intron
1 - How we can be sure that there are more than a few hundred partially efficient chips (2,3GH/s instead 5Gh/s) cut out of from center test wafer?
2 - Do we have any guarantee that TSMC this year let Bitfurry on their already heavily loaded production lines?
3 - In my opinion all money from preorder go to corrections that must be made to the chip. Than start small quantity test wafers. Unless Bitfury wants to introduce the production of something that had passed with the assumptions for more than 50%.
And if it is so weak performance whether for a few months, these chips will still work? Were the tests performed life-time chip in extremely adverse conditions to determine how quickly degrades structure?
Its very impressive start very risk development so high (55nm) and get working chip. And I am not speaking here against this project, but before invest will see what I'm dealing with.
1 - We've tested many chips so far (they all come mixed together, no sorting) and have found the yield to be outstanding. We have enough chips in hand to satisfy all orders.
2 - We don't work with TSMC. UMC is our foundry. We already have orders in place for next batch of chips and have no reason to believe that it would be delayed.
3 - We didn't go the MPW route. We went with full mask-set and produced a chip that works just as we expected this to. Hence we had a large engineering run, that was successful and can also be (and is in the process of being) deployed.
Did you plan correction on metal layer for improve performance ?
Why we can't buy July chip delivery on EU distribution ? On US page You can still order for 10 units for July.