Quote from: MrTeal on November 27, 2012, 10:34:15 PM

Quote from: meowmeowbrowncow on November 27, 2012, 10:25:37 PM

Quote from: MrTeal on November 27, 2012, 09:49:51 PM

Quote from: repentance on November 27, 2012, 09:43:51 PM

If Nasser is telling the truth, wouldn't it mean that either a) the chips haven't even been produced yet or b) the bullet run for the chips happened but they're going to scrap all the chips from that run and another bullet run (which will happen fuck knows when) will be needed to produce the chips with the new design?

According to Josh back on the 3rd, they had issues caused by refraction and that's why the chips were delayed with estimated delivery on the 23rd. I can't imagine how they wouldn't know that redesigning the clock buffers would cause a delay back when Josh posted this if it wasn't a new issue.

Quote from: BFL_Josh

As for why the delay, if the November issue of Bitcoin Magazine would come out it would make a lot more sense. Bottom line is, our chips are the most advanced chips by a wide, wide, margin compared to any other offering. Because of that, the design has taken longer than expected. Our process node also blows away any of the toy offerings from the other vendors and we had to deal with some unexpected refraction issues that needed to be addressed.

So, since we are left with differing statements from BFL I'll read between the lines.

The first run/prototypes underperformed. Makes sense if the official word is that "there was no flaw," but there was a revision.

They've said they have no working prototypes. Underperforming seems a tough pill to swallow, at the least they could be making a buttload of Little Singles using 8 chips but clocked low enough to get 30GH/s out of 8 chips. Even if power consumption was higher at 50W @ 30GH/s, they could easily showcase those.

Good point in regards to showcasing prototypes.

But at this point I'm not sure that BFL is acting completely rationally.