- Related Stories
-
Transmeta shows the fine print on Efficeon
October 14, 2003 -
Transmeta: Are the chips down?
November 5, 2001 -
Transmeta plans to raise more than $140 million in IPO
October 2, 2000
Transmeta will complete a "critical evaluation of the economics of its current business model of designing, developing and selling x86-compatible microprocessor products," the Santa Clara, Calif.-based company said in a statement.
Instead of making and selling processors, the company would instead focus more of its energy on licensing chip technology to third parties.
The shift is a reflection of the harsh realities the company faces. Designing, manufacturing and selling chips is exceedingly costly, even when the actual manufacturing gets outsourced. Price wars are also common. By contrast, licensing intellectual property often involves far less overhead. Rambus, one of the more successful so-called IP (intellectual property) companies, has only about 200 employees.
"There are some necessary economies of scale in the processor market, and they need to be about 10 times larger than they are," said Dean McCarron, principal analyst at Mercury Research.
Transmeta kicked off a licensing effort late in 2003. In the most recent quarter, licensing revenue came to $3.7 million, while chip sales came to $3.3 million. It also lost $27.5 million during the three months. Typically, Transmeta loses more than $20 million a quarter.
The company emerged in 2000 with a promise to bring energy-efficient processors to notebooks. The company's low-energy push spurred Intel to cut the energy consumption in its own chips. While Transmeta landed some early deals with Sony and others, it hit several manufacturing problems in 2001.
Revenue, and new deals, subsequently began to sink.
See more CNET content tagged:
Transmeta Corp., manufacturing, microprocessor, intellectual property, Intel
- Sad, but predictable
- I had high hopes for Transmeta, but Intel responded fast enought with the P-Ms to offset their design advantage. Unfortuantly it is hard to break into the hardware market, not matter how revolutionary your ideas are.
- Reply to this comment
- Java processor?
- The whole code-morphing architecture lends itself to running stuff that isn't just x86, and I'm sure I remember the idea of Transmeta being able to behave like a Java VM. What happened to all of that? Imagine an (almost) native Java machine - it's like the chip and VM were built for each other.
- Reply to this comment



