Mark Durcan's predecessor had a good look at Hyuandai Semiconductor when it was "almost" bankrupt in 2002 and could have bought that operation.
But the price, rumored to be between somewhere between $2 billion and $4 billion was considered too high and Steve Appleton walked away from it.
Hyundai, with a lot of good will from banking debtors who became equity holders, has become SK Hynix.
The lesson would appear to be, when it comes to memory, you have to be "all-in."
MIcron appears to be a perpetually loss making machine. In the last 5 FY, they made any profits in only 2. And this FY again it appears to have huge losses. http://www.google.com/finance?q=NASDAQ:MU&fstype=ii
Instead of Elpida, I am wondering how much time does micron have!
The best result for the incumbent DRAM makers might be that nobody buys Elpida, the company is officially wound up and the capacity is taken out of commission.
However, if somebody is going to buy Elpida and get the manufacturing capacity Micron might wish for it to be them to get additional economies of scale.
Ideally Durcan wants to be almost given Elpida so he has to neither borrow money or give away stock to effect the purchase.
No signs from Micron as to how soon they may buy or spurn Elpida. It must be remembered that companies always like to take the chance to look over another company's books even if they won't eventually do a deal.
Blog Doing Math in FPGAs Tom Burke 23 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...