Topic: ["WAIT LIST"] BFL Mini-Rig Order Date/Ship Date - page 3. (Read 32420 times)

 [P]ool management Settings [D]isplay options [Q]uit
 BFL  0:  41.5C         | 1377.0/1370.3Mh/s | A:32 R:0 HW:0 U:10.21/m
 BFL  1:  40.2C         | 1317.0/1279.0Mh/s | A:63 R:0 HW:0 U:20.10/m
 BFL  2:  41.0C         | 1317.0/1279.0Mh/s | A:70 R:0 HW:0 U:22.33/m
 BFL  3:  40.7C         | 1317.1/1279.0Mh/s | A:52 R:0 HW:0 U:16.59/m
 BFL  4:  38.2C         | 1317.0/1279.0Mh/s | A:70 R:0 HW:0 U:22.33/m
BFL  5:  42.4C         | 1377.2/1370.3Mh/s | A: 5 R:0 HW:0 U: 1.60/m
 BFL  6:  47.4C         | 1317.0/1279.0Mh/s | A:64 R:0 HW:0 U:20.42/m
 BFL  7:  42.5C         | 1377.0/1370.3Mh/s | A: 3 R:0 HW:0 U: 0.96/m
 BFL  8:  39.8C         | 1317.1/1279.0Mh/s | A:58 R:0 HW:0 U:18.51/m
 BFL  9:  43.8C         | 1317.0/1279.0Mh/s | A:58 R:0 HW:0 U:18.51/m
 BFL 10:  45.7C         | 1317.1/1279.0Mh/s | A:61 R:0 HW:0 U:19.46/m
 BFL 11:  38.1C         | 1317.1/1279.0Mh/s | A:54 R:0 HW:0 U:17.23/m
 BFL 12:  39.8C         | 1317.0/1279.0Mh/s | A:66 R:0 HW:0 U:21.06/m
 BFL 13:  41.6C         | 1494.1/1461.7Mh/s | A:65 R:0 HW:0 U:20.74/m
 BFL 14:  39.0C         | 1317.1/1279.0Mh/s | A:58 R:0 HW:0 U:18.51/m
 BFL 15:  41.0C         | 1317.0/1279.0Mh/s | A:53 R:0 HW:0 U:16.91/m
BFL 16:  37.5C         | 1377.0/1324.6Mh/s | A:16 R:0 HW:0 U: 5.10/m
 BFL 17:  42.6C         | 1317.1/1279.0Mh/s | A:54 R:0 HW:0 U:17.23/m
 BFL 18:  39.8C         | 1317.0/1279.0Mh/s | A:51 R:0 HW:0 U:16.27/m


Note #'s 0,5,7,and 16 - a result of the 1320 FW in the auto-tune package.  Don't use it!  

DOn't use the auto-tune package - just sayin'

Came from BFL with 4 units at 1500 and now only have 1 that will run 1500.  The other 3 bad flashed when the USB hub died and had to start all over.

So like a few others it seems I end up with a rig just making 25.2

18 x 1320 = 23.76 Gh/s
1 x 1500 = 1.5 Gh/s

25.26 Gh/s total 

Got mine on 8/14

It was shipped on 8/10

Probably safe to assume that everyone before you has received theirs as well.  Just not updated on the list.

Inaba set up my rig yesterday  (8/16/2012 for the OP)

U:344.2/m (found shares per minute) seems to be the important number. divided by seconds in a minute (/60) multiplied by the number of hashes needed to find a share (*2^32) and converted into Giga (/10^9) I get a hashrate of 24.6 GH/s. This is slightly below the advertised speed but still acceptable.

So, this is your unit?  I guess the list needs to be updated.
Probably means the JWU42 also received his.

Just trying to figure out the timeline.

 cgminer version 2.6.4 - Started: [2012-08-15 12:18:14]
--------------------------------------------------------------------------------
 (5s):27807.5 (avg):24947.1 Mh/s | Q:1591  A:20967  R:235  HW:0  E:1318%  U:344.2/m
 TQ: 0  ST: 21  SS: 0  DW: 150  NB: 10  LW: 25125  GF: 1  RF: 0
 Connected to ... with LP as user ...
 Block: 000001419794b1cf01e6d5abca6d92fe...  Started: [13:18:46]
--------------------------------------------------------------------------------
 [P]ool management [.]ettings [D]isplay options [Q]uit
 BFL  0:  45.7C         | 1309.7/1313.9Mh/s | A:1121 R:16 HW:0 U:18.40/m
 BFL  1:  45.8C         | 1309.8/1313.9Mh/s | A:1076 R: 9 HW:0 U:17.67/m
 BFL  2:  38.0C         | 1309.8/1313.9Mh/s | A:1109 R:12 HW:0 U:18.21/m
 BFL  3:  42.2C         | 1309.9/1312.8Mh/s | A:1082 R:11 HW:0 U:17.76/m
 BFL  4:  45.1C         | 1309.8/1312.8Mh/s | A:1100 R:12 HW:0 U:18.06/m
 BFL  5:  37.1C         | 1309.7/1312.8Mh/s | A:1087 R:14 HW:0 U:17.85/m
 BFL  6:  39.9C         | 1309.7/1312.8Mh/s | A:1110 R:13 HW:0 U:18.22/m
 BFL  7:  35.8C         | 1310.1/1312.8Mh/s | A:1122 R:15 HW:0 U:18.42/m
 BFL  8:  38.5C         | 1309.8/1312.8Mh/s | A:1136 R:14 HW:0 U:18.65/m
 BFL  9:  39.0C         | 1309.9/1312.8Mh/s | A:1141 R:14 HW:0 U:18.73/m
 BFL 10:  40.6C         | 1309.6/1312.8Mh/s | A:1114 R:13 HW:0 U:18.29/m
 BFL 11:  44.2C         | 1309.7/1312.8Mh/s | A:1102 R:10 HW:0 U:18.09/m
 BFL 12:  40.8C         | 1309.7/1312.8Mh/s | A:1164 R:16 HW:0 U:19.11/m
 BFL 13:  44.4C         | 1310.0/1312.8Mh/s | A:1063 R: 9 HW:0 U:17.45/m
 BFL 14:  38.7C         | 1309.8/1312.8Mh/s | A:1079 R:16 HW:0 U:17.72/m
 BFL 15:  39.1C         | 1309.8/1312.8Mh/s | A:1154 R:12 HW:0 U:18.95/m
 BFL 16:  41.1C         | 1309.8/1312.8Mh/s | A:1097 R:15 HW:0 U:18.01/m
 BFL 17:  35.2C         | 1309.8/1312.8Mh/s | A:1052 R: 9 HW:0 U:17.27/m
 BFL 18:  37.9C         | 1311.0/1313.9Mh/s | A:1080 R: 5 HW:0 U:17.73/m
--------------------------------------------------------------------------------

after downgrading to 1300MH/s all modules perform as expected. The hashrate reported by cgminer did not change much but the hash rate reported by the pool increased :-) In other words the "U" value is very important (something like "luck-rate" which should be ~ 100% [see slush pool / ztex driver] is missing in the statistics). I guess the 1500MH/s firmware causes the device to miss many shares (50%-80%) without reporting hardware problems.

I also have to admit that with "-S auto" I did detect 2 more devices (not shown) but these had communication problems, very low hashrate  and 100 times smaller utility (U) than the other devices, so I don't use them. I was not able to detect them with easyminer.

That's correct, rjk.  I've not seen a 20 unit rig yet, but there are 17 - 19 unit rigs shipping currently.

The number of modules has been reported anywhere from 17 to 20, and it seems that some are slower than expected so more are added to make sure the device as a whole performs as advertised.

Both of mine have 18, but each one hashes at the advertised speed.

This is because one MiniRig is actually 20 (+/- 1) Next generation Singles (which probably won't ever be sold individually).

I use cgminer but have little experience with this sa far and have problems with the interpretation of the output.

 cgminer version 2.6.4 - Started: [2012-08-14 20:53:04]
--------------------------------------------------------------------------------
 (5s):24426.6 (avg):25884.2 Mh/s | Q:478  A:4912  R:26  HW:0  E:1028%  U:263.3/m
 TQ: 0  ST: 20  SS: 0  DW: 65  NB: 3  LW: 7930  GF: 0  RF: 0
 Connected to ... with LP as user ...
 Block: 0000066476ac30aede3828015e1fe66d...  Started: [21:10:31]
--------------------------------------------------------------------------------
 [P]ool management [.]ettings [D]isplay options [Q]uit
 BFL  0:  46.5C         | 1491.5/1492.8Mh/s | A:154 R:1 HW:0 U: 8.26/m
 BFL  1:  47.0C         | 1491.7/1492.8Mh/s | A: 45 R:0 HW:0 U: 2.41/m
 BFL  2:  37.9C         | 1491.7/1489.0Mh/s | A:186 R:1 HW:0 U: 9.97/m
 BFL  3:  42.3C         | 1491.6/1489.0Mh/s | A: 64 R:0 HW:0 U: 3.43/m
 BFL  4:  45.6C         | 1491.7/1489.0Mh/s | A: 76 R:0 HW:0 U: 4.07/m
 BFL  5:  37.6C         | 1491.7/1489.0Mh/s | A: 93 R:0 HW:0 U: 4.99/m
 BFL  6:  37.5C         | 1314.9/1316.3Mh/s | A:357 R:3 HW:0 U:19.14/m
 BFL  7:  40.7C         | 1314.9/1316.3Mh/s | A:335 R:2 HW:0 U:17.96/m
 BFL  8:  43.8C         | 1314.9/1316.3Mh/s | A:346 R:2 HW:0 U:18.55/m
 BFL  9:  39.9C         | 1314.9/1316.3Mh/s | A:329 R:3 HW:0 U:17.64/m
 BFL 10:  43.4C         | 1314.9/1316.3Mh/s | A:327 R:1 HW:0 U:17.53/m
 BFL 11:  37.8C         | 1313.9/1312.4Mh/s | A:305 R:1 HW:0 U:16.35/m
 BFL 12:  39.1C         | 1313.9/1312.4Mh/s | A:344 R:2 HW:0 U:18.44/m
 BFL 13:  35.5C         | 1313.8/1312.4Mh/s | A:325 R:1 HW:0 U:17.42/m
 BFL 14:  37.9C         | 1314.0/1312.4Mh/s | A:329 R:3 HW:0 U:17.64/m
 BFL 15:  38.5C         | 1313.8/1312.4Mh/s | A:310 R:1 HW:0 U:16.62/m
 BFL 16:  40.4C         | 1313.9/1312.4Mh/s | A:331 R:3 HW:0 U:17.74/m
 BFL 17:  34.6C         | 1313.9/1312.4Mh/s | A:369 R:2 HW:0 U:19.78/m
 BFL 18:  37.4C         | 1313.9/1312.4Mh/s | A:304 R:0 HW:0 U:16.30/m
--------------------------------------------------------------------------------

I see that:
- there are 19 devices
- the devices that report higher hash rate (MH/s) have few accepted shares (A:...) and low utility (U:...)
- there are no hardware (HW:0) errors

... to see this in easyminer I would have to connect it to a windows machine :-(

Three weeks ago (there may be an updated version out now?), the auto-tune package hosed about 10 of the FPGAs in my two rigs.  Sonny sent me two firmware packages to load manually, which fixed the issue.  I also found that updating the firmware on one FPGA while the rest are mining almost always resulted in a bad firmware update, requiring me to do it over several times until I realized what was wrong.

Here are the two firmware packages Sonny sent to me.
https://www.box.com/s/e8cd6a2e651cc87b604c

What I did in the end was to use EasyMiner to view the properties of each FPGA and identify which ones were using the 1320 MH/s firmware.  I updated each of them to use the 1500 MH/s firmware and then mined with them for a while to see if errors showed up.  All but one had a 20%+ error rate, so I flashed them back to the 1320 MH/s firmware.

The Mini-Rig FPGAs don't throttle when there are errors, so I guess it was worthwhile since I got one running at 1500 MH/s with only a 2% error rate in exchange for a 13% speed increase.

Thanks, running. I had to add -S for all devices by hand though (debian). Did anybody try to tune the speed with the bfl driver: "Auto-Tune Package – For speed tuning Mini Rigs" ?

Plug in to USB, install and run cgminer, shouldn't be any more complicated than that.

I have picked up my rig today ... there are no instructions in the box :-) how do You use the rig ? do I have to install the easy-miner from the bfl page (http://www.butterflylabs.com/drivers/) ? I hope I can use this on a linux machine :-(

Very good to hear Inspector, take care of yourself.    Back on topic:  #of mini-rigs purchased by me = 100................in a dream I had.