I must confess to never having used a MIPS processor, instead I had leaned towards the ARM family early on. I am wondering what the IP portfolio is that would be of value to other companies. I would like to know what the analysis think was the limiting factor in the MIPS line? Was it the overall speed, performance, or cost that was the limiting factor in the "ARMs" race?
If some quality buyer comes, he will be better able to utilized the quality resources of MIPS, it totally depended on the decision makers where they want to take the company, the quality manpower will always there at the bottom level.
The last article Peter Clarke wrote about MIPS sale was inaccurate (he has a habit of doing that). Now he rewrites something Bloomberg reported. Can't EEtimes hire competent reporters these days, instead of just rehashing articles written by real reporters?
MIPS lost out when IBM took over the last generation of video games with IBM's Cell processor and the like. A huge royalty revenue soon dried up for MIPS, which was never replaced. A couple new spins of MIPS cores came out, but those didn't go far.
Interests in China, which seems highly motivated to advance in high-tech computing, could pick up a great portfolio of processor patents and know-how by buying MIPS Technology, also resolving many potential disputes in regards to IP. The Godson processor knocks off the MIPS architecture. If a company in China bought MIPS, it might be a strange reversal to send IP revenues to China.
But it is sad that such an icon seems to be falling to its knees.
I used MIPS for almost ten years, but their patents for the unaligned instructions that Lexra never used, should not have been awarded (prior processors had unaligned access). MIPS closed down a company that was winning more customers. When you loose Sony, Nintendo etc you obviously upset people. Gouging out eyeballs in Nintendo's case was clearly visible in the annual statements. The SysAD and SysCMD bus was the next worst thing to try and use, plus the higher prices for interface chips to use this silly standard. PowerPC took out the high end, and ARM with reasonable licence fees took out the low end. Instruction sets are unimportant when writing in C; the ecosystem is more important. ARM have done well at MIPS expense.