Topic: Cointerra or Investment? (Read 2107 times)

You have to rely on BTC value increasing, or already have something like a Jupiter order.

Also, at the moment there is no cost efficient way to buy 100GH/s and make a profit or even the costs back, in some cases. (w/out power costs taken into consideration)

The likelihood is you will see systems which max out at 3TH per power supply, and you'll be able to use multiple supplies to overcome it (or multiple boxes, same thing).

The biggest change will be cost per GH. 28nm is here to stay awhile.

5 TH/s for a single system seems potentially reaching too high...   the problem is how many TH's will fit in each box (not the asics, but the power & cooling considerations override how many asics will fit).

Cointerra is already on the limit right now with 2 TH/s, with chips consuming 0.6 W/GH/s, however once you factor in the power conversion losses of the dc/dc converters, the power supply losses, and add in the cooling system and controller consumption etc... you're really talking 0.8+ W/GH at the wall...   So it seems unreasonable that you'd get to HALF power consumption in the next generation as the non-asic's power consumption and losses remain in the system regardless of how low your asics can sip the power.  It makes the drag on achieving low power very high.

So... lets assume 0.5W/GH at the wall is the best case for the next generation (and i suspect it won't be because people prefer to overclock and have higher performance rather than lower power etc).

5 TH/s is 2.5 KW in one box.  Cooling 2.5 KW is pretty hard... and also, finding a low cost power supply for a 2.5 KW load is actually pretty hard (actually, you need more, as you can't run power supplies at 100% load without them being very inefficient so you'd need a 3 KW power supply or more!).  You'd need two very expensive atx power supplies, for instance.  They take up a *LOT* of room... and are expensive.  for instance, two of these... ($600)  http://www.newegg.com/Product/Product.aspx?Item=N82E16817171069

So.. my gut feeling is that achieving 5 TH in a single box is going to be challenging and i suspect quite impractical for the time being.  For best cost & performance, its best to keep it lower, so you can use more affordable cooling and power supply solutions...   there's no magic here.  the system cost is higher than the asic cost !

-- Jez

knc machines are not going for 10k on ebay. if they pop up, they go for about 15 to 17 lol...

AFAIK they will be going back into heavy R&D after they ship all their outstanding orders.  Gen 2 is expected March/April next year.  Specs should be at least 5TH/s if they want to stay competitive.

I'd like to know as well

when will knc gen 2 come out and what specs it will have?

Folks overuse this word a lot. Cointerra is not a scam, the product will ship. On time? Who knows, but it will.

To the OP, if you can find a first or second gen KNC unit for less than original cost, it'll do you well. I'm on the fence about Feb/March 2TH mining myself. If the value of BTC stabilizes at $500, it seems to be a good idea to get a miner.

Holding BTC is always a win in the long term.

Keep in mind, you can re-sell your mining equipment and more than likely break even or make profit. $5k KNC units are going on ebay for $10k right now. The same will occur in Feb/Mar, ESPECIALLY if BTC rises.

At worse you will find a way to break even.

i have done both. ive bought btc, and ive bought mining gear.  cointerra and hashfast are only vapourware til end of december at most and then they will be shipping.  thats pretty close.  But i do agree, if you want a miner, NOW, you can buy KnCMiner (or BitFury).  Actually, KnC has announced a small delay so even they arent shipping from stock this minute.  Buying a pre-order of cointerra or hashfast gets you more hashing power for your money, but with a delivery date a month or two further away.

Buy + hold or KnC Gen 2.  Cointerra and HashFast are vaporware.

Put simply, my question is whether it's going to be profitable to mine with a 2TH/s machine, in late-February or March, given the purchase cost, operating cost and the difficulty mining will be at, in the coming months.

Versus investing $6k-$10k directly into BTC

-

I've been doing the math & even though I'd love to have more machines running, it might not be the best solution.
-

Thanks