Never heard of them.  Probably a scam.  Search this forum and see if anyone has done business with them here. 

Caveat emptor. It is insaney easy to make a website promising someone the moon and once they pay Bitcoin there is no getting them back.

How did you learn of them?

So what does this have to do with Bitcoin?

Maybe.  Winter is probably less of an issue.  The relatively humidity is lower (cold, dry air) so there isn't as much water to condense out.  If you get condensation (more likely in Spring) it will be because the roof is colder than the dew point and water condenses out of the air and drips onto the rigs.  If the heat from your rigs keeps the underside of the roof higher than the dewpoint then there isn't much risk.

If you can change the setup from a roof with open walls to an enclosed box that keeps the temp of the air inside the box above the dew point the chance goes to zero.

OP might also find this interesting:
https://bitcointalk.org/index.php?topic=281279.0

It is kinda sad.  The miner sub forums at one time were a place to experiment, today it if you deleted every post which deals with return on investment it would be a ghost town.  I wouldn't do anything today.  The "daily value" of a new rig is still too high.  The downtime and potential for problems are more expensive when the revenue rate is this high.  In time difficulty will go up and what a rig makes per day will go down and then it will make more sense to experiment again (you have less to lose). 

Keep in mind KNC used a custom bar to attach the I30/A30 heatsink to the pcb.  The bar has holes in the right place to attach to that heatsink, they aren't using the "standard" bracket which comes with the heatsink.  Using a different heatsink (or waterblock) would either need mounting holes in the exact same spots (unlikely) or someone making a "custom mounting bracket".  It would have been easier if KNC used the standard AMD or Intel mounting holes, then any compatible waterblock by any vendor would fit.

And difficulty will be even higher to compensate.  At 10x current difficulty a 0.8 J/GH 28nm chip will use the same amount of power per BTC as an Avalon does now.  The network is self adjusting that way.

No.  Wafer is the uncut silicon.  The wafer is cut into individual dies (about 1000 per wafer).  The dies are then attached to the substrate (4 dies per chip in this case) and enclosed with a metal "lid" to form a single chip.

It depends on what hardware.

On a CPU, GPU, or FPGA they will consume more power than the BTC is worth.

However ASICs are a different story.  Looking at an "inefficient" 1st gen Avalon rig it uses ~ 8 J/GH.
Difficulty today is ~510 million.  So it takes on average 2^32 * 500M / 25 = 8.76 *10^16 hashes to create a bitcoin (25 BTC per block).  This assumes no stale shares, hardware errors, downtime, or pool fees.
At 8 J/GH that is 700,938,662 J or 194.7 kWh (3,600,000 J per kWh)

If we assume the average miner pays $0.10 per kWh, then using an Avalon rig would require ~$20.00 in electricity to produce 1 BTC.  That is just using the dificulty today, most people believe the difficulty will go much higher, probably to triple or quadruple the current value by the end of the year.  That would mean ~$80 per BTC using an Avalon (@ $0.10 per kWh).  The factors for the elecctrical cost of one BTC are efficiency (J/GH), electrical rates (USD per kWh), BTC exchange rate (USD per BTC), and difficulty (determines # of GH per BTC).  

With future not yet delivered hardware (highest 28nm specs) and $0.05 per kWh the cost may get close to $1 at current difficulty, however difficulty will be much higher by the time those devices are available (Dec/Jan).

Unless storage is 0 it will always be about quantity.  Not even sure how quality applies to a concept like the storage requirements.  Still I removed the first paragraph.

You can trade storage for security.  One example is a DHT.  Split the blockchain across 100 nodes and you reduce the requirement per node by a factor of 100.    That probably is a bad compromise to make when dealing with money.  If someone attacks the torrent DHT worse case scenario you download the wrong content (or thinking the content doesn't exist).  If someone attacks a Bitcoin DHT worst case scenario is they trick you into thinking you were paid when you weren't.

The blockchain can't be compressed, however it can be pruned.  Despite all the gnashing of teeth this is why the client change to make tx with outputs below the dust threshold (roughly 50% of min mandatory fee) was very important.   Economical outputs generally will be spent thus there are new outputs added to the UXTO and spent outputs removed from it.   The blockchain may grow exponentially but the pruned blockchain or the UXTO will grow closer to linearly.  So while pruning today may reduced the blockchain by X% in the future it will reduce it significantly more.

I think eventually many nodes will not download the unpruned blockchain but rather just the pruned version or at least pruned only for blocks deep enough in the chain.   Most things can be accomplished with "only" the pruned blockchain even mining and the storage savings are significant.

So you may eventually see three classes of nodes:
archive nodes - complete unpruned blockchain (complete canonical record of Bitcoin since the genesis block).
full nodes - pruned blockchain.
SPV nodes - real time query to full nodes for relevant portions of the blockchain.

Would be interesting to see what the chips do with water cooling.  The nice thing about water is it has a high heat capacity so it takes a lot of heating or cooling to change the temp more than a degree.  Some variable speed fans on a radiator and you might be able to keep the chip "warm" without needing to keep the ambient temp warm.  Certainly would be more wife/girlfriend friendly than devoting one room to tropical temps.

Not on any decade long scale.  In the short term we may grow faster than that BUT the current transaction volume is a tiny fraction of what current hardware can handle so there is a "cushion" built in.  We could increase tx volume by a factor of 20x and not be limited by hardware.  So if we grow at double (or triple) Moore's law for a couple years it simply eats into that cushion.  There is no scenario where Bitcoin growth can exceed Moore's law on a multi-decade timeline, there are only so many humans on the planet.

In 30, 40, 50 years, the discussion of not enough GBs look as silly as saying "video games will never take off because it in the future they will require thousands of floppy disks".  I bought a 1 GB drive for $300 in 1995.  That would be $300,000 per TB, and in less than 20 years the cost to store 1 TB has fallen to $50.  In 20, 30 years I fully expect the cost to store 1 PB to be <$300.  This isn't ancient history we are talking about just the effect of Moore's law in a single lifetime.

However in the long run most users will use SPV clients.  The security of the network doesn't require 100% of users running a full node, just that a critical mass continues to do so.  Between merchants, , developers, exchanges, service providers, hobbyists, and just die hards who refuse anything less than an equal peer on the network I don't see it being an issue.

HashFast announced a week ago there is at least a three week delay.  That would put it mid Dec.  Beyond that it is all just speculation.

I wasn't really talking about mining.  Mining secures the network (and introduces new coins into the network) the value of Bitcoin is based on what Bitcoin "can do".  Note "price" may vary from value or utility but only so long.  If price gets too far ahead of value/utility it will eventually crash back down.   Rising value and utility is based on increased adoption and those users doing more with Bitcoins.  If you double the number of people using Bitcoin and double the number of non speculative transactions they each make then the value or utility of the network will go up and with it the continued climb of price against the dollar.  

However currently most Bitcoin users are relatively sophisticated when it comes to technology but as Bitcoin "taps out" that market to continue that exponential growth will require taping into markets with a lot less sophistication.

If your grandmother going to learn how to use the QT client, or deal with pywallet to manually remove tx that get stuck due to incorrect fees settings?
Is a Billionaire going to buy $1B in Bitcoins by wiring a billion dollars to companies he has never heard (MtGox) located in countries he may not be comfortable (BTC-e)?
If a a thousand "casual users" puts 1 BTC into an eWallet and the operator runs away with it  will they all continue to use Bitcoin in the future like a "diehard believer" might?
Are millions (and someday tens of millions) of users going to be able to keep Bitcoins secure form hackers despite being HORRIBLY bad at security and running a Windows OS full of malware because they disable security updates that take too long? 

I would guess the answer is no at least not today with the wallets, exchanges, and service providers that exist.  The bitcoin infrastructure is going to have to improve, it is improving but at a much slower pace than adoption and eventually that will be the bottleneck.  Growth will slow until Bitcoin can be made easy, safe, and painless for the "masses".  I don't care to speculate on where or when that will happen only that there is a good chance it will eventually happen.

"Can",  I thought it was guaranteed?

It will not be a problem to have them in the same wallet with "big coins".  The coin selection algorithm attempts to minimize tx size.

Of course if you transfer them to a new wallet (or address or same address) you will consolidate them into a single output worth 5 mBTC but it will be a ~3KB tx and thus 0.3 mBTC in fees.  It won't be free but that is the cost of small inputs and now you have a single 5 mBTC output.  Honestly I would't worry about it much but if you want to that is always an option.  

In general if you can control it (a cashout from a pool or exchange) always try to keep your outputs as large as possible.  5 mBTC in 1 transaction is going to be cheaper to use than 5 mBTC in 15 transactions (or 150).

Stupid analogy.  What will 41,000 BTC a year be worth in 27 years?  It is entirely possible it will be worth more than the block reward is now.

If she didn't sell, the 200 BTC bounty is worth $70,000 today.  Not too bad for getting a tattoo.  I hate tattoos and I might just get one if someone paid me $70K.

The proof is in the pudding.  The sold out in what an hour?  Now if there were still 190 of 200 unsold and tons of people complaining saying they won't place an order until the shipping is fixed you might have a point.  Not saying it is the best strategy but it worked (and honestly it will probably work again and again and again and ... ok now I am just wasting bytes).

Noobs asking noobs for piles of money with no real business plan. 

This is going to end well.