Topic: **US** BitFury Setup Guide - page 23. (Read 53265 times)

so which pool works best with bitfury ? people should have an idea by now
friend reporting huge difference between slushs pool and web panel

Not sure if it helps with power distribution and potentially moving noisy bad chips away from good chips. However one thing I know for sure it makes it easy to detect problems as they will stand out and be more obvious. Also later on if you do happen to get better boards you can add them in place of the poor performing ones.

Sitarow, thank you. That as a very detailed reply that I was able to actually follow. I'll be working on it tomorrow.

Question: why move the boards position? I'm just curious.

First move the boards with 880 to the front of the line then the 825 ones and the 770 to the back.

so I would start with
Move 7 to 0 and put 0 on 4 and put 4 to 6
Move 6 to 7 as I think that one has chips that auto tune is still having troubles with.
Move 5 to 3 and 3 to 5.

After you do that then wait about 30 minutes, then run nano /run/shm/.stat.log

If you notice any chips pushing 0.600 or below turn them off by replacing AIfDSo to aifdso on the same line number as the chip. This is important because auto tuning will keep working with them if it can causing errors that may throw off some pools.

Note: I would only consider turning off auto tuning if I am in a temperature controlled environment where the ambient temperature is constant. OR if I am not as aggresive on my speed settings of the chips.

Below is an example of an "EOL" board that was included with my kit. I have simply turned off those chips but will be later cleaning the board with isopropyl alcohol to see if i can clear up some potential issues.

240     aifdso  50      0.000   0.000   0       0       0       0       0       [E:F]   756    
241     AIfDSo  55      1.432   1.871   100     4       0       0       177     [F:0]   36      
242     AIfDSo  55      1.632   1.807   114     3       0       0       171     [F:1]   72      
243     aifdso  51      0.000   0.000   0       0       0       0       0       [F:2]   756    
244     AIfDSo  55      1.761   1.575   123     4       0       0       149     [F:3]   0      
245     AIfDSo  55      1.775   1.733   124     0       0       0       164     [F:4]   42      
246     AIfDSo  55      1.933   1.881   135     2       0       0       178     [F:5]   0      
247     AIfDSo  55      1.847   1.892   129     2       0       0       179     [F:6]   21    
248     AIfDSo  55      1.747   1.797   122     1       0       0       170     [F:7]   0      
249     aifdso  50      0.000   0.000   0       0       0       0       0       [F:8]   756    
250     AIfDSo  55      1.904   1.924   133     2       0       0       182     [F:9]   0      
251     AIfDSo  55      1.675   1.786   117     2       0       0       169     [F:A]   0      
252     AIfDSo  55      1.432   1.776   100     5       1       0       168     [F:B]   72      
253     aifdso  50      0.000   0.000   0       0       0       0       0       [F:C]   756  
254     AIfDSo  55      2.090   1.860   146     1       0       0       176     [F:D]   0      
255     AIfDSo  55      1.689   1.934   118     3       0       0       183     [F:E]   72      
256     AIfDSo  55      1.804   1.913   126     0       0       0       181     [F:F]   36

F:      715     23.179  23.739  1619    26      0       0       13      0       3       (1.449/chip)    78%


I used the CPU speed as a marker for my personal notes for later review. The chips I set to aifdso 50 is the first wave of disabled chips that showed obvious problems. The ones set with 51 are for me to note my second wave of disabled chips.

Remember to only tweak the file after you have moved the cards to the more appropriate slots.
Remember to save the file to /opt/bitfury/best.cnf so that it will keep the changes you made to the file.

Sitarow,

This is what mine displays. Looks like 5 bad and one off?

What would I look for now and how to disable?

speed:6710 noncerate[GH/s]:220.432 (1.722/chip) hashrate[GH/s]:237.606 good:15397 errors:860 spi-err:5 miso-err:0 jobs:322 cores:95% good:122 bad:5 off:1 (best[GH/s]:229.165) Tue Sep 10 01:42:09 2013
0:      825     29.936  31.836  2091    100     0       0       15      1       0       (1.871/chip)    95%
1:      880     31.081  33.083  2171    101     0       0       16      0       0       (1.943/chip)    100%
2:      880     28.146  30.821  1966    35      0       0       16      0       0       (1.759/chip)    98%
3:      770     26.285  29.690  1836    193     1       0       14      2       0       (1.643/chip)    86%
4:      770     25.641  28.031  1791    145     1       0       14      2       0       (1.603/chip)    92%
5:      825     29.664  31.286  2072    85      1       0       15      0       1       (1.854/chip)    93%
6:      880     21.289  22.873  1487    94      1       0       16      0       0       (1.331/chip)    97%
7:      880     28.390  29.986  1983    107     1       0       16      0       0       (1.774/chip)    99%

My other looks worse:

speed:7260 noncerate[GH/s]:203.467 (1.413/chip) hashrate[GH/s]:224.172 good:14212 errors:749 spi-err:6 miso-err:0 jobs:323 cores:89% good:132 bad:8 off:4 (best[GH/s]:212.257) Tue Sep 10 01:47:15 2013

http://punin.com/README_BFSB_v0.1.txt

Thanks.  Where can I find out what each column represents?

check also that the cards are properly seated in the socket. Also if you get any chips that are operating at sub 1.0 speed then set it to disabled and see if that helps.


type this in ssh console

nano /run/shm/.stat.log

scroll to the bottom by pressing page down and see if you have any bad or off chips that may have been detected by autotune.

If you do i start by disabling them to get the board working good.

Yup I did all that and found that it didn't help.  I think there's a problem with the card.  Sometimes it hashes at regular rates, other times it stops hashing for some time.  It's like sometimes the chips are partially failing...

check what your /opt/bitfury/proxy_pools.conf is set too.

I would consider also restarting the stratum server first before restarting the miner.

Hi all,

I just noticed I am running into problems after 2 days hashing on slush.  It is now showing 0 noncerate.  I've tried start/stop miner but no good.  ssh'd in and tried this:

root@bitfury:/opt/bitfury# ./start-miner-console.sh
INIT: 0 chips detected
./start-miner-console.sh: line 1:  2274 Floating point exception/opt/bitfury/chainminer/miner

Also tried rebooting rpi.  The only thing I haven't done is power off the whole unit -- which I cannot do till I get home.

Did something die?

Thanks Sitarow!

set your diff 22  and 190 wait for 30 minutes or more

Can someone tell me why I'm seeing the below hashrate differences for my 400 gh/s kit?

Bitfury admin pages noncerate is 417.51 gh/s. Slush's pool is 2.9 gh/s!

Bitfury admin page 1 (xxx.xxx.x.250)
Noncerate: 45.04GH/s
pool config: stratum.bitcoin.cz, 3333, idxxx.worker1

Bitfury admin page 2 (xxx.xxx.x.251)
Noncerate: 372.474GH/s
pool config: stratum.bitcoin.cz, 3333, idxxx.worker2

Slush's Pool
                     curr                                                sugg.
login               shares  score         Mhash/s   stratum  diff.
idxxx.worker1   954      2043.1946  1411.922   yes       32
idxxx.worker2   1009    2129.3236  1493.323   yes        256

Average hashrate in last 10 rounds: 3708 Mhash/s
Unconfirmed reward: 0.01430835 BTC
Confirmed reward:   0.00182180 BTC
These rewards are for about 1 day of mining.


Thank you.

its in the bottom right on the board to the right of the usb ports.

Congrats Af_newbie -

To get the extra performance, did you manually configure autotune disabled from the problem chips and then set their speed to 0?

Thanks guys I'll try both tomorrow. Rig is not with me now. Is the Rpi light on the bottom or top?

Thanks everybody for help!! finally got my kit start hashing 25G+ THANK YOU VERY MUCH 

When you power on the raspberry pi you should see a red led indicating power and activity led light will flash at about the same time if your SDCard is inserted and detected.

If all you get is the Red LED then perhaps the sd card is the issue?

Do you have a mini-USB cable for any other device?  Cell phone, Kindle, iPad?  If so, take the rPI off the m-board and power it up separately using that mini-USB, see if it works running separately.