I'm not sure if this was asked before. By the time they deliver the units, wouldn't the difficulty rate be at a point which would render the 5mhs or 25mhs you're getting useless?
The asics will not provide and exponential hash rate increase like we've seen with sha-256. In fact GPU's will be able to compete as far as hashing power goes, but the Asic will win on less energy consumption so the electric bill wont be so astronomical.
^-- This is the best reason for these machines. If I can cut down from using 100amps of power and get the same--even more--hashrate.. I'd be super thrilled.
Yeah great at what expense though? That is a hefty price to pay for a cut in the energy bill, upfront, and who knows where the price will be next year.
If I examine it purely from an electric bill standpoint, the unit would pay for itself instantly as soon as it began mining. How could you not see the benefit? Even if the machine only mined at half the estimated speed and took up 1,000watts, it's still ahead of the game. At least for me. I use scryptmining to pay for its electric bill, and to still provide profit. Something like this is a benefit, despite the cost.
You are all assuming that the company will deliver and will stand behind their promises. If they do deliver then yea, you will double your profit because you won't pay astronomical electric bills.
On the other hand but you are paying upfront 2-6 months in advance and that money isn't mining while you wait, meanwhile you could purchase GPUs and start mining, and by the 4 month mark your rig will have paid for itself even with the electric bills.
Keep funding them with your money for a promise that doesn't even have such a good return/risk ratio

In a situation like this, you should only risk what is acceptable to lose. Scryptmining has beenvery/ lucrative for myself and others, especially among the recent pump/dump coins which have popped up making thousands of BTC.
With that in mind, a 30% BTC upfront preorder isn't that much to risk, and if the company seems to go south.. you're out 30% instead of 100% other companies over the past couple of years have demanded.
All the while, my systems are mining for me.. it's not a loss. It's also not for everyone. ASIC is coming to scrypt, be it through this organization or another, there are 3 different groups working on it. It's coming. It's not a bad deal to take a 30% risk on an unestablished company if you can risk it.