Bitcoin Forum
December 05, 2024, 10:36:38 AM *
News: Latest Bitcoin Core release: 28.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 [21] 22 23 24 25 26 27 »
  Print  
Author Topic: Ultra Under-overclock image for A2 Innosilicon by Emdje - V5.0  (Read 79804 times)
BakSAj
Full Member
***
Offline Offline

Activity: 157
Merit: 100


View Profile
December 09, 2015, 07:54:54 PM
 #401

Hi all, Emdje, did you manage to fix pool-switching problem (mainly betarigs) on 4.0 image? E.g. by putting back the old stratum, but still supporting the xnonce? Thanks.
emdje (OP)
Hero Member
*****
Offline Offline

Activity: 686
Merit: 500


View Profile WWW
December 10, 2015, 12:10:44 AM
 #402

Hi all, Emdje, did you manage to fix pool-switching problem (mainly betarigs) on 4.0 image? E.g. by putting back the old stratum, but still supporting the xnonce? Thanks.

I am not sure will have to check that.
Prelude
Legendary
*
Offline Offline

Activity: 1596
Merit: 1000



View Profile
December 12, 2015, 05:18:45 PM
 #403

I have a question that hopefully someone will be able to answer.

I'm trying to run my 60 and 88mh units using a server power supply. I cut the ATX harness off of a dead PSU, and connected all of the 12v 5v and 3.3v wires on the harness to the appropriate server PSU outputs. I thought for sure that would be all that's needed, but the controller doesn't work. Any idea what I'm missing?
BakSAj
Full Member
***
Offline Offline

Activity: 157
Merit: 100


View Profile
December 12, 2015, 06:41:08 PM
 #404

Hi all, Emdje, did you manage to fix pool-switching problem (mainly betarigs) on 4.0 image? E.g. by putting back the old stratum, but still supporting the xnonce? Thanks.

I am not sure will have to check that.

Seems like betarigs is out of game for good, so Im back to 4.0 and MRR.
mjgraham
Full Member
***
Offline Offline

Activity: 188
Merit: 100


View Profile
December 13, 2015, 04:07:50 AM
 #405

I have a question that hopefully someone will be able to answer.

I'm trying to run my 60 and 88mh units using a server power supply. I cut the ATX harness off of a dead PSU, and connected all of the 12v 5v and 3.3v wires on the harness to the appropriate server PSU outputs. I thought for sure that would be all that's needed, but the controller doesn't work. Any idea what I'm missing?
What server PSU are you attempting to use? How much of it does not work, any , some lights or just dead? have you checked to see if there is output on the lines?

Prelude
Legendary
*
Offline Offline

Activity: 1596
Merit: 1000



View Profile
December 15, 2015, 02:14:34 PM
 #406

I have a question that hopefully someone will be able to answer.

I'm trying to run my 60 and 88mh units using a server power supply. I cut the ATX harness off of a dead PSU, and connected all of the 12v 5v and 3.3v wires on the harness to the appropriate server PSU outputs. I thought for sure that would be all that's needed, but the controller doesn't work. Any idea what I'm missing?
What server PSU are you attempting to use? How much of it does not work, any , some lights or just dead? have you checked to see if there is output on the lines?

The PSU is a PFE1100. It's breakout board outputs 12v and 5v, and I'm using a buck converter to make 3.3v from the 12v output.
mjgraham
Full Member
***
Offline Offline

Activity: 188
Merit: 100


View Profile
December 15, 2015, 07:09:26 PM
 #407

I have a question that hopefully someone will be able to answer.

I'm trying to run my 60 and 88mh units using a server power supply. I cut the ATX harness off of a dead PSU, and connected all of the 12v 5v and 3.3v wires on the harness to the appropriate server PSU outputs. I thought for sure that would be all that's needed, but the controller doesn't work. Any idea what I'm missing?
What server PSU are you attempting to use? How much of it does not work, any , some lights or just dead? have you checked to see if there is output on the lines?

The PSU is a PFE1100. It's breakout board outputs 12v and 5v, and I'm using a buck converter to make 3.3v from the 12v output.

So nothing comes on? I don't really have a A2 miner but I have see photos of the controller board, I am going to assume that you do have power on the 12 and 5V rails , I don't know if the 3.3v is needed nothing that I have seen needs 3.3 just 12 and 5 and it looked like there might have been regulators on the board already for a 12->5 conversion. I would check power on the Rpi, you should be able to probe the capacitor at the normal input jack , if you got 5v there and it is still not on then you might have a big problem. On the connectors going down to the hash boards I *think* that pins 2&4 are +5 and 6,8&10 are gnd so you could check that as well. Really I think all the controller board does is bridge on the ATX psu and power the PI, I dont really know why it would need +12 at all unless it does some basic power monitoring to see if there is power there.

This are all guesses really , if you could get me a good straight on shot of front and back, I am working on a new controller board, I just only have one card I have adapted to the RPi, that power supply is pretty $$$ I hope you didn't have to pay that much for it.

emdje (OP)
Hero Member
*****
Offline Offline

Activity: 686
Merit: 500


View Profile WWW
December 16, 2015, 02:50:37 PM
 #408

Also check the polarity of the power going to the boards with the original power supply. I had to switch the plus and minus with the pcu I am using due to the connector being different.
ZeroGee
Member
**
Offline Offline

Activity: 92
Merit: 10


View Profile
January 02, 2016, 06:39:58 AM
Last edit: January 02, 2016, 07:49:20 AM by ZeroGee
 #409

So I decided to take a crack at overvolting and overclocking my A2 110. The guide provided earlier in this thread does not cover the V4 boards, and I could not find one on the forums so I decided to post it here.

The A2 110 V4 blade buck controllers are based on Linear Technology's LTC3856. Datasheet. Following Pg 29, the feedback resistors will be connected to FB (pin 2), and dictate the output voltage via the equation Vout=0.6*(1+RB/RA). The controllers are set up to use the chip's diffout configuration, but the equation remains the same.

There are 3 different controller layouts on this board:
One side
Other side
End of blade

As we see in the photo, there are two resistors labeled 01C and 66B, 10kΩ and 4.75kΩ respectively.
This yields a stock voltage of 0.6*(1+4.75/10)=0.885V

I figure I want around 0.925V, following the target voltage earlier in this thread.

There are two ways to handle this. We can decrease RA or increase RB. I had a lot of 24kΩ and 27kΩ resistors left over from a previous project. It just so happened that if I used one of each value in parallel with each RA and RB, 27kΩ||10kΩ=7.297kΩ and 24kΩ||4.75kΩ=3.965kΩ yields 0.926V.

So I soldered a 27kΩ across the resistor marked 01C and a 24kΩ across the resistor marked 66B. I ended up with ~0.933V on average.

If you want to only solder one resistor, Rp in parallel with RA (01C), it would follow this equation: Vout=0.6*(1+(4750/(10000*Rp/(10000+Rp)))). Solve for Rp given a target Vout.

I've yet to do long term overclock testing. I'm testing 1200MHz stability for now. On average the power consumption is up ~2A to ~15A on the 12V rail. My stock power supply is not able to supply more than 4 blades stock, much less 4 overclocked blades. I just split the load half and half between the stock power supply and another 1kW power supply. I should have some numbers soon.

After 1hr of 1300MHz, it seems successful. I'll run it overnight and see how it fares.
mjgraham
Full Member
***
Offline Offline

Activity: 188
Merit: 100


View Profile
January 04, 2016, 12:49:01 PM
 #410

I had went with single parallel option on the 10k, seemed to work fine, running at 0.94v although math says 0.92v, wish I had done some more testing before hand though but hey that is the way it is. Good job!, while crude this is a chart I did with this board, I just ran 50k shares at each frequency, I should have ran more but in the interest of not taking 3 hrs for a test.

Frequency|
1100
1120
1140
1160
1180
1200
1220
1240
1260
1280
1300
1320
1340
1360
1380
1400
HW errors|
2
2
0
4
0
5
4
1
10
9
17
14
36
43
90
89
Current on 12v|
14.28
14.45
14.67
15.06
15.23
15.46
15.64
15.76
16.13
16.37
16.63
16.75
17.2
17.44
17.53
17.89
Watts
173.502
175.5675
178.2405
182.979
184.8922
187.6844
189.8696
191.3264
195.8182
198.5681
201.7219
203.1775
208.636
211.1984
212.2883
216.6479

ZeroGee
Member
**
Offline Offline

Activity: 92
Merit: 10


View Profile
January 04, 2016, 02:23:32 PM
 #411

Good to know someone got some use from this information! I speculate the increased voltage might have to do with the values they've setup diffout. I did notice a slight change in voltage (+10mV) under load. I didn't trace the circuit out completely, but I think it's a pretty logical conclusion.

That's good information. What does HW error mean for you? Does that include conditions that call for board reinit or just invalid nonce? The current draw is a little bit different from my measurements, but I'm using a particularly cheap DC clamp meter for current measurement for now.

I've been running it at 1300MHz since my post, getting about 112MH/s accepted according to cgmrrd and the pool's stats. Though, I've yet to overvolt one of the boards. I'm running it at 1200MHz and it's being used as a reference.

I'll be trying values between 1320MHz and 1380MHz today to determine my highest accepted rate.

1400MHz is not stable for me though. I get around 80MH/s accepted. I only tested this for 30 minutes and switched back to 1300MHz. Oh well. It could be a power supply thing, though. This may require further testing at a later date once I have a better power solution.
mjgraham
Full Member
***
Offline Offline

Activity: 188
Merit: 100


View Profile
January 04, 2016, 04:12:50 PM
 #412

yea there are a lot of options, those were just the HW errors when cgminer exits out on the board, yea I guess really I am talking just a few mV that isn't to bad, I may put the board back to stock and sweep the frequencys again , as for current I am using one of these http://www.alliedelec.com/fw-bell-bb-100/70103851/ works pretty well but have to get the offset and do some math but the meter takes care of that , oh yea the +-15 supply that is annoying as well .

I notice there is a thermal pad on the top heatsink, did  they use one on both sides or just the top or is that something you did?

The blue board I have has some no name dc-dc converter ICs and it looks like the feedback is something odd like a 11k and a 13k so right now I don't know the math to make that work out unless there are other parts in line else where.

ZeroGee
Member
**
Offline Offline

Activity: 92
Merit: 10


View Profile
January 04, 2016, 05:19:48 PM
 #413

Gotcha. When using the normal equipment it appears the webUI interface separates invalid nonce and the other chip specific errors into two categories, HW and device rejected. Even then, it gives a percentage value that I don't think is weighted properly. That's why I ask. That's not a bad sensor from the look of it. Probably more accurate than my current instrument, assuming the maths and biases are correct.

These have a very large silpad for the topside and some thermal paste (silver type by appearance, not a whole lot...) for the bottom side. The silpads are not very accurately cut, as you can see in some of the photos.

Oh? That's interesting. I couldn't find any photos in your photo collection of the chips you're talking about. They're not the same chips posted in the guide from earlier in the thread? Which I'm pretty sure happen to be the same chips on the Zeusminer x32 boards, RT8809B if I remember correctly.

PS: 1320MHz seems to be yielding more than 1300MHz in terms of pool accepted. Time to move to 1340MHz.
mjgraham
Full Member
***
Offline Offline

Activity: 188
Merit: 100


View Profile
January 04, 2016, 07:34:52 PM
 #414

if your saving log files or anything and like perl I wrote a little thing to parse the files and tell you which chips had errors you get some output like this, as you can see I have one really happy chip on the second board.


Board 0 Chip  1 Errors  2228
Board 0 Chip  2 Errors   767
Board 0 Chip  3 Errors   390
Board 0 Chip  4 Errors  1504
Board 0 Chip  5 Errors   408
Board 0 Chip  6 Errors     0
Board 0 Chip  7 Errors  1331
Board 0 Chip  8 Errors   775
Board 0 Chip  9 Errors   318
Board 0 Chip 10 Errors   701
Board 0 Chip 11 Errors   567
Board 0 Chip 12 Errors     0

Board 1 Chip  1 Errors  2731
Board 1 Chip  2 Errors   755
Board 1 Chip  3 Errors   935
Board 1 Chip  4 Errors   271
Board 1 Chip  5 Errors   548
Board 1 Chip  6 Errors     0
Board 1 Chip  7 Errors     2
Board 1 Chip  8 Errors  1048
Board 1 Chip  9 Errors 15264
Board 1 Chip 10 Errors   557
Board 1 Chip 11 Errors  2558
Board 1 Chip 12 Errors     0

ZeroGee
Member
**
Offline Offline

Activity: 92
Merit: 10


View Profile
January 04, 2016, 10:03:11 PM
 #415

Good idea! That chip seven is super happy.

I decided to make a (not so elegant) shell script to take the cgminer.log and do something similar.

I'm putting it here in case it's useful to anyone, not because I'm proud of it. It's very rolled out and hastily written...


Code:
#!/bin/bash

sudo rm chiperror.rep
sudo touch chiperror.rep
sudo chmod 777 chiperror.rep

_0_1=$(grep -o 'chip(cs0) 1' cgminer.log | wc -l)
_0_2=$(grep -o 'chip(cs0) 2' cgminer.log | wc -l)
_0_3=$(grep -o 'chip(cs0) 3' cgminer.log | wc -l)
_0_4=$(grep -o 'chip(cs0) 4' cgminer.log | wc -l)
_0_5=$(grep -o 'chip(cs0) 5' cgminer.log | wc -l)
_0_7=$(grep -o 'chip(cs0) 7' cgminer.log | wc -l)
_0_8=$(grep -o 'chip(cs0) 8' cgminer.log | wc -l)
_0_9=$(grep -o 'chip(cs0) 9' cgminer.log | wc -l)
_0_10=$(grep -o 'chip(cs0) 10' cgminer.log | wc -l)
_0_11=$(grep -o 'chip(cs0) 11' cgminer.log | wc -l)

_1_1=$(grep -o 'chip(cs1) 1' cgminer.log | wc -l)
_1_2=$(grep -o 'chip(cs1) 2' cgminer.log | wc -l)
_1_3=$(grep -o 'chip(cs1) 3' cgminer.log | wc -l)
_1_4=$(grep -o 'chip(cs1) 4' cgminer.log | wc -l)
_1_5=$(grep -o 'chip(cs1) 5' cgminer.log | wc -l)
_1_7=$(grep -o 'chip(cs1) 7' cgminer.log | wc -l)
_1_8=$(grep -o 'chip(cs1) 8' cgminer.log | wc -l)
_1_9=$(grep -o 'chip(cs1) 9' cgminer.log | wc -l)
_1_10=$(grep -o 'chip(cs1) 10' cgminer.log | wc -l)
_1_11=$(grep -o 'chip(cs1) 11' cgminer.log | wc -l)

_2_1=$(grep -o 'chip(cs2) 1' cgminer.log | wc -l)
_2_2=$(grep -o 'chip(cs2) 2' cgminer.log | wc -l)
_2_3=$(grep -o 'chip(cs2) 3' cgminer.log | wc -l)
_2_4=$(grep -o 'chip(cs2) 4' cgminer.log | wc -l)
_2_5=$(grep -o 'chip(cs2) 5' cgminer.log | wc -l)
_2_7=$(grep -o 'chip(cs2) 7' cgminer.log | wc -l)
_2_8=$(grep -o 'chip(cs2) 8' cgminer.log | wc -l)
_2_9=$(grep -o 'chip(cs2) 9' cgminer.log | wc -l)
_2_10=$(grep -o 'chip(cs2) 10' cgminer.log | wc -l)
_2_11=$(grep -o 'chip(cs2) 11' cgminer.log | wc -l)

_3_1=$(grep -o 'chip(cs3) 1' cgminer.log | wc -l)
_3_2=$(grep -o 'chip(cs3) 2' cgminer.log | wc -l)
_3_3=$(grep -o 'chip(cs3) 3' cgminer.log | wc -l)
_3_4=$(grep -o 'chip(cs3) 4' cgminer.log | wc -l)
_3_5=$(grep -o 'chip(cs3) 5' cgminer.log | wc -l)
_3_7=$(grep -o 'chip(cs3) 7' cgminer.log | wc -l)
_3_8=$(grep -o 'chip(cs3) 8' cgminer.log | wc -l)
_3_9=$(grep -o 'chip(cs3) 9' cgminer.log | wc -l)
_3_10=$(grep -o 'chip(cs3) 10' cgminer.log | wc -l)
_3_11=$(grep -o 'chip(cs3) 11' cgminer.log | wc -l)

_4_1=$(grep -o 'chip(cs4) 1' cgminer.log | wc -l)
_4_2=$(grep -o 'chip(cs4) 2' cgminer.log | wc -l)
_4_3=$(grep -o 'chip(cs4) 3' cgminer.log | wc -l)
_4_4=$(grep -o 'chip(cs4) 4' cgminer.log | wc -l)
_4_5=$(grep -o 'chip(cs4) 5' cgminer.log | wc -l)
_4_7=$(grep -o 'chip(cs4) 7' cgminer.log | wc -l)
_4_8=$(grep -o 'chip(cs4) 8' cgminer.log | wc -l)
_4_9=$(grep -o 'chip(cs4) 9' cgminer.log | wc -l)
_4_10=$(grep -o 'chip(cs4) 10' cgminer.log | wc -l)
_4_11=$(grep -o 'chip(cs4) 11' cgminer.log | wc -l)

_5_1=$(grep -o 'chip(cs5) 1' cgminer.log | wc -l)
_5_2=$(grep -o 'chip(cs5) 2' cgminer.log | wc -l)
_5_3=$(grep -o 'chip(cs5) 3' cgminer.log | wc -l)
_5_4=$(grep -o 'chip(cs5) 4' cgminer.log | wc -l)
_5_5=$(grep -o 'chip(cs5) 5' cgminer.log | wc -l)
_5_7=$(grep -o 'chip(cs5) 7' cgminer.log | wc -l)
_5_8=$(grep -o 'chip(cs5) 8' cgminer.log | wc -l)
_5_9=$(grep -o 'chip(cs5) 9' cgminer.log | wc -l)
_5_10=$(grep -o 'chip(cs5) 10' cgminer.log | wc -l)
_5_11=$(grep -o 'chip(cs5) 11' cgminer.log | wc -l)

echo -e "Board \t1 \t2 \t3 \t4 \t5 \t6" | tee chiperror.rep
echo -e "Chip 1 \t$_0_1 \t$_1_1 \t$_2_1 \t$_3_1 \t$_4_1 \t$_5_1" | tee chiperror.rep
echo -e "Chip 2 \t$_0_2 \t$_1_2 \t$_2_2 \t$_3_2 \t$_4_2 \t$_5_2" | tee chiperror.rep
echo -e "Chip 3 \t$_0_3 \t$_1_3 \t$_2_3 \t$_3_3 \t$_4_3 \t$_5_3" | tee chiperror.rep
echo -e "Chip 4 \t$_0_4 \t$_1_4 \t$_2_4 \t$_3_4 \t$_4_4 \t$_5_4" | tee chiperror.rep
echo -e "Chip 5 \t$_0_5 \t$_1_5 \t$_2_5 \t$_3_5 \t$_4_5 \t$_5_5" | tee chiperror.rep
echo -e "Chip 6 \t$_0_7 \t$_1_7 \t$_2_7 \t$_3_7 \t$_4_7 \t$_5_7" | tee chiperror.rep
echo -e "Chip 7 \t$_0_8 \t$_1_8 \t$_2_8 \t$_3_8 \t$_4_8 \t$_5_8" | tee chiperror.rep
echo -e "Chip 8 \t$_0_9 \t$_1_9 \t$_2_9 \t$_3_9 \t$_4_9 \t$_5_9" | tee chiperror.rep
echo -e "Chip 9 \t$_0_10 \t$_1_10 \t$_2_10 \t$_3_10 \t$_4_10 \t$_5_10" | tee chiperror.rep
echo -e "Chip10 \t$_0_11 \t$_1_11 \t$_2_11 \t$_3_11 \t$_4_11 \t$_5_11" | tee chiperror.rep

exit 0

Output:
Code:
Board 1 2 3 4 5 6
Chip 1 21 20 14 35 32 12
Chip 2 7 15 4 14 9 21
Chip 3 0 1 5 2 7 8
Chip 4 0 13 2 5 9 4
Chip 5 0 2 16 3 1 9
Chip 6 7 18 8 9 8 8
Chip 7 6 35 8 10 21 10
Chip 8 7 8 10 20 8 11
Chip 9 8 11 8 16 13 4
Chip10 3 3 3 7 0 6

Now have it run with cgmrrd, taking out the "| tee" for ">>" and put the output file on a reports page for easy access.

You could almost determine which sets of chips to investigate into regarding the voltage they're getting. Hmmm... My first chip across all boards consistently having errors is interesting.
mjgraham
Full Member
***
Offline Offline

Activity: 188
Merit: 100


View Profile
January 05, 2016, 01:26:52 AM
 #416

yea got one super happy and one super sad on the same board, i have noticed that every now and then they will be very evenly spread, normally when I am tracking down something. Here is the code to my thing , not amazing but works . I got some heatsinks on the way and maybe two more cards, don't know yet.

perl bad_chips.pl /path/to/logfile


bad_chips.pl
Code:
my @a2;
my $boards = 2;
init_var();
open ($fh , '<', $ARGV[0]) or die "No input cgminer Log file\n";
while ($row = <$fh>) {
chomp $row;
if ($row =~ /invalid/ && $row =~ /cs/) {
@temp = split(/ /, $row);
$temp_cs = substr($temp[3],7,1);
$a2[$temp_cs][$temp[4]]++;
}
}
report();
close $fh;
sub init_var {
for ($cs=0;$cs<$boards;$cs++) {
        for ($chip=1;$chip<13;$chip++) {
                $a2[$cs][$chip] = 0;
        }
}
}
sub report {
        for ($cs=0;$cs<$boards;$cs++) {
                for ($chip=1;$chip<13;$chip++) {
printf ("Board %1d Chip %2d Errors %d\n",($cs),($chip),$a2[$cs][$chip]);
                }
print "\n";       
}
}

emdje (OP)
Hero Member
*****
Offline Offline

Activity: 686
Merit: 500


View Profile WWW
January 05, 2016, 10:49:55 AM
 #417

May I ask which version software you guys are running? (because you seem to be running more than 8 chips on a board)

On a side note, personally I am running mine stably on 1440 MHz, and on average 17.7 Mh/s on nicehash per 8 chip board (I can run them up to 1500 MHz). If you guys are interested I can adapt the driver so that it should accept more chips and/or boards.
ZeroGee
Member
**
Offline Offline

Activity: 92
Merit: 10


View Profile
January 05, 2016, 06:21:58 PM
 #418

I'm running Anx/MarkAz's image using cgminer/3.9.0 in 12chip v0p5. I've modified a good bit of it to suit my desires in terms of graphs and information display, but the core software is the same. We have A2 110 boards. I've got V4 boards.

I would love to try a version of yours with an increased chip count and provide feedback/information.

For me 1320MHz seems to yield the highest accepted hash rate.

Diff: 2048
BA2 0 is running 1200MHz @ 0.885V
BA2 1-5 is running 1320MHz @ 0.930-0.935V

Code:
[2016-01-05 12:12:37] Started at [2016-01-04 14:47:08]                   
 [2016-01-05 12:12:37] Pool: stratum+tcp://mine.zpool.ca:3433                   
 [2016-01-05 12:12:37] Runtime: 21 hrs : 25 mins : 29 secs                   
 [2016-01-05 12:12:37] Average hashrate: 117.6 Megahash/s                   
 [2016-01-05 12:12:37] Solved blocks: 86                   
 [2016-01-05 12:12:37] Best share difficulty: 260M                   
 [2016-01-05 12:12:37] Share submissions: 61629                   
 [2016-01-05 12:12:37] Accepted shares: 61467                   
 [2016-01-05 12:12:37] Rejected shares: 162                   
 [2016-01-05 12:12:37] Accepted difficulty shares: 125884416                   
 [2016-01-05 12:12:37] Rejected difficulty shares: 331776                   
 [2016-01-05 12:12:37] Reject ratio: 0.3%                   
 [2016-01-05 12:12:37] Hardware errors: 6201                   
 [2016-01-05 12:12:37] Utility (accepted shares / min): 47.82/min                   
 [2016-01-05 12:12:37] Work Utility (diff1 shares solved / min): 98718.30/min
                   
 [2016-01-05 12:12:37] Stale submissions discarded due to new blocks: 333                   
 [2016-01-05 12:12:37] Unable to get work from server occasions: 4                   
 [2016-01-05 12:12:37] Work items generated locally: 1159802                   
 [2016-01-05 12:12:37] Submitting work remotely delay occasions: 1                   
 [2016-01-05 12:12:37] New blocks detected on network: 4337
                   
 [2016-01-05 12:12:37] Summary of per device statistics:
                   
 [2016-01-05 12:12:37] BA20                | (5s):18.58M (avg):18.09Mh/s | A:19900416 R:32768 HW:637 WU:1559                   
 [2016-01-05 12:12:37] BA21                | (5s):20.37M (avg):19.90Mh/s | A:20572160 R:73728 HW:1342 WU:161                   
 [2016-01-05 12:12:37] BA22                | (5s):20.43M (avg):19.89Mh/s | A:22257664 R:71680 HW:636 WU:1745                   
 [2016-01-05 12:12:37] BA23                | (5s):20.43M (avg):19.88Mh/s | A:20719616 R:53248 HW:1340 WU:162                   
 [2016-01-05 12:12:37] BA24                | (5s):20.44M (avg):19.89Mh/s | A:20803584 R:45056 HW:1153 WU:163                   
 [2016-01-05 12:12:37] BA25                | (5s):20.43M (avg):19.89Mh/s | A:21630976 R:55296 HW:1093 WU:169 

Errors per chip:
Board 1 2 3 4 5 6
Chip 1 375 253 191 591 514 293
Chip 2 104 157 93 304 121 233
Chip 3 7 32 45 29 126 91
Chip 4 20 155 45 89 95 74
Chip 5 1 49 355 144 43 152
Chip 6 84 173 46 77 66 153
Chip 7 43 638 86 105 294 212
Chip 8 100 43 76 260 149 115
Chip 9 163 159 67 182 200 67
Chip10 58 46 50 257 29 166
mjgraham
Full Member
***
Offline Offline

Activity: 188
Merit: 100


View Profile
January 06, 2016, 04:02:07 AM
 #419

I am using the one of your versions, from maybe v4 or the mini, to be honest I don't think the driver determines the chip count so much. Looking at the SPI traffic with the logic analyzer I see that the software asks for a chip count from the uC on the board it replies with a 0x000C which is 12 then it steps through them one at a time getting info from them, of course 6 and 12 return no data , the thing I find odd is chips 9 - 12 always return 0x14 for the temperature which makes me think they didn't modify the code on the uC on board when they went from the 8->10 chip versions or not that part of it, at least I think that is it, I know there are 4 chips that are 20C all the time.

I think a  lot of my issues in overclocking is heat, should have some heatsinks this week , they run around 55C right now I hope that will help some.

I have done a lot of thinking about more boards , should be no problem to do 8 (with 3 cs lines) on hardware but power and space is another issue, you looking at 150A+ at 12V that gets to be a lot while not impossible though.

ZeroGee
Member
**
Offline Offline

Activity: 92
Merit: 10


View Profile
January 06, 2016, 06:57:48 AM
Last edit: January 06, 2016, 07:12:55 PM by ZeroGee
 #420

The driver should have something to do with it at least. emdje's version supports 8 chips per board as a max in the current releases, according to him. If it's anything like the A1 source code, it should matter.

Huh. That's interesting. Taking a look at the datasheet it doesn't directly have a register for temperature, but has flags and configs for the flags. I figured that the temp readings are based on external sensors. This also implies that we don't really know what the die or junction temperature is. On the subject of temperature, I installed a air guide above the blade closest to the power supply, as it was 5C above the others, and 7C when overvolted/overclocked bringing it to 47-50C. Post guide installation it has brought it down considerably, to 41C.

It's very possible, though I have not compiled the data to find a direct hash/temp correlation. Though the datasheet does mention some recommendations. Just smack some fans on those suckers and I'd imagine it would do much better.

I almost purchased some dead 8 chip boards of various versions on ebay to expand it, hoping to get at least one working blade. If one could break out the last two cs lines and set the driver to 8 board maximum instead of 6 as it is now, I don't see why it isn't possible. In your case it's easily possible since you have FPGA's the controller board, given that the driver was rewritten. (Good job by the way. You should release the verilog/VHDL for that. I've got some FPGAs sitting around.)

You're looking at 240V power supply land. Maybe one of those 2880W supplies. I'm tempted to get one myself so I can drive all my miners on it. (Zeus Thunder x3, and A2 110MH/s)

Now that I've modified and overclocked the final board, my accepted hashrate is 111-119MH/s at 1320MH/s.  Grin
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 [21] 22 23 24 25 26 27 »
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!