So on January 28th, everyone in Batch 1 will receive 4 more modules if I understand this correctly.
Yeah I can't see how difficulty combined with late start will result in more than a 25% return of purchase price by end of Jan (a return of less than 25% of purchase price is needed for the max MPP compensation). I mean even if difficulty growth slows down significantly it seems very likely. Of course if the return is higher by end of Jan then less boards will be issued by the MPP but I don't see that happening.
If they ship mid-December but start the MPP on Oct 31, that essentially means that between mid-late December and Jan 29th (let's say 40 days) you would need to earn back 25% of the purchase price. Even at the current difficulty a 400GH/s Babyjet would generate 0.3937 BTC/day, or 15.748BTC. That's already almost 25% the cost of the Babyjet.
BTW - It will be interesting to see how they implement it. They are saying the MPP program will start Oct 31st even though they're delayed, but this is the example from their website.
Example 2 You buy a Baby Jet for 50 BTC, and we ship it on October 25th. January 23, 2014 comes and goes, and it turns out that at the Baby Jets nominal hashrate, it would only have generated 25 BTC during that 90 day period. Calculated out, it would have taken an additional 400 Ghash/s of capacity shipped with your Baby Jet on October 25, 2013 to generate 50 BTC. We cant go back in time to give you that 400Ghash/s. Instead, HashFast will double that and give you 800 Ghash/s in additional capacity. In this example, HashFast will give you two additional Golden Nonce ASICs, each with 400 nominal GHash/s.
Ignoring the Oct 25th to Oct 31st difference, if that scenario played out and at the end of January HF sends out one extra module because hey, a BabyJet would have earned 25BTC between Oct 31st and Jan 29th they will have a lot of pissed off customers.