Topic: KnCMiner Jupiter Miner First Impressions - page 10. (Read 65993 times)

that's some terrible readings

did you look inside the case?   is everything tight and seem right?

Well my Jupiters a bit shit.

4 VRM Model.

Average Hash 0.94 and 0.95 410Gh/s

Average Hash 0.96 278Gh/s

Sent support and email with my asic_test log file that's added when upgrading to 0.96, looks like and I have some dodgy Chips lots of bad cores.

ASIC board #0 has 4 DC/DC
Found a KnC miner with 192 cores
KnC ASICs: good cores: 53 from 192
KnC ASICs: good cores after re-test: 83 from 192
KnC ASICs: good cores after re-test: 105 from 192
Good cores map:
 00 00 00 00 00 00 00 00 00 00 10 C3 3C FB DD EF FE FF EE F3 FF FE FF FF

ASIC board #2 has 4 DC/DC
Found a KnC miner with 192 cores
KnC ASICs: good cores: 85 from 192
KnC ASICs: good cores after re-test: 120 from 192
KnC ASICs: good cores after re-test: 133 from 192
Good cores map:
 EC B5 FE FD FE 3D 00 00 00 00 00 00 0C 32 C3 3C F0 CF 00 00 00 00 00 00

ASIC board #3 has 3 DC/DC
MEGADLYNX[0] error: Bad PMBUS revision 0xFFFFFFFF
I2C test failed!
Found a KnC miner with 192 cores
KnC ASICs: good cores: 57 from 192
KnC ASICs: good cores after re-test: 65 from 192
KnC ASICs: good cores after re-test: 74 from 192
Good cores map:
 00 00 00 00 00 00 6C B3 E3 FF F3 FF 3E BE EB 3E BF FF BE FB FF FF FF FF


ASIC board #4 has 4 DC/DC
Found a KnC miner with 192 cores
KnC ASICs: good cores: 176 from 192
KnC ASICs: good cores after re-test: 188 from 192
KnC ASICs: good cores after re-test: 188 from 192
Good cores map:
 00 00 00 00 03 0C 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

I have no idea - I've never seen one
(and I don't expect to ever see one either)

Odd though that they don't put all those extra stats in the API stats ... but instead you have to run some extra program.

Kano you think
That firmware of knc miners
Have any big bug or problem?

Hint ... W:0/0 ... bad ...

maybe I should have scrolled down the page a little 

No, need to edit manually. Just check this option : http://www.suxorz.com/images/2013/10/17/jKTWM.jpg

if your webUI 'mining stats' aren't displayed like below.



edit /config/cgminer.conf and replace false with true for the "api-listen" and "api-network" see below

wow, that is good to know.   I would think that anyone with full 8 VRM boards wouldn't bother going past .93 or .94 since those boards can handle it nicely.  Not sure if they will ever do specific tuning for 8 VRM boards (or even acknowledge they exist lol)

I saw .96 clean up two of my ASICs so they ran with no cores dropping, but I still had one board with lots of drops and that other board with the bad VRM that is either on with all cores or off with none (almost always off)

My VRM issue doesn't take the whole board down unless I run .93 or earlier firmwares, then FPGA errors are all over the place and make everything unusable.

Hello, Is there bertmod which is working fine with 0.96?

0.95 -> 0.96

8 VRM jupiter down from 530 GH -> 485 GH

sorry I'm leaving now, if can join #kncmienr @ freenode, his nickname is pZombie

with 0.96 he said that it has 3 die that works perfectly and the last one with all cores disabled

Can you ask him if his dead card was taking the whole system down?  I had to unplug the PCI-E from mine when I had it.

Better as in hashrate or just better because of the web portal showing more details?

DPoS how much better ?

So you still have core off but is less core off?
I am understand well that?

Dpos

adding to what sickpig posted, you can just copy out to /www/pages/ and hit the page there

 perl ./asic_status.pl > /www/pages/stat1.html


.96 does seem better than .95 but I haven't seen any miracles from it yet

I'll try a few full power cycles to see if all the VRMs/Cores can at least start fresh again

thanks.  It probably means nothing then.  My other Jupiter's have the same configuration...

1 2
2 2

And those run fine.

was able to run the asic status and .96 looks to have cleared up two good boards/chips so no cores at all off there

one board that drops cores alot is still doing it

one board that loves to show a low VRM now shows all the cores off (before it would think they were on)

here's the two problem boards
 
4 Temperature sensor: 37.0 C

Die ID Cores ON Cores OFF %
0 23 25 47.9
1 48 0 100
2 47 1 97.9
3 48 0 100
DC/DC ID ON/OFF Status Input Voltage Output Voltage Output Current
0 ON OK 12 V 0.739 V 23.6 A (17.4 W)
1 No DC/DC detected
2 ON OK 12.1 V 0.735 V 38.2 A (28.1 W)
3 No DC/DC detected
4 ON OK 12.1 V 0.754 V 39.2 A (29.6 W)
5 No DC/DC detected
6 No DC/DC detected
7 ON OK 12 V 0.757 V 39.7 A (30.1 W)
 
5 Temperature sensor: 43.5 C

Die ID Cores ON Cores OFF %
0 48 0 100
1 48 0 100
2 48 0 100
3 0 48 0
DC/DC ID ON/OFF Status Input Voltage Output Voltage Output Current
0 ON OK 12.1 V 0.737 V 39.8 A (29.3 W)
1 No DC/DC detected
2 ON OK 12.2 V 0.745 V 39.7 A (29.6 W)
3 No DC/DC detected
4 ON OK 12.2 V 0.744 V 37.2 A (27.7 W)
5 No DC/DC detected
6 No DC/DC detected
7 ON OK 12.1 V 0.735 V 0.188 A
 

they added some nice features to the mining tab but made the status page a brickwall
just a static page showing avg hash but never updates and no 'get status' button anymore

might have to try another browser

Edit-  IE not friendly with it but firefox is

PS- KNC, we are used to core/VRM data now, please include that if you do not want use running bertmod scripts on the miner directly - we still need to troubleshoot - thx

PSS- the update does restart again after 2 minutes of restarting it for the update