Topic: Swedish ASIC miner company kncminer.com - page 1231. (Read 3049528 times)

Did you try admin/admin or root/admin or both?

ATTENTION PEOPLE WITH DISABLING CORES OR HIGH ERROR RATES:

I got my Saturn a few days back. Didn't hear any noise from possibly loose fans or anything. Hooked it up, ran it for a few hours, it was underperforming a little bit and showed disabling cores on one die. Went to Walmart, bought a new fan and torx screwdrivers, opened the case and saw that one of the fans' ground was disconnected and fan wasn't spinning. Got scared of course and thought that my disabled cores were the result of that few hours of running without a fan (although temperatures never exceeded 60 degrees). Connected the ground, fan started spinning. Anyway, these cores disabling on one of the dies continued on and it was really bugging me. Tried enablecores, changing firmwares etc, nothing helped. Today I found enough courage to start messing with it a little bit deeper. Found out which board was disabling cores (and it was not the one which had fan disabled), changed slots, pushed down on vrm's as some suggested. Heard a click on two of them, and thought, that's it. Put everything back together but the core disabling continued. I was very short from starting messing with resistors on the back of the pcb behind the chip, when it came to me, so here what I suggest you try:
Disclaimer: I am not responsible for any loss of warranty or other issues as the result of doing this.
1. Find out which asic is giving you problems (just counting, for ex if 0-131 is disabled, then that's the one connected to the first slot)
2. Shutdown the miner, and then REMOVE the fan (but not the radiator) cooling that chip (unscrew and then it pops out, or just unplug the power to that fan), turn off your own cooling if you added any.
Turn on your miner.

One by one cores enabled and I have 0 disabled cores for about an hour now and hardware error rate is steadily falling. Temperature is at about 58 right now on that chip without the fan, so I reckoned it's not too bad. It appears that kncminers from cold Sweden like warm weather.

If this helps, donations are appreciated

Mercury does about that.

Trying to keep the faith in .97...  updated Jupiter 2 hours ago and KnC interface says 555 but pool is reading around 500 or a little less   : (   Not sure what WU means but it is at 7979.

From anyone's experience with .97, will the pool rate come up over time?

I ordered and paid july and got mine Yesterday.
Maybe they are assembling one type at a time in batches. Mercs, then back to Sats, then Jups?

downgrade to 0.96

deliver enough cooling

jala makes 0.3 a day? dont think so.

Hey - just as an update to earlier ...

Ended up getting a Coolermaster Silent Pro 1000w to replace the XFX Pro 1050w Black.

Popped in the exact same paperclip and it worked first try - been hashing at 554 stable for 4-5 hours now.

So, O'rama, I dunno .... maybe it was just that supply, or maybe it's an XFX thing, but I couldn't paperclip trick that particular supply.

Guy at the store had a thermaltake power supply tester, and it was giving all the right numbers on that, so ... I dunno.

That was my experience anyway.

Tried that XFX a few more times, jiggling the clip, jamming it in, just plain monkeying with it ... and I could never get it to spin up.

And again, thanks to the forum members for chipping in with the advice earlier. Nice to know it wasn't just me being puzzled.

I didn't do anything out of the ordinary - just configured it via the web UI, as always.

It has slowed down some over time though...  Looks like I am getting a sustained 245 GH/s at the moment.  WU is still high at 4136.

Holy, after an hour or two it is up to 286-7 steady at 4100 wu!
(And 2.2% errors now.)

Seems like the errors were all at the start and now it isn't generating nearly as many...interesting behaviour.

Looks like Hashfast is going to be a month late, which works to our advantage. Hopefully after this increase this will mean at least one difficulty increase that will be minor.

0.97 works great, getting 278gh/s avg on a Saturn, awesome!

It is coming at 2 days 400 diff

At a month Jupiter will be like jalapeño 0.30 per day

Using the web interface?

Or did you go in with ssh and fiddle with the cgminer timeouts and such?

(I have only been using the web interface as the ssh does not seem to like the username/password the web interface shipped with.)

-MarkM-

 

when did you order/pay?

0.97 on my Saturn has brought me up to 282gh and 4100 wu/m.

At a 3.4% HW error rate as well!

Quite happy I updated!

I have yet to see the flushwork occur hopefully it is recovering much faster now.

I updated by "bad" jupiter from 0.95 (with enable cores) to 0.97 about 12 hours ago

The WU increased from 69xx to 76xx-78xx, but the error rate is high 37% and the cgminer Avg is the same, despite the increased WU
 (5s):504.3G (avg):482.4Gh/s | A:2669800  R:20300  HW:1004701  WU:7644.5/m

The pool reported speed remains lower on 0.97, currently 405.04Gh/s

Mine have been running without any of them going offline.

I got lucky with mine - all 384 cores work all the time.

So it's official that 2 Saturns > 1 Jupiter (not by much, but still).