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
DPoS how much better ? |
|
|
|
With FW 0.96 my numbers have gone from 520,525 and 540 to 450,500 and 520 Ghash. This is quite the decrease in performance.
I have been running them for about an hour so it might still be variance, anyone else getting similar results?
//DeaDTerra
are those numbers at the pool or reported back form cgminer ? CgMiner //DeaDTerra thanks anyway FeedbackLoop could be right: https://bitcointalk.org/index.php?topic=170332.msg3356843#msg3356843e.g. slush found 3 block in less than one hour I got a hold of Marcus and told him about my numbers, I told him to not start investigating it yet and that I would get back to him in 24 hours with my new numbers as it is possible that it is due to the flushwork as you guys said. //DeaDTerra perfect! really appriciated, maybe the'll finally fix it. edit1: since u have a direct way to contact Marcus would u mind to ask him what are the mining of the stickers they put on top os the PCBs (1,2,3,4,5) ? thanks |
|
|
|
With FW 0.96 my numbers have gone from 520,525 and 540 to 450,500 and 520 Ghash. This is quite the decrease in performance.
I have been running them for about an hour so it might still be variance, anyone else getting similar results?
//DeaDTerra
are those numbers at the pool or reported back form cgminer ? CgMiner //DeaDTerra thanks anyway FeedbackLoop could be right: https://bitcointalk.org/index.php?topic=170332.msg3356843#msg3356843e.g. slush found 3 blocks in less than one hour |
|
|
|
With FW 0.96 my numbers have gone from 520,525 and 540 to 450,500 and 520 Ghash. This is quite the decrease in performance.
I have been running them for about an hour so it might still be variance, anyone else getting similar results?
//DeaDTerra
are those numbers at the pool or reported back form cgminer ? |
|
|
|
HW backed to 10% with 0.96
care to elaborate? |
|
|
|
check it and let me know asap if core off
make it core on
as soon as possible anyway pZombie on kncminer irc channel @ freenode said that .96 resurect a completely dead ASIC slot I am at work and can't really get on IRC. What sticker number did he have on his dead board? Mine was "1" while the rest were "2". I've asked him but I lost connection while getting the response (cause I'm on a train now), I'll ask again edit: i said #5. it seems they could be related to production line rather than quality. I'll soon I get more ifno I'll share it here edit2: #5 on the faulty one and #1 on the other (he has a saturn) |
|
|
|
I'll give .96 a try on my troubled miner.. I have to keep bringing back VRMs from the dead (once a VRM shows < 1 amp output, the whole die is not used even though bertmod won't show them 'off') I can get it back but usually turns back off in 6-8 hours of hashing you mean the core come on and then come off again? after some hours?  jelly not the cores but the VRMs, they are the DC/DC converters you can see in your PCB |
|
|
|
check it and let me know asap if core off
make it core on
as soon as possible anyway pZombie on kncminer irc channel @ freenode said that .96 resurect a completely dead ASIC slot |
|
|
|
Can't say I remove browser cache files very often for firefox. Just clear the history? Who wants to volunteer to flash this firmware? I did the deed on .92 and .93 for folks. A bit hesitant to do .96 as i've had great luck with .95. I'm not home so luckily I get to watch others be the guinea pig. guys join the kncminer irc channel @ freenode there are already a bunch of brave miners that have flushed their box with the new firmware. pZombie have a dead module and 0.96 seems to fix it |
|
|
|
See you in the first impressions thread   edit: DeaDTerra is goign to test it and report back in kncminre irc channel, i'm travelling now but as soon as I get home I will try it for sure  Did you flash .96? No, but If I were home I would have done it in a breathe |
|
|
|
See you in the first impressions thread edit: DeaDTerra is goign to test it and report back in kncminre irc channel, i'm travelling now but as soon as I get home I will try it for sure |
|
|
|
Thank you KnC for fixing the issue with my hosted miner. I can now report that I am very happily hashing away at an acceptable rate and the folks at KnC have made me whole.
Brief shipping delays aside, KnC has acted in a completely professional manner and have exceeded my expectations. Even in dealing with rude and pushy customers.
Queue the replies from said rude and pushy customers in 1...2...
were they able to improve your hashrate without replacing faulty parts? from which level to which level ? edit: forgot to say that i'm really glad that knc sorted out your problem |
|
|
|
Hash Rate's going to be crazy! And wtf: 128 seconds 831.04 Gh/s Is your miner hosted @KnC datacenter? if yes maybe they are deploying a new fw, who knows? |
|
|
|
yes there is a sticker on each board with number 1, what does it mean?
The stickers 1, 2, 3 are from the assembly lines 1, 2, 3... Yes, there are 3 assembly lines doing this. sticker 1 mean that you have the best chip
that the reason you have that speed
So it is just number or really shows the hash speed of particular chip? Anyone knows for sure... ? I'm gathering info about that and till now the evidence confirms that the lower the number the higher the quality. I've measured number of enabled cores to evaluate chip quality. I've already posted on this forum instructions on how to it, anyway it boils to download bertmod firmware, the file despite the bin extension is a tar.gz file. Untar it. There u'll find among others a file named asic_status.pl, copy it into your mainers (scp), log into ur miner using ssh and execute perl ./asic_status.pl > statud.html Copy status.html somewhere and use a browser torender it. It will tell you what u need to know. E.g. my asic stamped with a #3 have 3 out 4 die full functioning whereas the last one has all cores disabled. The very same asic run a lot cooler (10 Celsius) than the others and consume a lot less energy. As soon as i will be able to use a proper device with a proper internet connection i'll post here the link to the step by step instructions. edit1: it come to my mind that yesterday on kncminer irc channel (freenode) someone told me his miner has #5 on all boards and in fact is jup hash @340 |
|
|
|
you're lucky man at the pool mine mint at ~490 since your case is open do you mind to check if your PCB boards have a sticker with a number on top of them? |
|
|
|
an you direct me how to unpack or mount the u-boot rootfs.cpio file? I believe I have found a method, but I'm out during the day so I ask if you have a quick solution to that. Then I will enable to explain A LOT! (Can't wait to get inside, knock! knock!)
did you already solve this problem or do you still need help? I was away, so, no I didn't mount the image yet. can you send me a PM with the method you used? No need to PM i think anbody will benefit from sharing knowledge assuming you're using a linux version of some sort. 0) download knc firmware (let say kncminer-0.95.bin) 1) tar xfz kncminer-0.95.bin 2) dd if=Angstrom-0.95.u-boot of=Angstrom-0.95.zip bs=64 skip=1 3) gunzip Angstrom-0.95.zip 4) cpio -i -F Angstrom-0.95 if you run 4) as normal user you'll get a bunch of errors due to mknode trying to create a few devices. edit1: useful links: http://www.isysop.com/unpacking-and-repacking-u-boot-uimage-files/http://www.linuxquestions.org/questions/linux-newbie-8/how-to-extract-a-ascii-cpio-archive-svr4-with-no-crc-4175436617/#post4827380Edit2: 2) dd if=Angstrom-0.95.u-boot of=Angstrom-0.95.gz bs=64 skip=1 3) gunzip Angstrom-0.95.gz |
|
|
|
I like the fact that bitcoin price is steadily climbing higher....I really like the fact that China's version of google, called BAIDU, is going to now start accepting bitcoin for some of its services....seems like China, India and Brazil are areas of blossoming growth for bitcoin.
it's not actually baidu but one of the many controlled companies (just acquired in Sept 2013). http://jiasule.baidu.com/it's a sort a cloudfare a good sign nonetheless |
|
|
|
|
Show Posts
