Wireless and broadband giant Motorola
MRAM has the potential to revolutionize the $48 billion semiconductor memory market. It is fast and should be cost competitive, as it eliminates the need for multiple memory technologies in a single product. That combination should allow MRAM to fight to replace Flash, DRAM, and all but the fastest SRAM. Motorola's first target market will be Flash memory in wireless devices.
And MRAM is not that far off. Sample shipments will be made this year. Production will begin in 2004.
"Non-volatile memory" is another important feature. Computer users are accustomed to waiting for their systems to boot up when powered on. Today's computers are "volatile" -- their memories dump when powered down. With MRAM, computers (and cell phones too) start where they were when powered off.
MRAM is also a light power user. Cell phone and notebook computer users will appreciate longer battery life.
What's not to like?
For one, Motorola is not alone in the MRAM marketplace. IBM
Moreover, MRAM is not the only non-volatile memory being developed. Ramtron
More importantly, as sexy as MRAM is, look at Motorola's semiconductor operations as of its last quarterly report. The unit posted quarterly sales of $1.1 billion, down 11%, compared with the same period last year, losing $125 million for the quarter. Meanwhile, orders declined 23% to $1 billion.
Yes, Motorola has the early lead in MRAM, but getting market penetration will be difficult. If nothing else, buying Motorola now to gain access to the semiconductor business means taking on technology risk at a time when the fundamental business is not strong. That's a tough way to make money in the market.
W.D. Crotty can be reached by e-mail at [email protected].