For this calc example I assumed I would get it at the last of October, or 137 days from now.
I then assumed a 2% difficulty increase per day until then, which gives me a starting difficulty of 239941346.93285772
If I now put this into
http://www.coinish.com/calc/ and run it over 12 months with a 100% utlization, my estimated income is -3149.71USD
So considering this, it doesn't seem like the greatest idea anymore.
You've clearly got the hang of this. My only quibble might be that your rate of increase in difficulty of 2% / day is 32% per two weeks, which may be a bit steeper than we could expect (Avalon Batch 3 are still not shipping, and BFL still at a trickle). Redoing the calc with the default 1.0306% gives diff 92M in 137 days and breakeven 44 days later (with continued diff increase).
Which explains why KNCMiner have so many keen customers. But it all turns to dust if there is any significant delay in the project (and ASICs are very prone to over-running the schedule). Remember these will be prototype devices in the first batch from the foundry, and KNCMiner are expecting to ship these to customer. User KS pointed me at a good article
http://www.edn.com/design/systems-design/4319891/The-economics-of-structured-and-standard-cell-ASIC-designs (a little dated at 2006 but still relevant). The key point being the 12-26 week delay from prototypes to full production.
Anyway this discussion is best continued on the KNCMiner threads in CUstom Hardware rather than in newbie land (hmm still only 1hr 42mins logged in, they key is to keep browsing the forums, refresh at least every 15 minutes, Mining Speculation is a good starting place
https://bitcointalk.org/index.php?board=81.0 )
