Swedish ASIC miner company kncminer.com

[Gyrsur]

They will be out of business VERY soon.

Vegas

and why?

[LongAndShort]

They will be out of business VERY soon.

Vegas

and why?

It is not even close to being a correct or accurate assumtion of kncs' future.

[zybc]

Yestarday we  have removed the Titan chip CAP Today will test it using chip water cooling, to check if the thermal grease was a problem..
https://i.imgur.com/EMrGqjDs.jpg

How did you remove the Titan chip cap?

We used BICs razor blade Cut a half of it to make it thinner and inserted it between the cap and chip to extract the glue (actually it seems to be a thermal grease)
As there are many elements around it took I think 3 hours. (can try makinng some drawings to understand the way)
The second one will be faster because we know how to and have som exp.
Now I am testing it without cap.
We placed indium foil instead of grease and put the cooling radiator on it. We set additional radiators on the aluminium clamp.
But as I see the temperature is not lower than on not modified boadrs but higher! CPU is higher 3-5 degree (Celsium) and DCDC 4-5 degree

[Mattzees]

Yestarday we  have removed the Titan chip CAP Today will test it using chip water cooling, to check if the thermal grease was a problem..

How did you remove the Titan chip cap?

We used BICs razor blade Cut a half of it to make it thinner and inserted it between the cap and chip to extract the glue (actually it seems to be a thermal grease)
As there are many elements around it took I think 3 hours. (can try makinng some drawings to understand the way)
The second one will be faster because we know how to and have som exp.
Now I am testing it without cap.
We placed indium foil instead of grease and put the cooling radiator on it. We set additional radiators on the aluminium clamp.
But as I see the temperature is not lower than on not modified boadrs but higher! CPU is higher 3-5 degree (Celsium) and DCDC 4-5 degree

Yes, the chip cap (heat spreader) is there for a reason.  You should put it back on.  I only asked because I was interested in replacing the thermal compound under the heat spreader.

That being said, I just performed an interesting heat sink modification, which worked OK, but only lowered the temps maybe 3C in total.  If I actually machined a new heat sink, instead of modifying the old one, it might work better.

[vegasguy]

When these miners actually work, they can be very profitable. Think I just broke my record solomining 21 blocks in 1 minute!! LOL!!

[zybc]

We used BICs razor blade Cut a half of it to make it thinner and inserted it between the cap and chip to extract the glue (actually it seems to be a thermal grease)
As there are many elements around it took I think 3 hours. (can try makinng some drawings to understand the way)
The second one will be faster because we know how to and have som exp.
Now I am testing it without cap.
We placed indium foil instead of grease and put the cooling radiator on it. We set additional radiators on the aluminium clamp.
But as I see the temperature is not lower than on not modified boadrs but higher! CPU is higher 3-5 degree (Celsium) and DCDC 4-5 degree

Yes, the chip cap (heat spreader) is there for a reason.  You should put it back on.  I only asked because I was interested in replacing the thermal compound under the heat spreader.

That being said, I just performed an interesting heat sink modification, which worked OK, but only lowered the temps maybe 3C in total.  If I actually machined a new heat sink, instead of modifying the old one, it might work better.

We have second CUBE that we modified. We just replaced the thermal grease under heat exchanger to Indium foil. The situation is the same - temperatures are higher! Actually I need to reduce Frequency to lower the temp of DCDC because it reach 95 degree..

[vegasguy]

You need to be careful.Your going to destroy that cube. Please wait for the video.

Vegas

[Mattzees]

When these miners actually work, they can be very profitable. Think I just broke my record solomining 21 blocks in 1 minute!! LOL!!

You're pumping a dead coin? 

And cross-posting about it?..
https://bitcointalk.org/index.php?topic=491058.msg12351947#msg12351947

[vegasguy]

Ya sorry, didnt know there were rules about mining a dead coin. Was there an addendum I dont know about? Can I get a list of whats ok to mine? I was thinking Casino coin , because Im from Las Vegas, but I need to get your ok on that first.

Vegas

[Mattzees]

I was thinking Casino coin , because Im from Las Vegas, but I need to get your ok on that first.

Vegas

You're OK to mine casino coin.  Go nuts.

[LongAndShort]

Ya sorry, didnt know there were rules about mining a dead coin. Was there an addendum I dont know about? Can I get a list of whats ok to mine? I was thinking Casino coin , because Im from Las Vegas, but I need to get your ok on that first.

Vegas

Pretty sure they asked a question, two even.

When these miners actually work, they can be very profitable. Think I just broke my record solomining 21 blocks in 1 minute!! LOL!!

~snip

You're pumping a dead coin?  

And cross-posting about it?..
https://bitcointalk.org/index.php?topic=491058.msg12351947#msg12351947

[zybc]

You need to be careful.Your going to destroy that cube. Please wait for the video.

Vegas

I would like to make two phase immersion cooling for it, but have no resources for it

But I do not see the problem to cool without cap if I place the radiator carefully.
The part of a problem actually I think is that the airflow and radiator are in different levels.
And also can't understand why the temp off DCDC-1 raises after my intrusion...

[zybc]

What about burning power socket?? How to avoid this effect?
I have -0.0513 Voltage. The Mhz were 300/300/325/325

[Mattzees]

What about burning power socket?? How to avoid this effect?
I have -0.0513 Voltage. The Mhz were 300/300/325/325

This thread is almost 2100 pages, so maybe this would be better discussed on a Titan hardware mod thread.  Until that happens...

I would think that if you're doing immersive cooling (especially two-phase) that would keep the connector from melting even if you're exceeding the power spec for the connector.  In the excessively hot spots, you have to watch out for cavitation (look it up... also look up "Chernobyl" while you're at it).

Regarding my recent heat sink mod, I used a 12-ton press to reshape the conduction plate on top of the copper heat sink so that it now has direct contact with 6 of the 8 VRMs.  The remaining 2 VRMs still use a thermal gap pad.  I used Arctic Ceramique between the heat spreader and the copper pipes, then between the conduction plate and the copper heat sink.

Because some of the VRMs are now in direct contact with the conduction plate, I also used Ceramique on those to create a good thermal bridge between the two.

My reward for all that is a 3C-degree temp reduction, which is rather disappointing.  I do believe that if I machined a new conduction plate to the proper shape (instead of bending the old one) I'd get a better result.

[GenTarkin]

GenTarkin/Vegaguy,

Awesome works you guys are doing here!  The overall Error Rate % is nice to see, but is it possible to break it down to Hardware % error rate per die like what it was done by the KnCMinion?  I found this information very helpful to find the lowest voltage setting for each die while keeping the HW % below 1%.  On some problematic die, I actually have to up the voltage and lower the clock frequency to keep HW Error % to an acceptable level.

https://github.com/p4fg/kncminion

{"STATUS":[{"STATUS":"S","When":1441579822,"Code":9,"Msg":"13 PGA(s)","Description":"bfgminer 5.2.0"}],"DEVS":[{"PGA":0,"Name":"KNC","ID":0,"ProcID":0,"Enabled":"Y","Status":"Alive","Device Elapsed":37548,"MHS av":11.558,"MHS 20s":12.863,"MHS rolling":12.863,"Accepted":202,"Rejected":0,"Hardware Errors":165,"Utility":0.323,"Stale":1,"Last Share Pool":0,"Last Share Time":1441579605,"Total MH":433976.2463,"Diff1 Work":101.43922169,"Work Utility":0.162,"Difficulty Accepted":101.16761480,"Difficulty Rejected":0,"Difficulty Stale":0.50195311,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579820,"Device Hardware%":0.6338,"Device Rejected%":0.0000},{"PGA":1,"Name":"KNC","ID":0,"ProcID":1,"Enabled":"Y","Status":"Alive","Device Elapsed":37559,"MHS av":18.602,"MHS 20s":21.531,"MHS rolling":21.531,"Accepted":311,"Rejected":1,"Hardware Errors":145,"Utility":0.497,"Stale":0,"Last Share Pool":0,"Last Share Time":1441579820,"Total MH":698653.6059,"Diff1 Work":163.30589201,"Work Utility":0.261,"Difficulty Accepted":155.19976348,"Difficulty Rejected":0.50195311,"Difficulty Stale":0,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579820,"Device Hardware%":0.3470,"Device Rejected%":0.3074},{"PGA":2,"Name":"KNC","ID":0,"ProcID":2,"Enabled":"Y","Status":"Alive","Device Elapsed":37559,"MHS av":19.561,"MHS 20s":13.142,"MHS rolling":13.142,"Accepted":330,"Rejected":1,"Hardware Errors":189,"Utility":0.527,"Stale":2,"Last Share Pool":0,"Last Share Time":1441579624,"Total MH":734674.2886,"Diff1 Work":171.72550035,"Work Utility":0.274,"Difficulty Accepted":164.96378601,"Difficulty Rejected":0.50195311,"Difficulty Stale":1.00390622,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579822,"Device Hardware%":0.4298,"Device Rejected%":0.2923},{"PGA":3,"Name":"KNC","ID":1,"ProcID":0,"Enabled":"Y","Status":"Alive","Device Elapsed":37559,"MHS av":20.385,"MHS 20s":24.047,"MHS rolling":24.047,"Accepted":365,"Rejected":3,"Hardware Errors":174,"Utility":0.583,"Stale":1,"Last Share Pool":0,"Last Share Time":1441579735,"Total MH":765645.0294,"Diff1 Work":178.96471647,"Work Utility":0.286,"Difficulty Accepted":181.85140454,"Difficulty Rejected":1.50585933,"Difficulty Stale":0.50195311,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579822,"Device Hardware%":0.3798,"Device Rejected%":0.8414},{"PGA":4,"Name":"KNC","ID":0,"ProcID":1,"Enabled":"Y","Status":"Alive","Device Elapsed":37559,"MHS av":18.436,"MHS 20s":22.090,"MHS rolling":22.090,"Accepted":316,"Rejected":1,"Hardware Errors":162,"Utility":0.505,"Stale":0,"Last Share Pool":0,"Last Share Time":1441579757,"Total MH":692446.0360,"Diff1 Work":161.85491153,"Work Utility":0.259,"Difficulty Accepted":157.48261559,"Difficulty Rejected":0.50195311,"Difficulty Stale":0,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579820,"Device Hardware%":0.3910,"Device Rejected%":0.3101},{"PGA":5,"Name":"KNC","ID":0,"ProcID":2,"Enabled":"Y","Status":"Alive","Device Elapsed":37559,"MHS av":20.151,"MHS 20s":20.133,"MHS rolling":20.133,"Accepted":376,"Rejected":2,"Hardware Errors":194,"Utility":0.601,"Stale":1,"Last Share Pool":0,"Last Share Time":1441579622,"Total MH":756853.7682,"Diff1 Work":176.90981438,"Work Utility":0.283,"Difficulty Accepted":188.05362907,"Difficulty Rejected":0.77699278,"Difficulty Stale":0.50195311,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579822,"Device Hardware%":0.4282,"Device Rejected%":0.4392},{"PGA":6,"Name":"KNC","ID":2,"ProcID":0,"Enabled":"Y","Status":"Alive","Device Elapsed":37559,"MHS av":20.258,"MHS 20s":22.370,"MHS rolling":22.370,"Accepted":323,"Rejected":2,"Hardware Errors":186,"Utility":0.516,"Stale":1,"Last Share Pool":0,"Last Share Time":1441579704,"Total MH":760880.3000,"Diff1 Work":177.85099091,"Work Utility":0.284,"Difficulty Accepted":160.31554703,"Difficulty Rejected":0.77699278,"Difficulty Stale":0.50195311,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579821,"Device Hardware%":0.4085,"Device Rejected%":0.4369},{"PGA":7,"Name":"KNC","ID":0,"ProcID":1,"Enabled":"Y","Status":"Alive","Device Elapsed":37559,"MHS av":20.148,"MHS 20s":21.810,"MHS rolling":21.810,"Accepted":320,"Rejected":1,"Hardware Errors":233,"Utility":0.511,"Stale":0,"Last Share Pool":0,"Last Share Time":1441579604,"Total MH":756736.3277,"Diff1 Work":176.88236340,"Work Utility":0.283,"Difficulty Accepted":160.39808179,"Difficulty Rejected":0.50195311,"Difficulty Stale":0,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579822,"Device Hardware%":0.5139,"Device Rejected%":0.2838},{"PGA":8,"Name":"KNC","ID":0,"ProcID":2,"Enabled":"Y","Status":"Alive","Device Elapsed":37559,"MHS av":20.170,"MHS 20s":21.810,"MHS rolling":21.810,"Accepted":361,"Rejected":1,"Hardware Errors":236,"Utility":0.577,"Stale":1,"Last Share Pool":0,"Last Share Time":1441579811,"Total MH":757558.4113,"Diff1 Work":177.07844184,"Work Utility":0.283,"Difficulty Accepted":180.07050554,"Difficulty Rejected":0.50195311,"Difficulty Stale":0.50195311,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579822,"Device Hardware%":0.5199,"Device Rejected%":0.2835},{"PGA":9,"Name":"KNC","ID":3,"ProcID":0,"Enabled":"Y","Status":"Alive","Device Elapsed":37559,"MHS av":20.248,"MHS 20s":20.972,"MHS rolling":20.972,"Accepted":350,"Rejected":0,"Hardware Errors":203,"Utility":0.559,"Stale":0,"Last Share Pool":0,"Last Share Time":1441579796,"Total MH":760477.6468,"Diff1 Work":177.75687326,"Work Utility":0.284,"Difficulty Accepted":173.64136756,"Difficulty Rejected":0,"Difficulty Stale":0,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579819,"Device Hardware%":0.4459,"Device Rejected%":0.0000},{"PGA":10,"Name":"KNC","ID":0,"ProcID":1,"Enabled":"Y","Status":"Alive","Device Elapsed":37559,"MHS av":20.242,"MHS 20s":22.929,"MHS rolling":22.929,"Accepted":324,"Rejected":0,"Hardware Errors":200,"Utility":0.518,"Stale":3,"Last Share Pool":0,"Last Share Time":1441579644,"Total MH":760276.3203,"Diff1 Work":177.71373600,"Work Utility":0.284,"Difficulty Accepted":161.49824047,"Difficulty Rejected":0,"Difficulty Stale":1.50585933,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579822,"Device Hardware%":0.4394,"Device Rejected%":0.0000},{"PGA":11,"Name":"KNC","ID":0,"ProcID":2,"Enabled":"Y","Status":"Alive","Device Elapsed":37559,"MHS av":20.175,"MHS 20s":14.820,"MHS rolling":14.820,"Accepted":354,"Rejected":0,"Hardware Errors":270,"Utility":0.566,"Stale":2,"Last Share Pool":0,"Last Share Time":1441579775,"Total MH":757742.9606,"Diff1 Work":177.11765753,"Work Utility":0.283,"Difficulty Accepted":176.10300689,"Difficulty Rejected":0,"Difficulty Stale":1.00390622,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579822,"Device Hardware%":0.5943,"Device Rejected%":0.0000},{"PGA":12,"Name":"KNC","ID":0,"ProcID":3,"Enabled":"Y","Status":"Alive","Device Elapsed":37559,"MHS av":20.056,"MHS 20s":20.412,"MHS rolling":20.412,"Accepted":348,"Rejected":2,"Hardware Errors":262,"Utility":0.556,"Stale":1,"Last Share Pool":0,"Last Share Time":1441579811,"Total MH":753263.4440,"Diff1 Work":176.07059865,"Work Utility":0.281,"Difficulty Accepted":173.31820167,"Difficulty Rejected":1.00390622,"Difficulty Stale":0.50195311,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579822,"Device Hardware%":0.5802,"Device Rejected%":0.5702}],"id":1}

Yeah, whats you do is ssh into the pi & do screen -r then "m"
That will show you the per die HW% error rate.

Im not really bent on putting that much info in the webgui, cuz eventually it will just get cluttered n ugly and have stuff not everyone needs.

[Wesly]

GenTarkin/Vegaguy,

Awesome works you guys are doing here!  The overall Error Rate % is nice to see, but is it possible to break it down to Hardware % error rate per die like what it was done by the KnCMinion?  I found this information very helpful to find the lowest voltage setting for each die while keeping the HW % below 1%.  On some problematic die, I actually have to up the voltage and lower the clock frequency to keep HW Error % to an acceptable level.

https://github.com/p4fg/kncminion

{"STATUS":[{"STATUS":"S","When":1441579822,"Code":9,"Msg":"13 PGA(s)","Description":"bfgminer 5.2.0"}],"DEVS":[{"PGA":0,"Name":"KNC","ID":0,"ProcID":0,"Enabled":"Y","Status":"Alive","Device Elapsed":37548,"MHS av":11.558,"MHS 20s":12.863,"MHS rolling":12.863,"Accepted":202,"Rejected":0,"Hardware Errors":165,"Utility":0.323,"Stale":1,"Last Share Pool":0,"Last Share Time":1441579605,"Total MH":433976.2463,"Diff1 Work":101.43922169,"Work Utility":0.162,"Difficulty Accepted":101.16761480,"Difficulty Rejected":0,"Difficulty Stale":0.50195311,"Last Share Difficulty":0.50195311,"Last Valid Work":1441579820,"Device Hardware%":0.6338,"Device Rejected%":0.0000},{"PGA":1,

Yeah, whats you do is ssh into the pi & do screen -r then "m"
That will show you the per die HW% error rate.

Im not really bent on putting that much info in the webgui, cuz eventually it will just get cluttered n ugly and have stuff not everyone needs.

I am aware of the "screen -r" command.  I used it on my Neptune to check the HW % for each individual die.  However for some reason, it doesn't work on any one of my Titan Raspberry pi controllers.  It always gives me the error "There is no screen to be resumed" message.

I don't meant to ask to display every single individual stats, I think the HW error % next to each die is helpful to have for optimizing the Titan since most dies can handle under-clocking to run cooler and use less power while maintaining less than 1% HW error rate.  Just a suggestion to try to make the Titan firmware mod even better.

root@raspberrypi:~# screen -r
There is no screen to be resumed.

root@raspberrypi:~# ps -ef | grep screen
root     26242  4568  0 04:55 pts/1    00:00:00 grep screen

root@raspberrypi:~# ps -ef | grep bfgminer
pi       18249     1  0 Sep07 ?        00:00:10 /usr/bin/SCREEN -S cgminer -t cgminer -m -d sh -c while true; do waas >/dev/null 2>/dev/null ; /home/pi/bfgminer/bfgminer --api-listen --api-allow=W:127.0.0.1/24 -c /config/cgminer.conf -Q 10 --scrypt -S titan:auto; done
pi       18250 18249  0 Sep07 pts/0    00:00:00 sh -c while true; do waas >/dev/null 2>/dev/null ; /home/pi/bfgminer/bfgminer --api-listen --api-allow=W:127.0.0.1/24 -c /config/cgminer.conf -Q 10 --scrypt -S titan:auto; done
pi       18529 18250 17 Sep07 pts/0    06:08:57 /home/pi/bfgminer/bfgminer --api-listen --api-allow=W:127.0.0.1/24 -c /config/cgminer.conf -Q 10 --scrypt -S titan:auto
root     26613  4568  0 04:55 pts/1    00:00:00 grep bfgminer

[mindtrip]

Need to login with user pi not root or it wont work

then do screen -r

[crackfoo]

For those looking for a reliable stable pool to park your KNC Titan's on we have a custom Titan port on www.xpool.ca

We've already done a great deal of testing and tweaking and currently the existing Titans are hitting a efficiency of +96% which is pretty good considering we're running a multipool.

Suggested config for the pool is:

{
"pools": [
{
"url": "mine.xpool.ca:1123/#xnsub#skipcbcheck",
"user": "coinaddress.workername",
"pass": "--diff 16384"
}

],
"scrypt-n": 10,
"failover-only": true,
"no-submit-stale": true
}

Valid coins to get paid in are BTC, LTC, NXT, DASH, BTCD, FIBRE and BITS.

Also with our custom account settings you can keep any coins you mine by setting each of your wallet addresses and saving the settings by using an encrypted message from your main coin/mining address.

You can view our support/info thread from the link in my sig.

Cheers!

[GenTarkin]

An update coming down the pipes for next release....
Fan failure protection w/ webgui output in big red letters when a potential fan failure has been detected!!!!

Up till now, my monitoring script has dynamically adjusted clocks up n down depending on environmental temperatures that increased / decreased gradually. It would attempt to clock the unit down as temps rose but this happened at the rate of KNC's monitoring script which is 1 poll per minute. A fan failure would easily go up in temperature much quicker than decreasing 25mhz ever 1 minute on respective dies would account for.
In the process of coding this I also found fault w/ KNC's waas command not actually being able to set a die OFF, rather bfgminer would have to be restarted and the advanced.conf file held a null value of any desired dies to be set as "OFF".
So... what Ive come up with is a small rewrite of setting dies to OFF logic(now all dies set to off while mining will invoke a bfgminer restart).
Next, onto the fan fail protection conditions...
1. If either of the DCDC temps of any die have detected as being above the user set threshold by 10C or more then the dies will be turn OFF.
OR
2. If either of the DCDC temps of any die have reached 115C or higher then that die will be set OFF.

After all dies which meet the condition above have been set to OFF, bfgminer will be issued a restart command.
A warning in big red letters on the STATUS page of webgui will warn about a possible fan failure detection.

To clear the warning the pi must be rebooted(I figured in the case of a genuine fan failure, the pi will be turned off along w/ the miner to service the cube w/ the failed fan).

*note.... if the 115C threshold ends up being an issue for anyone running their titans this hot well... A: your Titan will not last long B: I can always change this upward, but Im willing only to go up to 120C or so.

I will post when this release will be live.

[GenTarkin]

nother quick little feature going into next build:

Updated LCD display message =)
 

Got a question for yall, if a fan failure is detected, would it help anyone if I put a message on the LCD??? Like...... "FAN-FAIL DETECTED!"
I dont know how many people actually use the LCD =P