from them before you exceed the PSU's specs on those.

Go get a high-quality single-rail PSU with at least 700W on the 12V rail.

Here's a good one for $90 if you don't need it to be modular: http://www.newegg.com/Product/Product.aspx?Item=N82E16817182073
I try to find ones that get good reviews from JonnyGuru or HardOCP.

[Swimmer63]

I have two h-cards, v2.2 available if anyone is looking.

https://asktom.cf/index.php?topic=333237.0

[CMaxx22]

Hi! I download the raspberry pi image from the first post https://dl.dropboxusercontent.com/u/43738391/2013-Sep-27_BFSB_GM2.rar and write it to SD. Now how can I connect to Rpi? What is the default IP in this image?

[BTClobsta]

Okay I got it working. The only problem I got now is that I can only get bank number 1-11 working. I have 12 plugged in. but only 11 looks to be working.
I’m have a problem running the rig when I have all 16 h boards plugged in. it would just power up and shut off. But when I remove 4 (left with 12) it would power on and I would be able to hash. However it would only hash with 11 boards. i've also notice that after hashing for about 20min or so, some of the h boards stop hashing. only 4 would be hashing still and the rest would stop. why is that?

Any idea on how to fix this problem.

i doubt that its the psu. but i could be wrong. my psu is 700w and 570w towards 12v. i have 2 pci-e cables on the psu.
if you think its the psu. please direct me to the right place to buy the right psu.
psu model:
coolmax
cu700b

Mboard version ?

mboard is v3.0