Also, in order to sell chips, you actually have to have chips to sell. So, how many chips do you think are likely to be had from the sample batch in November? How many from low volume production in December?
Meanwhile LC will have been hashing with 4 TH/s from the 10th September at the latest and increasing that to 50 TH/s in October. You can mock that hashing power as being meaningless as much as you want, but it'll represent at least 5% of the network hash rate. If ActM had any chance of paying meaningful divs before December, you'd be singing a different tune.
Mindlessly buying hashing power is a sure-fire way to lose money. Just ask anyone that paid BTC1.99 for a block erupter.
Again, you are just speculating on the dates. ActM may also have it's high-volume process already at 100% in December, depending on NRE pay date.
LC is a whole different gamble than ActM. Have you even seen a working LC chip yet? I haven't. They are receiving untested chips and untested PCB's, relying 100% on "simulation" results. How nice. They've designed a chip using a sea-of-gates methodology similar to what BitFury did. BitFury even "simulated" their chip to reach 10GH/s! You might wanna go find out what it actually ended up working at!
I thought the untested chips were what BTCGarden was offering... Yikes.