Swedish ASIC miner company kncminer.com

[The Avenger]

Pages and pages of comments about me ignoring some idiot. Some of you guys don't have much to do with your lives if this is how you spend your day.

I'm not interested in making any response. Let's get back to discussing what this thread is about. The rest is irrelevant. Peace.

[1713954159]

1713954159

[1713954159]

1713954159

[1713954159]

1713954159

[rolling]

any stats on amps for Jupiters?   15 or 20?

I am figuring if setting up at home, separate circuits/breakers per miner is the way to go

It depends on the voltage you are getting.  In the US, it is typically 120 but that can vary from 110-120.

Watts = Amps x Volts

If you are in the US.
650 to 850 Watts / 120 Volts = ~5.5 to 7 Amps each Jupiter so 2 per dedicated circuit is a safe bet.

[erk]

Well, that's obvious, and the more pressing issue considering we are fighting difficulty rise isn't a stuffy room. The problem we are trying to overcome is achieving the safest most stable overclock speed possible. I can care less if the office AC has to work harder to get rid of the heat, I want to maximize the potential of my hardware.

[Edit]
The office that my miners will be stored in includes electricity so, no extra cost for additional cooling.

How do you know these devices can be overclocked?

[crumbs]

I've been thinking about the cooling solution that KNC provided, and have been trying to figure outa way to cool these things enough to squeeze out another 50GHz/s. I have been considering the notion of stripping the PCB from the case, stripping the fans from the heatsinks, and flipping the board 180 degrees upside-down with the Heatsinks immersed in a mineral oil bath instead of the traditional full board immersion techniques people have tried.

The immersion pool would have a fan cooled radiator and would utilize a large liquid pump such as the kind you would see in a large fishtank providing circulation for the oil. Anyone have any thoughts about this?

if you're going that far, why not watercool? ==>> Not messy.

I had the same idea as him. pretty much because I don't have the right formfactors for a watercooling device - mineral oil is quick and dirty but works imo

There are tons of waterblocks out there, the universal brackets are trivial to mod with a cheap drill (if for some odd reason the footprint is non-standard -- would be an odd design choice). Edit: I mean CPU waterblocks.

I never had good luck with oil -- heat transfer coefficient is 3 to 4 times worse for oil, it's messy, it wicks, is more viscous (doesn't flow as well) etc., etc.  Oil is also a pain to circulate well in the tank itself, especially around heatsinks designed for forced air, and don't forget that it wicks.  And it wicks.  Drop a wire into the tank, and have a puddle on the floor in the morning.  4realz.  Anything dipped in oil becomes a pain to rework, no matter how much solvent you use (trust me, it will get on the board even if the board itself is not submerged.  And it *does* have a smell.  There's fun stuff i've only read about -- boiling point ~30 - 40 C, but it costs way too much.  Something like that.

[Mota]

One word of caution to the mineral oil, if for some strange reason your heat exchange system (fan/radiator) fails, you will deep fry your circuitry, unless you have a large enough bath to carry the thermal load long enough to realize and fix.

I wonder if that would happen - heat dissipation from the circuits -> oil -> glas should be sufficient if you use a fishtank. oil has a heat dissipation of ~0,13 (a fifth of water) and glas is at around 0,76. If you take the surface of a fishtank to be around 1 m² the temperature should even out at max. 65°C if your room temperature is around 20°C

[DPoS]

any stats on amps for Jupiters?   15 or 20?

I am figuring if setting up at home, separate circuits/breakers per miner is the way to go

It depends on the voltage you are getting.  In the US, it is typically 120 but that can vary from 110-120.

Watts = Amps x Volts

If you are in the US.
650 to 850 Watts / 120 Volts = ~5.5 to 7 Amps each Jupiter so 2 per dedicated circuit is a safe bet.

thanks for the reassurance!!!

[JoelKatz]

You can do the math. If you have an 85% efficient power supply delivering 800W and your line voltage is 117V (standard in the US), you get:

800 / .85 / 117 = 8A

8 Amps.

[jelin1984]

Any news about the ASIC 28nm chip?

If bet 1 btc that not delivery at october
How btc I will win?

[Mota]

Any news about the ASIC 28nm chip?is only 10 days from the time that knc must have the chip at hand for not delay

Pray tell good Sir, how did you get that number of days for no delay? And no delay to where? If you live next door to them you should still be on time if they send them on the 26th...

[xephireusMMX]

Any news about the ASIC 28nm chip?

If bet 1 btc that not delivery at october
How btc I will win?

There is betting site for this:

http://bitbet.us/bet/472/kncminer-will-deliver-asic-devices-before-october-1st/

But hard to say how much you get by betting 1 btc and winning

[mwarrior]

Delivered.  1 Bad Ass Power Supply.
To be delivered.. 1 Bad Ass Asic miner.   ;-)  
Sad to think that the power supply will outlast the miner. 
Oh well. 

[btc_uzr]

 

sorry, but at first the platinum one is few bucks more only paying off itself in no time
next you wasted money, they proposed 850W..you remember their strategy 'margins upon margins upon margins'...so it should be sufficient even for next generation and overclocking (if possible).

For the same money I'd have preferred the corsair's digital platinum PSU with 7years warranty and 92% efficiency in worst case
http://www.newegg.com/Product/Product.aspx?Item=N82E16817139041

[blastbob]

sorry, but at first the platinum one is few bucks more only paying off itself in no time
next you wasted money, they proposed 850W..you remember their strategy 'margins upon margins upon margins'...so it should be sufficient even for next generation and overclocking (if possible).

For the same money I'd have preferred the corsair's digital platinum PSU with 7years warranty and 92% efficiency in worst case
http://www.newegg.com/Product/Product.aspx?Item=N82E16817139041

1000w might be a good choice for hardcore overclocking.. time will tell

[erk]

1000w might be a good choice for hardcore overclocking.. time will tell

How do we know KNCminer chips can be overclocked? Did I miss some announcement from them?

[utarinues]

Your funny guys, you are ordering PSUs too early

[ASIC-K]

Your funny guys, you are ordering PSUs too early

+1

when i get a tracking number, then ill head down to the computer store. lol

[mwarrior]

Delivered.  1 Bad Ass Power Supply.
To be delivered.. 1 Bad Ass Asic miner.   ;-)  
Sad to think that the power supply will outlast the miner.  
Oh well.  

Per a conversation with KNC.. I was advised that using a power supply like this one would be good for the efficiencyof the miner.  Secondly I work in IT.. and can assure that this bad boy will get used appropriately long after I blow holes in the Asic miner once it has served it's purpose.  Thank you drive through.  

[demonmaestro]

although i got to ask why a cooler master and not a corsair

[DeathAndTaxes]

although i got to ask why a cooler master and not a corsair

My guess is cost. Still many people might be surprised to know that Corsair doesn't manufacture any power supplies, and neither do Coolermaster, OCZ, PC Power & Cooling (after 2009), Thermaltake, Silverstone, EVGA, Rosewill, NZTX, XFX, etc.  There are only a half dozen companies which produce (good) ATX power supplies in the world.  It is sometimes funny to have two people arguing about which is the better PSU when both are the exact same power board made by the same company and the only thing that is different is the stickers and color of the cables.

http://www.tomshardware.com/reviews/power-supply-oem-manufacturer,2913.html

Most of the high end PSU are made by companies that most people have never heard of like Channel Well, FSP, Sirtec, HEC, Super Flower, etc. Two companies which actually make their own power supplies are Seasonic and Enermax (most newer models).  SeaSonic sells units under their own brand and they also make units for other brands.  One thing to watch out is the other brands often will change manufacturers even within a same "model" so without seeing the power supply label and looking up the UL code you really have no idea who made the PSU.  That is one reason I have always recommended SeaSonic.  Every SeasSonic is made by SeaSonic and nobody else.

[DeathAndTaxes]

For the same money I'd have preferred the corsair's digital platinum PSU with 7years warranty and 92% efficiency in worst case
http://www.newegg.com/Product/Product.aspx?Item=N82E16817139041

Platinum doesn't mean 92% efficiency "in worst case".  More like 89% to 92% depending on load on 120V.  On edit:  240V is generally 2% to 3% higher efficiency which is what btc_uzr was referring to. 

Still a very good PSU.  

http://www.plugloadsolutions.com/psu_reports/CORSAIR_75-001305_860W_ECOS%203323_Report.pdf

One thing to watch out for is that KNC for some reason decided to use 4 PCIe connectors instead of EPS12V connector + 3 PCIe connectors.  On the 860 Corsair 2 of the PCIe connectors are on the same PSU connector (i.e. single set of wires with 1 PSU connector on one end and 2x PCIe connector on other end).  Without testing it there is no way to know for sure what the overcurrent limit is for that connector (good PSU limit both overall current and current per wire).  The PCIe standard only mandates that two 8 pin connector supply 300W (1.25@ @ 12VDC).  Now the PSU probably can supply a lot more than that but just by looking at it there is no way to know for sure.  Most reviews only show total load not max load per connector and certainly not at amperages beyond what a PC would use anyways.  

To clarify imagine a Jupiter uses 790W that works out to 16.4A per connector however with this PSU two connectors are on the same set of wires meaning 32.9A.  If the PSU has the overcurrent set to say 40A per connector (connector on PSU) then it would work fine.  However lets say it is only 30A.  Despite the PSU having enough total power its current protection system would cause it to trip because too much if being pulled from one wire.   This could have been avoided if KNC had used EPS12V connector plus PCIE connectors instead.