cognoscente
Member

Offline
Activity: 111
Merit: 10
|
 |
October 28, 2013, 05:48:18 PM |
|
The mods have gone full circle. Now it's back to factory build with just a bit of cardboard on top of the heatsinks. =:)
Back to bone stock on .96 here.
|
|
|
|
|
Phoenix1969
Legendary
Offline
Activity: 938
Merit: 1000
LIR DEV
|
 |
October 28, 2013, 05:50:50 PM |
|
The mods have gone full circle. Now it's back to factory build with just a bit of cardboard on top of the heatsinks. =:)
well, yes... lol... I'm convinced that most of the problems out there related to cores dropping/dies activating, is a simple contact problem. This could surely help as a solution, opposed to attempting to shorten spacers thru dis-assembly & modifying hardware, and risking screwing things up. this is a simple & quick way to cure it. I was faced with this problem on a saturn, which had a stripped screw from assembly, and couldn't really be fixed without tearing it apart. I was able to get some pliers in there to turn it a bit, but it didnt last for long, because the threads are stripped. The pad trick worked. Yet another Sat had similar woes, but no stripped screw. The pad trick worked there too
|
|
|
|
chrono030
Member

Offline
Activity: 114
Merit: 10
|
 |
October 28, 2013, 05:52:48 PM |
|
also... an observation... The reason a machine works better with the cover on: There are little black padded squares stuck to the top of the heatsinks. These pads contact the top of the case, and add slight pressure to the entire asic assembly on each PCB. It is enough to make a difference in how the asic performs. and the added temp with the case on, actually enhances it. taping a small piece of cardboard(1layer) to the top of the pad helps to increase that pressure just a tad. just my 2 cents
Thanks for the details Phoenix. I find it that if my temperatures are around 50 C, that the number of disabled cores skyrockets up. My unit seems to MUCH better with the case off & a fan on it. Does anyone else have the same experience? By the way, are you on .97 or .96?
|
|
|
|
|
Phoenix1969
Legendary
Offline
Activity: 938
Merit: 1000
LIR DEV
|
 |
October 28, 2013, 05:54:00 PM |
|
also... an observation... The reason a machine works better with the cover on: There are little black padded squares stuck to the top of the heatsinks. These pads contact the top of the case, and add slight pressure to the entire asic assembly on each PCB. It is enough to make a difference in how the asic performs. and the added temp with the case on, actually enhances it. taping a small piece of cardboard(1layer) to the top of the pad helps to increase that pressure just a tad. just my 2 cents
Thanks for the details Phoenix. I find it that if my temperatures are around 50 C, that the number of disabled cores skyrockets up. My unit seems to MUCH better with the case off & a fan on it. Does anyone else have the same experience? By the way, are you on .97 or .96? 0.97 I have a feeling you will find that the cores enable/disable from temps due to the contact issue. was for 2 of 3 of my sats. a simple way to test it is by watching the putty window while pressing on the heatsink pipes for about 5-10 seconds to see if any reaction from cgminer. That's how I located each poorly connected asic.
|
|
|
|
|
timmmers
|
 |
October 28, 2013, 05:55:48 PM |
|
whoa, i dont know what i just did, but i did some messing around and my mercury is now reporting 163ghs on btc guild! lets see how long this lasts, lol
Messing? Looks like good messing, extra 20GH over mine which has 1% HW errors.
|
|
|
|
Biffa
Legendary
Offline
Activity: 3234
Merit: 1225
|
 |
October 28, 2013, 06:00:00 PM |
|
Anyone noticed their miner has the date 1 Jan 2000?
|
|
|
|
Phoenix1969
Legendary
Offline
Activity: 938
Merit: 1000
LIR DEV
|
 |
October 28, 2013, 06:01:31 PM |
|
Anyone noticed their miner has the date 1 Jan 2000?
clear browser cache & make sure your pc time is correct.. had that one... hehe
|
|
|
|
|
timmmers
|
 |
October 28, 2013, 06:03:46 PM |
|
Just spoke with Anna and she told me my miners are ready and to be picked up by DHL today very shortly  That's tomorrow to you in the UK most likely. Mine took almost exactly a day. Good to hear. How late in the day did you get your email from DHL with the tracking number? IT was picked up t 1pm but the email and KNC status were later that same day. It hung around in Belgium for a good long time. I think the crucial part was it arriving in the Midlands at 3am. Got to me mid afternoon that day. IF your make that plane and get to the UK early am they'll manage to get to your local depot and on the van for tomorrow. Make sure you have the ethernet leads...that would be irritating to have to go get some late pm lol Glad you finally got some action. Hope your rigs run as nicely as mine has so far.
|
|
|
|
|
NetTime
|
 |
October 28, 2013, 06:04:22 PM |
|
Indeed. Glad I stayed away from that stuff. Why can't they just get Kano to tweak CGminer for the hardware?
|
|
|
|
|
Phoenix1969
Legendary
Offline
Activity: 938
Merit: 1000
LIR DEV
|
 |
October 28, 2013, 06:05:44 PM |
|
Indeed. Glad I stayed away from that stuff. Why can't they just get Kano to tweak CGminer for the hardware? Kano isn't the maker of cgminer... (kano is an assistant, and refused to help)cklovas is... and he released that.
|
|
|
|
|
idee2013
|
 |
October 28, 2013, 06:07:02 PM |
|
also... an observation... The reason a machine works better with the cover on: There are little black padded squares stuck to the top of the heatsinks. These pads contact the top of the case, and add slight pressure to the entire asic assembly on each PCB. It is enough to make a difference in how the asic performs. and the added temp with the case on, actually enhances it. taping a small piece of cardboard(1layer) to the top of the pad helps to increase that pressure just a tad. just my 2 cents
Thanks for the details Phoenix. I find it that if my temperatures are around 50 C, that the number of disabled cores skyrockets up. My unit seems to MUCH better with the case off & a fan on it. Does anyone else have the same experience? By the way, are you on .97 or .96? my jup likes it cool and my sat likes it warm...it is very different
|
|
|
|
|
the-skeptic
Member

Offline
Activity: 91
Merit: 10
|
 |
October 28, 2013, 06:40:25 PM |
|
OK, I am in Beijing right now trying to talk my brother in Santa Fe through setting up a bitchy Jupiter that worked when it arrived, and then lost one hashing board after updating to 0.9.7. What success have people have, if any, bringing a dead board back to life? The web interface shows the 3 working boards at 45C and the dead one at 20-25C. Did 0.9.7 lower the amperage so that a slightly flaky board might not come alive due to a lack of power? Kinda like when you can't jump start your car in the cold winter months because the thin, cheap ass jumper cables won't carry enough amps to crank over the engine?
Dealing with being in the Beijing time zone, and having to communicate with both the Santa Fe Mountain time zone and CEST Stockholm time means I am almost up 24 hours a day trying to sort this out, and I wanted to jump off the Great Wall during our visit their today. I'm so frustrated. I just spent $7000 on plug-n-play device that doesn't share and play well with other children. I already have an autistic 8 year old son that does that, if I wanted another one, I'd go fuck my wife. Instead I feel like I just got fucked by some dude in Stockholm.
Can someone give me the Cliff's notes on CGMiner 3.6.6? How difficult is it to install over putty? I understand my web interface will probably go to shit, but I don't care. I want my 550GHs machine, not this gimpy three board machine barely cranking at 385Ghs. Does 3.6.6 increase amperage so that maybe weak/high resistance connections will make have enough power to come alive? Is that why the cores on 3.6.6 are running at 70C instead of 45C like mine are now?
Orsoc promised a 0.9.8 today, which I suspect will include some of CKolivas' magic, but if that doesn't do anything, is there a community opinion that 3.6.6 might Frankenstein the bad board?
I have missed over 150 pages of forum posts this week, and trying to read them all over this crappy 64kb 3g wifi dongle thing they use here in the Beijing countryside is gonna make me nuckin futs.
Thanks for any advice, links or condolences ahead of time. This community rocks. You would think that if one guy figured out how to fix everything, he'd keep it to himself to keep that advantage against the rest of us struggling, but people here feel like we're all on the same boat and want to share the information so that we can all succeed. That's why I love you guys (and maybe some girls too!).
It's noon in Santa Fe, and 2 am here in Beijing and I just got off the phone with Santa Fe, after spending an hour reading the riot act to some guy at Orsoc. KNC took the phone off the hook, apparently. Skype just keeps repeating that it's busy.
There is a one year warranty on the miners, we should all take advantage of that and insist that if they can't solve the problems with v1, all pre order customers should get a free trade-in on v2 (if they make it that far).
Ok, it's now 2:30 AM and I am starting to get wobbly. Night all.
|
|
|
|
|
sickpig
Legendary
Offline
Activity: 1260
Merit: 1014
|
 |
October 28, 2013, 06:56:26 PM |
|
|
Bitcoin is a participatory system which ought to respect the right of self determinism of all of its users - Gregory Maxwell.
|
|
|
Phoenix1969
Legendary
Offline
Activity: 938
Merit: 1000
LIR DEV
|
 |
October 28, 2013, 07:05:43 PM |
|
 3 sats...0.97, straight up... they love 70c Are you pointing out the 128 and 256 second averages?. My 128 & 256 jump from 1600 to 3200 Gh/s constantly and I consider them almost useless. no, mostly the 3 and 12 hr... lol 3 sats @ 275 is 825 ... I'm getting that @ the pool, which is very good IMHO OOOoohhhHH!!! Can't wait for 0.98!!!
|
|
|
|
|
optimator
|
 |
October 28, 2013, 07:06:25 PM |
|
All,
We have noticed an increase in failures with our boxes when the firmware has been updated to 0.97.
There has been a bug identified in this firmware which can cause drop in performance in certain circumstances . There is a fix right now undergoing development but it’s not quite ready to release yet as it must undergo more testing overnight.
We are very aware that our customers are waiting for this hotfix to rescue some boxes who’s performance has dropped after applying 0.97.
The 0.98 release which is scheduled for release tomorrow will contain
1, A solution to address the bug which can cause dramatically reduced performance.
2, A bug fix which will stop the slowly reducing performance over time,
3, A new version of the CGMiner binary provided to us by the CGMiner developer’s
The update will be available tomorrow form our website,
SWEET!!!
|
|
|
|
ASIC-K
Sr. Member
  
Offline
Activity: 280
Merit: 250
Hell?
|
 |
October 28, 2013, 07:07:06 PM |
|
ok mercury is now running 141ghs stable! finallllly. this is on .94 firmware. lol
|
|
|
|
|
|
WastedLTC
|
 |
October 28, 2013, 07:08:04 PM |
|
 3 sats...0.97, straight up... they love 70c Are you pointing out the 128 and 256 second averages?. My 128 & 256 jump from 1600 to 3200 Gh/s constantly and I consider them almost useless. no, mostly the 3 and 12 hr... lol 3 sats @ 275 is 825 ... I'm getting that @ the pool, which is very good IMHO Gotcha, thnx
|
|
|
|
|
Biffa
Legendary
Offline
Activity: 3234
Merit: 1225
|
 |
October 28, 2013, 07:08:23 PM |
|
Anyone noticed their miner has the date 1 Jan 2000?
clear browser cache & make sure your pc time is correct.. had that one... hehe No mine has that date on the miner, if I ssh in.
|
|
|
|
|
AFox
|
 |
October 28, 2013, 07:11:11 PM |
|
All,
We have noticed an increase in failures with our boxes when the firmware has been updated to 0.97.
There has been a bug identified in this firmware which can cause drop in performance in certain circumstances . There is a fix right now undergoing development but it’s not quite ready to release yet as it must undergo more testing overnight.
We are very aware that our customers are waiting for this hotfix to rescue some boxes who’s performance has dropped after applying 0.97.
The 0.98 release which is scheduled for release tomorrow will contain
1, A solution to address the bug which can cause dramatically reduced performance.
2, A bug fix which will stop the slowly reducing performance over time,
3, A new version of the CGMiner binary provided to us by the CGMiner developer’s
The update will be available tomorrow form our website,
SWEET!!! Will it be a install/apply with no possibility to go back to the actual version of CGMiner ?
|
|
|
|
xyzzy099
Legendary
Offline
Activity: 1068
Merit: 1109
|
 |
October 28, 2013, 07:14:05 PM |
|
All,
We have noticed an increase in failures with our boxes when the firmware has been updated to 0.97.
There has been a bug identified in this firmware which can cause drop in performance in certain circumstances . There is a fix right now undergoing development but it’s not quite ready to release yet as it must undergo more testing overnight.
We are very aware that our customers are waiting for this hotfix to rescue some boxes who’s performance has dropped after applying 0.97.
The 0.98 release which is scheduled for release tomorrow will contain
1, A solution to address the bug which can cause dramatically reduced performance.
2, A bug fix which will stop the slowly reducing performance over time,
3, A new version of the CGMiner binary provided to us by the CGMiner developer’s
The update will be available tomorrow form our website,
SWEET!!! Will it be a install/apply with no possibility to go back to the actual version of CGMiner ? You can always re-install an older firmware image. I think this IS the actual version of CGMiner, though 
|
Libertarians: Diligently plotting to take over the world and leave you alone.
|
|
|
|