Where is Transmeta Heading?

Autoversicherung writes "Transmeta, once the darling of Silicon Valley, employer of Linus Torvalds and heralded as the new Intel is facing bleak times. Having $53.7 million in cash and short-term investments in its coffers, enough for just under two quarter's worth of operations and a reported net loss of $28.1 million and revenues of $11.2 million for the fourth quarter of 2004 the company's future is everything but certain. Will the planned restructuring to a pure IP company help?"

Comments

Help .... who?

[zappepcs]

Isn't this how many technologies make it to the consumer? Company A invents it, goes broke trying to sell it, then the big players buy it cheap and finally the rest of us get to use it?

Oh, except for that famed 50+ mpg engine....

A purely IP company, huh?

[geminidomino]

That would mean that it would be in their best interests to support stupid laws like copyright-until-the-heat-death-of-the-universe laws and software patents.

Kind of a delicious irony there... employing Linus and striving to hamstring Linux...

Re:No

[eln]

The problem is the chips are just too damn slow. Intel's chips in notebooks run hot and suck power fast, but they don't run hot enough or suck power fast enough to make people want to significantly decrease the performance of their notebooks for less heat and longer battery life.

Intel has put billions into R & D over the years to make their chips small and fast, and they are now starting to put money into making them more power efficient. Transmeta can't compete with that sort of hardware engineering with software alone. In addition, the idea of running multiple instruction sets on the same chip is not that big of a deal in an x86-dominated world. Transmeta had a good idea from a software engineering standpoint, but there's no market for that idea.

Failed Expectations

[Proc6]

Only marketing points that ever stuck in my mind about their CPUs were,

- Could run other OS's through emulation.
- Would give your notebook insane long battery life.

The first point never mattered in a Windows / Linux world that ran on i386 anyway. The second point never really came to be. I remember looking at Sony Picturebooks with dinky screens and Transmeta CPUs and seeing them last like 2 hours. Big deal. If they didn't double battery life, the public wouldn't notice enough to buy Transmeta on purpose. Then Centrino came out and, well, yeah, thanks for playing.

why is slashdot still talking about transmeta

[Anonymous Coward]

the *only* reason why slashdot cared about transmeta was because linus was hired by them.. no other reason. the simple fact is that this company is a failure so could we please, please stop talking about it? it's going to go bankrupt like 99% of all startups, so it's really not that big of a deal. their technology really wasn't that great because intel smothered them with additional versions of their centrino chip. too bad so sad.

Power management, not code morphing

[Animats]

Transmeta's story is funny. Their big idea was supposed to be "code morphing", or on-the-fly recompiling for a different CPU. But it turned out that they achieved some success because they were the first to take on-chip power management seriously. That gave them an edge for one development cycle. Then, Intel and AMD noticed that power management mattered, and fixed their parts. End of Transmeta.

"Code morphing" for the x86 instruction set never made too much sense, because making fast x86 machines is well understood, although painful. AMD already did some "code morphing" at cache load time; they inflate all the instructions to a constant length. (Intel does it differently.) For a CISC instruction set with inherent speed problems (the DEC VAX comes to mind) "code morphing" could be a big win. But there's no market for a fast VAX at this late date.