Swedish ASIC miner company kncminer.com

[GenTarkin]

OK ... go ahead and do a git pull, think I got the syntax nailed. It worked on my end for a fake hard reset & a live soft reset so... yeah

Also, I made it so in my loop it will attempt a soft reset every 60 seconds up to 7 times, if it fails all those it will do a hard reset at that point.
Now that my code is working, since that die of yours constantly shows 0 amps even with successful soft resets, it will succumb to my hard reset loop after it attempts 7 times. =(

ok I'll try it out

I can write a custom version for you that will tell the KNC loop to ignore the die u have clocked at 150mhz that way it snot constantly doing a hard reset on you every 20mins =) Let me know.

let me see how it goes, I might take you up on that

yeah, It lasted about 12 min before the hard reset, I wonder why it took 5 min before starting the resets?  maybe it initially has an amp reading??

maybe create a file that can be edited with a custom list of dies to ignore??

Alrighty, just pushed a commit for you. You can add dies in knc-asic/bypass_dies.conf file for any sort of resetting(bypasses the current checking in KNC's main loop).
If you wanted you can add the dies that you have flagged a "off" as well to the file =)
If you read the contents of that file, I think ull figure it out.
In your case it would be this below the line in the file:

5 1

[mishax1]

http://www.kncminer.com/blog/newsarchive

I guess they will find out how worthless the xapo card is...   hahahaha

How come?   Whats your point of view on Xapo ?    ( If I were a Greece resident I would disagree with you).

I, like most others, have been waiting over a year for their Xapo card to be delivered.
Feedback from the 100 or so beta testers is pretty harsh as well.
Most people who have actually received their card are reporting a long list of card rejections across the board... not working at all for the most part.
They talk a great game... but have nothing yet.

I've heard the same from other users. Some places(like gas stations and some stored rejected the card, in other stores it was accepted)  You do know that you need some BTC balance for the card to be clarified and delivered, right ?

[TXSteve]

OK ... go ahead and do a git pull, think I got the syntax nailed. It worked on my end for a fake hard reset & a live soft reset so... yeah

Also, I made it so in my loop it will attempt a soft reset every 60 seconds up to 7 times, if it fails all those it will do a hard reset at that point.
Now that my code is working, since that die of yours constantly shows 0 amps even with successful soft resets, it will succumb to my hard reset loop after it attempts 7 times. =(

ok I'll try it out

I can write a custom version for you that will tell the KNC loop to ignore the die u have clocked at 150mhz that way it snot constantly doing a hard reset on you every 20mins =) Let me know.

let me see how it goes, I might take you up on that

yeah, It lasted about 12 min before the hard reset, I wonder why it took 5 min before starting the resets?  maybe it initially has an amp reading??

maybe create a file that can be edited with a custom list of dies to ignore??

Alrighty, just pushed a commit for you. You can add dies in knc-asic/bypass_dies.conf file for any sort of resetting(bypasses the current checking in KNC's main loop).
If you wanted you can add the dies that you have flagged a "off" as well to the file =)
If you read the contents of that file, I think ull figure it out.
In your case it would be this below the line in the file:

5 1

ok just installed it and looking good, haven't even noticed any soft resets yet -- running about 20 min,  I think you got it 

[2015-08-15 00:52:57] --Start monitoring!--
DCDC temp monitoring ENABLED & threshold set to: 90
Manually disabled die detected, skipping dead die detection. ASIC# 4, DIE# 2
Manually disabled die detected, skipping dead die detection. ASIC# 4, DIE# 4
ASIC# 5 DIE# 1 manually bypassed for current check
Manually disabled die detected, skipping dead die detection. ASIC# 4, DIE# 2
Manually disabled die detected, skipping dead die detection. ASIC# 4, DIE# 4
ASIC# 5 DIE# 1 manually bypassed for current check
Manually disabled die detected, skipping dead die detection. ASIC# 4, DIE# 2
Manually disabled die detected, skipping dead die detection. ASIC# 4, DIE# 4
ASIC# 5 DIE# 1 manually bypassed for current check
Manually disabled die detected, skipping dead die detection. ASIC# 4, DIE# 2
Manually disabled die detected, skipping dead die detection. ASIC# 4, DIE# 4
ASIC# 5 DIE# 1 manually bypassed for current check

[TXSteve]

GenTarkin I think we got that problem fixed, now I have something else for you to look at  :|

in this Bfgminer 5.2 I am getting a ton of stale share errors when using load balancing, it's like the buffer isn't getting properly flushed after a new block is found. Eventually it clears out and starts submitting valid shares. And it usually just effect 1 or 2 pools that are being load balanced. btw these pools run great when not load balanced

I reverted back to 5.1 and these issues went away

I realize that these are probably BFgminer bugs, but since 5.2 is part of the upgrade I'm throwing it out there on the off chance there is anything you can do about it

[TXSteve]

genTarkin someone selling your work on ebay  ---  hope it's you ....  vegaslaptoppartsbuyer

"personally tested this version on multiple sets of titans for weeks without issue. You will also get access to the latest versions that come out about once a week, free for life!"

http://www.ebay.com/itm/KncMiner-Titan-custom-modified-firmware-with-lifetime-updates-for-Batch-1-2-/281773449397?hash=item419b01a0b5

Kind of wondering what this is:  rofl, like they read my prior post and threw it in there

"Loadbalancing not fully implemented yet, it does function 100%.  but you will need to add 1 small line manually. I will give you this information to you if you want it, after payment."

[Prelude]

genTarkin someone selling your work on ebay  ---  hope it's you ....  vegaslaptoppartsbuyer

"personally tested this version on multiple sets of titans for weeks without issue. You will also get access to the latest versions that come out about once a week, free for life!"

http://www.ebay.com/itm/KncMiner-Titan-custom-modified-firmware-with-lifetime-updates-for-Batch-1-2-/281773449397?hash=item419b01a0b5

Kind of wondering what this is:  rofl, like they read my prior post and threw it in there

"Loadbalancing not fully implemented yet, it does function 100%.  but you will need to add 1 small line manually. I will give you this information to you if you want it, after payment."

Man, people are seriously shitty. A big "go fuck yourself, cunt" to whoever is trying to sell the firmware. Looks like it could be the poster from vegas earlier in this thread.

[TXSteve]

genTarkin someone selling your work on ebay  ---  hope it's you ....  vegaslaptoppartsbuyer

"personally tested this version on multiple sets of titans for weeks without issue. You will also get access to the latest versions that come out about once a week, free for life!"

http://www.ebay.com/itm/KncMiner-Titan-custom-modified-firmware-with-lifetime-updates-for-Batch-1-2-/281773449397?hash=item419b01a0b5

Kind of wondering what this is:  rofl, like they read my prior post and threw it in there

"Loadbalancing not fully implemented yet, it does function 100%.  but you will need to add 1 small line manually. I will give you this information to you if you want it, after payment."

Man, people are seriously shitty. A big "go fuck yourself, cunt" to whoever is trying to sell the firmware. Looks like it could be the poster from vegas earlier in this thread.

not real familiar with github, but what's the license terms? Isn't it open source? Don't think you're allowed to sell open source, but I am just speculating -- I am sure someone knows

[TheRealSteve]

not real familiar with github, but what's the license terms?

GenTarkin hasn't set a license.  He'd also have to derive it from KnC's original license anyway.

Isn't it open source? Don't think you're allowed to sell open source, but I am just speculating -- I am sure someone knows

For most open source licenses, there's absolutely nothing disallowing the sale of software.  I could take LibreOffice and start selling it on DVD through e-bay as a service, for example, and charge $50/DVD.  The GPL in particular allows you to sell software for whatever price you like (it's simply not specified).  The GPL also requires you to make the source code available to any recipient of the software who asks for the source, however, and while you can charge for making that source code available, there's a limit on the amount you can set that comes down to only covering e.g. shipment costs.

What you're legally allowed (or at least not disallowed) to do and what is morally right are two different things, though

[mindtrip]

can't believe someone is selling what he has done and posted here for free, people on ebay never cease to amaze me

[vegasguy]

Guys relax. It is me, and I actually asked permission from GenTarkin in advance, and he gave me expressed permission.  So nothing shady here. I promised to split the profits with GenTarkin from all of the sales. You guys can monitor ebay and he will get a cut of every single sale. It will also be for sale on my website. Everything is on the up and up, and I personally discussed this in advance with GenTarkin. This is NOT my work, and so I asked his permission FIRST, and promised to split profits, and he gave the ok. Its a great opportunity were we can both make a bit of money, and that has been going on for centuries. As for my experience. I own multiple computer repair shops for 20 years in Las Vegas and I stopped working at my shops and began working in crypto full time in 2013. I worked in crypto 7 days a week, and about 12-14 hours a day. So Ill correct my post to read about 8,000 hours in crypto. That would be more accurate. Thank you for pointing that out. It FELT like 50,000 hours. LOL.

Actually its a good thing that people are protective of good crypto. I like that, as I am the same way, and probably would have reacted the same , had I saw someone selling GenTarksins work and didnt know permission was granted. It probably would have been a good idea to post here and let everyone know whats going on. Sorry guys, I failed to do that. For the record, I would NEVER sell someones work unless permission was granted first, even if it is open source. Ethically its the right thing to do, and if GenTarkin would have said no, then I would not have disrespected him and listed it.

Thanks

Vegas

[TXSteve]

well, as long as GenTarkin is onboard I hope you sell a lot so he can continue development

[GenTarkin]

Yeah, guys its ok if he sells it, It brings more publicity I guess to make people aware that such firmware exists. If they want to seek out the free version of it then so be it. I dont know much about licensing crap either so... w/e.
So yeah, he has my permission as long as a cut is coming to me =)

Ive never actually written software people are willing to pay for so, this is all new to me. ROFL!

Far as my latest work goes, theres been a lot done since my latest release. Will be making a new release soon once all the changes are tested thorougly, I may make it a non prerelease as well if all goes well =)

[GenTarkin]

GenTarkin I think we got that problem fixed, now I have something else for you to look at  :|

in this Bfgminer 5.2 I am getting a ton of stale share errors when using load balancing, it's like the buffer isn't getting properly flushed after a new block is found. Eventually it clears out and starts submitting valid shares. And it usually just effect 1 or 2 pools that are being load balanced. btw these pools run great when not load balanced

I reverted back to 5.1 and these issues went away

I realize that these are probably BFgminer bugs, but since 5.2 is part of the upgrade I'm throwing it out there on the off chance there is anything you can do about it

Well, I could revert to 5.1 in the build, but prefer to stay with the latest. Since that seems what people want.

[TXSteve]

GenTarkin I think we got that problem fixed, now I have something else for you to look at  :|

in this Bfgminer 5.2 I am getting a ton of stale share errors when using load balancing, it's like the buffer isn't getting properly flushed after a new block is found. Eventually it clears out and starts submitting valid shares. And it usually just effect 1 or 2 pools that are being load balanced. btw these pools run great when not load balanced

I reverted back to 5.1 and these issues went away

I realize that these are probably BFgminer bugs, but since 5.2 is part of the upgrade I'm throwing it out there on the off chance there is anything you can do about it

Well, I could revert to 5.1 in the build, but prefer to stay with the latest. Since that seems what people want.

yeah I prefer the latest too, I didn't see anything in the changelog re load balance, so I'll play around with it some more

[TXSteve]

GenTarkin I think we got that problem fixed, now I have something else for you to look at  :|

in this Bfgminer 5.2 I am getting a ton of stale share errors when using load balancing, it's like the buffer isn't getting properly flushed after a new block is found. Eventually it clears out and starts submitting valid shares. And it usually just effect 1 or 2 pools that are being load balanced. btw these pools run great when not load balanced

I reverted back to 5.1 and these issues went away

I realize that these are probably BFgminer bugs, but since 5.2 is part of the upgrade I'm throwing it out there on the off chance there is anything you can do about it

Well, I could revert to 5.1 in the build, but prefer to stay with the latest. Since that seems what people want.

yeah I prefer the latest too, I didn't see anything in the changelog re load balance, so I'll play around with it some more

I set up load balance on a 2nd rig and imported the same pools and settings

this is bfgminer 5.1:
 [2015-08-15 19:34:24] Rejected 000264b5 KNC 0d pool 2 Diff 417m/72m (Job not found)
 [2015-08-15 19:34:26] Accepted 0003de13 KNC 0c pool 4 Diff 258m/125m
 [2015-08-15 19:34:30] Accepted 0009b4c9 KNC 0b pool 2 Diff 103m/72m
 [2015-08-15 19:34:30] Accepted 000374cf KNC 2d pool 5 Diff 289m/125m
 [2015-08-15 19:34:31] Accepted 0003d80c KNC 1d pool 2 Diff 260m/72m
 [2015-08-15 19:34:32] Accepted 0001f17b KNC 2c pool 4 Diff 514m/125m
 [2015-08-15 19:34:32] Accepted 00061ef2 KNC 0a pool 2 Diff 163m/72m
 [2015-08-15 19:34:33] Accepted 00005ee4 KNC 0c pool 4 Diff 2/125m
 [2015-08-15 19:34:35] Accepted 00082b2e KNC 1d pool 2 Diff 122m/72m
 [2015-08-15 19:34:39] Accepted 000a73e2 KNC 0b pool 2 Diff 95m/72m
 [2015-08-15 19:34:42] Accepted 00008e9f KNC 0d pool 4 Diff 1/125m
 [2015-08-15 19:34:46] Accepted 000288d7 KNC 2d pool 5 Diff 394m/125m
 [2015-08-15 19:34:46] Accepted 0006248f KNC 0d pool 4 Diff 162m/125m
 [2015-08-15 19:34:47] Accepted 000034e7 KNC 1a pool 5 Diff 4/125m
 [2015-08-15 19:34:47] Accepted 000030b2 KNC 1c pool 4 Diff 5/125m
 [2015-08-15 19:34:50] Accepted 00015eff KNC 1b pool 5 Diff 729m/125m

this is bfgminer 5.2:
[2015-08-15 19:34:58] Pool 3 stale share detected, submitting as user requested
 [2015-08-15 19:34:59] Rejected 0007a5d4 KNC 1b pool 3 Diff 130m/125m (prevhash-stale)
 [2015-08-15 19:34:59] Rejected 003027c6 KNC 0d pool 2 Diff 20m/9m (Job not found)
 [2015-08-15 19:34:59] Pool 2 stale share detected, submitting as user requested
 [2015-08-15 19:34:59] Rejected 005145fc KNC 0c pool 2 Diff 12m/9m (Job not found)
 [2015-08-15 19:35:00] Pool 4 stale share detected, submitting as user requested
 [2015-08-15 19:35:00] Rejected 0001aac8 KNC 2c pool 4 Diff 599m/125m (prevhash-stale)
 [2015-08-15 19:35:00] Pool 2 stale share detected, submitting as user requested
 [2015-08-15 19:35:00] Rejected 003f49d1 KNC 0c pool 2 Diff 15m/9m (Job not found)
 [2015-08-15 19:35:01] Pool 2 stale share detected, submitting as user requested
 [2015-08-15 19:35:01] Pool 2 stale share detected, submitting as user requested
 [2015-08-15 19:35:02] Rejected 006018b6 KNC 0c pool 2 Diff 10m/9m (Job not found)
 [2015-08-15 19:35:02] Pool 2 stale share detected, submitting as user requested
 [2015-08-15 19:35:02] Rejected 00639377 KNC 0c pool 2 Diff 10m/9m (Job not found)
 [2015-08-15 19:35:02] Rejected 0029d73b KNC 0c pool 2 Diff 23m/9m (Job not found)
 [2015-08-15 19:35:02] Pool 2 stale share detected, submitting as user requested
 [2015-08-15 19:35:02] Pool 2 stale share detected, submitting as user requested
 [2015-08-15 19:35:02] Rejected 00656354 KNC 0d pool 2 Diff 9m/9m (Job not found)
 [2015-08-15 19:35:03] Rejected 001aed83 KNC 0d pool 2 Diff 37m/9m (Job not found)
 [2015-08-15 19:35:03] Pool 2 stale share detected, submitting as user requested
 [2015-08-15 19:35:03] Rejected 005594e0 KNC 0c pool 2 Diff 11m/9m (Job not found)
 [2015-08-15 19:35:05] Pool 2 stale share detected, submitting as user requested
 [2015-08-15 19:35:05] Rejected 0005e63b KNC 0c pool 2 Diff 169m/9m (Job not found)
 [2015-08-15 19:35:06] Accepted 0000ebe6 KNC 1d pool 4 Diff 1/125m
 [2015-08-15 19:35:09] Accepted 00051c4f KNC 0d pool 5 Diff 195m/125m
 [2015-08-15 19:35:11] Accepted 0007f9be KNC 0a pool 4 Diff 125m/125m

[GenTarkin]

If I had to guess its because the vardiff is set so low on that one pool, look its 9m when comparing it to the higher vardiff value in 5.1 Seems to be in pool 2 ...

[jelin1984]

what diff per worker per titan must be put?

[TXSteve]

If I had to guess its because the vardiff is set so low on that one pool, look its 9m when comparing it to the higher vardiff value in 5.1 Seems to be in pool 2 ...

pool 2 is using a ridiculously slow vardif -- because of the rejects it never gets up to speed, and constantly cycles from  the lowest vardif on up, that's why I put 5.1 back in

also it's effecting them all, we are seeing mostly pool 2 because it is a small sample and the vardif is low so more blocks submitted in the sample, the other pools are set at 8192, but they still get a surge of rejects

5.2 rejects run around 20%, 5.1 around 2-3%, failover rigs run around .5%,  there is some performance overhead when using load balance but 20% is pretty high

I'll ask in the bfgminer thread later, see what they say -- maybe they know of an issue with it

thx

[TheRealSteve]

meanwhile:

Stockholm, Sweden / Taipei, Taiwan – August 14, 2015- KnCMiner AB, the leading provider of Bitcoin mining machine, and Alchip Technologies, Ltd. (TWSE: 3661), a leading fabless ASIC company, announced that they have cooperated in designing and manufacturing a 16nm FinFET AISC for KnCMiner’s latest Solar bitcoin mining machine. The chip samples were delivered in May and started mass production in July.

[vegasguy]

meanwhile:

Stockholm, Sweden / Taipei, Taiwan – August 14, 2015- KnCMiner AB, the leading provider of Bitcoin mining machine, and Alchip Technologies, Ltd. (TWSE: 3661), a leading fabless ASIC company, announced that they have cooperated in designing and manufacturing a 16nm FinFET AISC for KnCMiner’s latest Solar bitcoin mining machine. The chip samples were delivered in May and started mass production in July.

And anyone that sends money to those liars deserves what they get.

Vegas