Swedish ASIC miner company kncminer.com

[DeathAndTaxes]

I'm not dodging anything. They were an excessive margin that is of no use now, why would you not cut back on unnecessary components, when proven unrequited to perform optimal? Makes no sense. Why not add 4 extra wheels to your car because you can?!

The car was designed to use 4 wheels.  Adding 4 wheels or removing 3 of them IN production would be equally stupid.  Glad you agree.

The time for version 2 is once orders have shipped and the new CHEAPER (which benefits KNC not customers) 4 VRM version is extensively tested.   Not on the fly right in the middle of production.  You know that.  There is absolutely no reason to switch in production.  Nobody is saying 8 VRMs today, 8 VRMS tomorrow 8 VRMS forevers but the board was designed, simulated, and tested one way.  Then at the last minute the margins upon margins where chopped off with minimal testing.   Oops it turned out to be a bad idea.

[ElGabo]

And O'rama!

Thaks a lot for your good work.

I love you fuckin much

[Paladin69]

My Jupiter is scheduled for delivery today. Of course I bought a Corsair HX850 - is there any *safe* way to use it, or should I not even plug the thing in?

What if I power on the HX850 and plug in the PCI-E connectors one at a time?

Don't plug it in.  Go buy something else at your local electronics store.  It will pay for itself in 24 hours.  You do NOT want a downed machine currently.

Also - .95 seems to be working very well for me:

 (5s):585.5G (avg):556.7Gh/s | A:65024  R:512  HW:3533  WU:7784.1/m

Did I miss something?  What is wrong with his PSU?  This thread is moving too fast.

[Phoenix1969]

Then why not leave the 8 VRMs on for that point??

why are you dodging that?  what was the main gain for everyone on taking off the 4 VRMs??   You could keep shipping with 8 until you tested taking off 4.. .how can you try to keep avoiding this logic?

Simple.  KNC chose to use an off the shelf power module.  This means rapid prototyping but it also means higher cost.  Those modules run $25 each.  8 per board means $200 per board in cost just for the power supply.  That is $800 per Jupiter.  If they leave 4 of them off that is an extra $400 in profit per unit.  1500 units @ $400 ea = $600,000.

There is absolutely no other reason to change the extensively designed, tested and fully operational boards right in the middle of peak production.

manufacturing time... takes less time to mount 4 converters than 8....times that by about ten thousand boards....

[mjsherman]

Just wanted to echo what everyone else is saying as well.  New 0.95 firmware is a huge upgrade.

My Saturn was running at ~210 Gh/s for the last few days, 10-12% hardware errors, 70C temp.  New firmware its running at ~300 Gh/s (only been 30 min), less than 1% HW errors, and 45-50C temp.

[rrrix]

My Jupiter is scheduled for delivery today. Of course I bought a Corsair HX850 - is there any *safe* way to use it, or should I not even plug the thing in?

What if I power on the HX850 and plug in the PCI-E connectors one at a time?

JESUS

I have absolutely no idea how people with this amount of experience find themselves buying these devices.  Really?

Literally in awe when I'm reading some of the questions in here.
[/quote]

brb, gonna go study electrical engineering for a few years and then ask a question that by then I'll already know the answer to. Thanks for clarifying, at least I asked first, instead of frying my stuff.

[DeathAndTaxes]

Did I miss something?  What is wrong with his PSU?  This thread is moving too fast.

KNC is claiming that Corsair (one of the most trusted enthusiast brands) "ramps up current too quickly" and that is why a mining board (powered by Corsair PSU) blew a capacitor.  Of course the fact that a high current switching supply SHOULD operate that way should be ignored.  The obvious explanation is that Corsair doesn't know how to build power supplies.  

[The Avenger]

Then why not leave the 8 VRMs on for that point??

why are you dodging that?  what was the main gain for everyone on taking off the 4 VRMs??   You could keep shipping with 8 until you tested taking off 4.. .how can you try to keep avoiding this logic?

Simple.  KNC chose to use an off the shelf power module.  This means rapid prototyping but it also means higher cost.  Those modules run $25 each.  8 per board means $200 per board in cost just for the power supply.  That is $800 per Jupiter.  If they leave 4 of them off that is an extra $400 in profit per unit.  1500 units @ $400 ea = $600,000.

There is absolutely no other reason to change the extensively designed, tested and fully operational boards right in the middle of peak production.

Ding! Hits the nail directly on the head. Spot on DeathAndTaxes.

[Transam808]

Because you wanted boards hashing ASAP, and got them.

Then why not leave the 8 VRMs on for that point??

why are you dodging that?  what was the main gain for everyone on taking off the 4 VRMs??   You could keep shipping with 8 until you tested taking off 4.. .how can you try to keep avoiding this logic?

I'm not dodging anything. They were an excessive margin that is of no use now, why would you not cut back on unnecessary components, when proven unrequited to perform optimal? Makes no sense. Why not add 4 extra wheels to your car because you can?!

All 8 might of not been needed, but at least get the firmware fished and working right with the 4 VRMs before mass producing them and sending them to customers, they could of kept sending out the 8 VRMs boards tell they had stable firmware for the 4, just my opinion

[The Avenger]

Then why not leave the 8 VRMs on for that point??

why are you dodging that?  what was the main gain for everyone on taking off the 4 VRMs??   You could keep shipping with 8 until you tested taking off 4.. .how can you try to keep avoiding this logic?

Simple.  KNC chose to use an off the shelf power module.  This means rapid prototyping but it also means higher cost.  Those modules run $25 each.  8 per board means $200 per board in cost just for the power supply.  That is $800 per Jupiter.  If they leave 4 of them off that is an extra $400 in profit per unit.  1500 units @ $400 ea = $600,000.

There is absolutely no other reason to change the extensively designed, tested and fully operational boards right in the middle of peak production.

manufacturing time... takes less time to mount 4 converters than 8....times that by a few thousand units.....

Have you seen this video? Takes seconds http://www.youtube.com/watch?v=vFfAHjSmlZM

Time is no excuse here

[Mota]

Because you wanted boards hashing ASAP, and got them.

Sorry, Orama. You know that I am pretty much objective concerning facts and stuff, but they fucked up the shipping, easy as that (among other stuff). They even admitted it in the last post. Well, email with the monition is out and we will see if they finally give me a satisfying reply or if I am on the forever ignore list.

[Paladin69]

Did I miss something?  What is wrong with his PSU?  This thread is moving too fast.

KNC is claiming that Corsair (one of the best brands in power supply industry) allows their boards "ramps up current" too quickly and that is why a board blew using that PSU.  Of course the fact that a high current switching supply SHOULD operate that way makes the "reason" suspect but Corsair obviously doesn't know how to build power supplies. 

Is there a "safe" list?  Never heard of a PSU being blamed before...

[jimmy3dita]

Kudos to Knc and the latest firmware.

No "flu" anymore on my Jupiter, it's working at 550W, with low temps and very high performance (550GH/s). Almost no errors.

I'll post an update tomorrow to see the results on the long term.

[Vega]

I'm running 0.93. Just tried to upgrade to 0.95, and it seems to work, but after restart the web interface still says that the current firmware revision is 0.93. I tried to upgrade (downgrade) to other versions but it's the same.

It is possible that the upgrade was successful and the web interface displays the wrong version number? Is there a way to check firmware version through ssh?

[DPoS]

Then why not leave the 8 VRMs on for that point??

why are you dodging that?  what was the main gain for everyone on taking off the 4 VRMs??   You could keep shipping with 8 until you tested taking off 4.. .how can you try to keep avoiding this logic?

Simple.  KNC chose to use an off the shelf power module.  This means rapid prototyping but it also means higher cost.  Those modules run $25 each.  8 per board means $200 per board in cost just for the power supply.  That is $800 per Jupiter.  If they leave 4 of them off that is an extra $400 in profit per unit.  1500 units @ $400 ea = $600,000.

There is absolutely no other reason to change the extensively designed, tested and fully operational boards right in the middle of peak production.

+1

I too say fine with removing 4 VRMs     **AFTER**  testing

it would take 1 minute to see the wattage increase 70 PER BOARD and heat rise 10C PER BOARD

[Bitcoinorama]

We all bought an unrefined product did we? Odd, I didn't see those words or anything like that when I paid for mine.
Not in any of the back patting videos or interviews either.

We bought a miner that we were told confidently would have margins upon margins capable of 400Gh and even today no-one has said that this (and the oct 15th date) has changed. Except of course to up it to 550 on the site and most promos ...which currently is false advertising ..as the past 20 pages here bear witness to.

As for all this shite about power consumption...right now is that a big priority to get down? I'd rather have a hashing rig while it's making enough BTC to make a few quid power bill extra irrelevent..later on the power consumption will matter more but not initially compared to a working rig. As long as the PSU handles it, we can wait for a firmware fix to get it down. Hashing reliably without smoke is the priority for users.

The sarcastic tone of Oramas post amazed me, people using fans to reduce temps ..planking? It worked. How bizarre is that? More to the fucking point no-one was talking from KNC end despite much asking...so what else were people to do? "We have a specialist flying in and we'll get back to you as soon as we know something."  < see, took me 3 seconds to type that, and even some poor sweaty overworked KnC human could manage that surely?

The way to deal with trainwrecks like this isn't arrogance, sarcasm and silence ..it's communication and apologies and in many cases here where buyers have been so badly let down with multi hosted jupiters...fair compensation.

I wont deal with KnC again, nothing to do with the rigs, it's the denial and head in the sand stuff I can't be living with.

+100000
Totally.

Timmers you are being sensitive. Avenger I'm starting to believe you are a troll. Of course it's unrefined as you well knew. You knew perfectly well they were delivering a product to be placed in your hands in an expedited fashion upon receiving the chips. As such margin upon margins were in place to ensure no post chip receipt refinement was required prior to you being able to hash. Obviously this would mean improvements would then succeed the initiation of delivery, and will continue to do so. There is yet more to come. This is not a first for just Bitcoin mining, this is pretty much a first for the entire IC industry. It's insane. To get this, this right, first time, on 28nm tech is unheard of. Those involved; fab, suppliers, backend, etc. think this is remarkable, and it is.

[Biomech]

My Jupiter is scheduled for delivery today. Of course I bought a Corsair HX850 - is there any *safe* way to use it, or should I not even plug the thing in?

What if I power on the HX850 and plug in the PCI-E connectors one at a time?

JESUS

I have absolutely no idea how people with this amount of experience find themselves buying these devices.  Really?

Literally in awe when I'm reading some of the questions in here.

brb, gonna go study electrical engineering for a few years and then ask a question that by then I'll already know the answer to. Thanks for clarifying, at least I asked first, instead of frying my stuff.
[/quote]

Good on you! This place is not newbie friendly. I know. Not long since I was in the same boat. (I wouldn't have made the mistake you did, but I made plenty of others!)

Seriously, though. Nearly all ac to dc power supplies are NOT hot pluggable. Not just computer equipment, that's across the board. You can get away with it on a lot of things, but the only one I can think of that's designed to be hot pluggable is a USB phone charger.  That's not a slam on ya, btw. Just sayin'.

[r3animation]

I'm running 0.93. Just tried to upgrade to 0.95, and it seems to work, but after restart the web interface still says that the current firmware revision is 0.93. I tried to upgrade (downgrade) to other versions but it's the same.

It is possible that the upgrade was successful and the web interface displays the wrong version number? Is there a way to check firmware version through ssh?

You need to shift+refresh probably to see the updated numbers.

[Bitcoinorama]

Then why not leave the 8 VRMs on for that point??

why are you dodging that?  what was the main gain for everyone on taking off the 4 VRMs??   You could keep shipping with 8 until you tested taking off 4.. .how can you try to keep avoiding this logic?

Simple.  KNC chose to use an off the shelf power module.  This means rapid prototyping but it also means higher cost.  Those modules run $25 each.  8 per board means $200 per board in cost just for the power supply.  That is $800 per Jupiter.  If they leave 4 of them off that is an extra $400 in profit per unit.  1500 units @ $400 ea = $600,000.

There is absolutely no other reason to change the extensively designed, tested and fully operational boards right in the middle of peak production.

+1

I too say fine with removing 4 VRMs     **AFTER**  testing

it would take 1 minute to see the wattage increase 70 PER BOARD and heat rise 10C PER BOARD

Actually no. There are many compatibility issues that arise, even with identical components populating the board. It really is like a group of in laws that don't get on with eachother, being made to behave amicably with eachother over time.

[RoadStress]

Firmware 0.95 solves the VRM problem and boards seem to work fine with only 4. Thank you Orama and KnC! Welcome to working with customers Orama and enjoy your ride Of course people are whining, but some are just not realizing that you are trying to push things out of the door as fast as you can and this hiccups are totally understandable. At least for me. The good thing is that you are pushing firmwares and fixes as fast as you can too. Even if KnC isn't perfect it seems that it's the best and the most reasonable "pre-order" ASIC company yet. Great job and keep doing your stuff even if people are complaining as hell.