It seems strange to me that they don't have a working product. Making a large order and just hoping that everything works out seems risky. I don't see how they can be so sure everything will work correctly.
This is a new age....throw tons of money at a business that has no working product for months....just have "FAITH"
I'm only worried about mass producing something before testing it.
I agree here. Testing should be done on small quantities (i.e. a prototype ...DUH). BFL must have some magic potion that allows them to divert the normal development process in any new technology.
Unicorns, leprechauns, pots of golds, lucky charms, rainbows!
You remember that little fiasco where they were using a QFN chip, but then when they tested it, the plastic couldn't conduct the heat out of the chip and into the heatsink? Well, lets just think about that for a second. They had chips, and they were even testing the chips on one of their boards. If they really had a faulty design, and the chips really couldn't hash, then they would have known about it in October.
I'm not sure how they tested the heat, but if they had a hashing ASIC they wouldn't be hiding it. They could have also figured out the spec for the unit by now if they had one. It's more likely they just skipped testing and went straight to production hoping everything works out.
The fact that BFL was able to switch package types mid-stream
might indicate that they never had a full production of wafers.
Some of the heat testing places use ceramic elements to mimic the thermal load and get a general sense of how the board and it's ancillary components will perform.
(and IIRC this thermal testing only took place last November)
My suspicion is both are true. I chuckle when Josh says, "We've done the impossible!," by changing the packaging type. When this most likely means that BFL was
never anywhere close to published timelines
ever.
