I just wanted to point out that this statement is a big oversimplification.  A chip design and a pcb design are completely separate beasts.  I have done freelance electronics design (pcb) for 10 years, and I can say with some certainty the dc/dc to generate 1V at 100amps range (and there are likely to be several similar for different rails) to power a ~250W ASIC is not easy in a number of ways.  There is a whole mini-industry of chips and designs to handle this difficult problem.  A small loss of efficiency could be the difference between a viable and not overall design - it could function just fine, but creates too much heat/uses too much power/etc.
For an example of this look at cointerra's blog.
Cryptx says they are doing 3 pcb designs in parallel and WOW how fortunate that they have the resources to do it this way - the best possible way when time is of the essence.  Since there is obviously no time for a pcb respin once the ASICs arrive, the which-design-is-best contest has to be completed beforehand.

On a different topic, if cryptx asked to change the reinvestment ratio from the 65/35 to 60/40 or 55/45 or 50/50, I'd likely vote yes.  I see the inevitable future to be a small number of large mining operations, and if by doing a larger reinvest petamine can climb to being one of those, all the better for the shareholders in long run.  All else equal (ha there's a broad statement) if cryptx reinvests a larger % than the difficulty rise, petamine will have a seat at the grownups table.  By having enough $ to be a lead investor/customer for each next generation round of ASICs petamine can stay ahead of all the others buying the same chips.

Thoughts?
Jesse