cnbcs129

Analysts Say Rambus Ride Not Over


By Hal Plotkin
Silicon Valley Correspondent

Fasten your seat belts, and prepare for the wild ride to continue, say the divided pack of analysts who cover Rambus Inc.’s {RMBS} highly volatile stock.

“It’s a very controversial stock,” says Seth Dickson, semiconductor analyst at UBS Warburg, based in San Francisco. “The question is whether a company with less than 200 employees is going to be able to tell a $30 billion industry what to do.”

Dickson is betting Rambus will get the job done. He slapped a “strong buy” rating and a $350 price target on the stock earlier this year when it was trading in the $170 range.


Rambus 52-week stock performance

Rambus’s fast-moving stock is one of the hottest topics in online-investment chat rooms. The Mountain View, Calif.-based company designed and owns royalty rights to a new type of dynamic random access memory (DRAM) chip used in personal computers and other devices, such as Sony Corp.’s {SNE} newest game console.

Backers of the new memory chip claim it is faster than preexisting and competing technologies and say its comparatively higher cost will come down with volume production in the months and years ahead. The Rambus naysayers, however, note that major DRAM manufacturers have been extremely reluctant to switch to a new technology that offers lower production yields, higher prices for end-users, and for which they also have to pay royalties.

Rambus’s stock nonetheless shot up last February after Intel Corp. {INTC} publicly reiterated its support for the company’s DRAM design, which it had already selected as the next-generation memory chip for personal computers. The stock fell back to earth during the recent technology-stock selloff and has given investors quite a roller-coaster ride since then.

“The best thing Rambus has going for it is the unwavering support of Intel,” Dickson says. “Intel and Microsoft {MSFT} control the future of the PC market. That’s why I put more Rambus stock in my IRA every month.”

Bill McClean, president of IC Insights, a semiconductor market-research firm based in Scottsdale, Ariz., says Dickson might want to consider some other investments.

McClean says Rambus will garner 7 percent to 10 percent of the DRAM market this year and perhaps 15 percent to 20 percent next year. After that, though, he says all bets are off.

“Rambus is a really hot topic right now, but it’s in a gray area,” McClean says. “If it weren’t for Intel being behind it, I’d have called it dead a long time ago.”

McClean adds that Rambus could still be on its way to the technology deathbed, despite some limited successes.

In a move that could signal major trouble for Rambus, major DRAM manufacturers recently formed an alliance with Intel, McClean notes, that has set an explicit goal of designing a royalty free next-generation post-Rambus memory chip. Rambus, which generates all its revenue from royalties paid by manufacturers who use its designs, elected not to participate in the new consortium.

“Intel is not going to be dictating the next generation chip,” McClean says. “And it’s looking more and more like Rambus’s solution will not be the one to follow on in the marketplace.”

Asked to guess whether Rambus will be a dominant force in the computer memory industry or just an asterisk five years from now, McClean quickly puts it in the latter category.

“There was just too much backlash from the DRAM makers,” McClean says. “I’d say they’re going to become an asterisk. They [DRAM makers] look at Rambus as an interloper. They don’t want some small little company in California telling them what to do. The feelings against that run too strong for it to become the accepted solution. The momentum is there to replace them.”

Drew Peck, an analyst at S.G. Cowen Securities in Boston, has a “neutral” rating on the stock and says it definitely isn’t a wise play for risk-averse investors.

“The problem with Rambus is it’s either a $300 stock or it’s worth zero,” Peck says. “There’s not much room for anything in between.”

Peck adds that Rambus either wins the battle to become the next memory chip used in most personal computers or it doesn’t. The outcome, he says, won’t be determined for at least another 12 to 18 months.

“The chat boards are going to be filled over the next year with people arguing both sides,” Peck says. “So we can expect more rancor between now and then. Personally, if I were a Rambus investor, I’d go and spend my money in Vegas, where the odds are about the same and at least you’d have a good time.”

But, like any big gamble, there are eager players on both sides.

Jim Handy, DRAM analyst at Dataquest, based in San Jose, Calif., for example, says he thinks there is little chance competing DRAM makers are ever going to agree on a design scheme for memory chips.

“There have been at least three previous similar consortiums, and none of them have ever turned out anything useful,” Handy says. “It’s highly likely this new consortium will fall into that same trap.”

Handy says that the complex process of designing memory chip architectures isn’t suited to work by a committee.

“An autocrat can always make things happen faster than a committee can,” Handy adds. “That advantage belongs to Rambus. ”

What’s more, Handy predicts that Rambus will make good progress in the marketplace over the coming months that will put it in position for its designs to be accepted for future iterations of DRAM chips.

There are some indications that could be beginning to happen. This week, for example, Dell Computer Corp. {DELL} announced that it will begin selling a Rambus-equipped PC for less than $1,200, about half the price of the first high-end PCs containing Rambus’s more-powerful memory chips that hit the market last year. Other PC makers are expected to follow suit with lower cost units, now that Rambus chips are becoming more available in the market.

“If everything goes well for Rambus and they do dominate the current generation they will have the upper hand for the next generation,” Handy says. “Rambus understands the heck out of this stuff. They’ve already tried a lot of things the consortium probably hasn’t even thought of yet. I don’t think the consortium has gone down all the roads Rambus has gone down. You can bet your bottom dollar Rambus is already working on the next best thing. To me, it all looks very positive for Rambus.”

It is a sentiment, Handy notes, that isn’t widely shared among professional analysts who, by and large, tend to be more skeptical about the long-term prospects for Rambus, given widespread opposition elsewhere in the industry.

McClean concedes that Rambus is making progress and could grab a larger market share than many now expect if end-user prices continue to fall.

“If they can get prices down to, say, a five or 10 percent premium [over standard DRAM] they can get some market share,” McClean says. “But there was a real reluctance at higher prices when the DRAM makers were already not making money. You can’t push non-economical goods to market.”

McClean adds that most PC users simply have no current need for the increased memory speed made possible by the Rambus technology. “There is no reason for people to pay higher prices for something they don’t need in an industry that competes on price,” he says. “Faster memory is just not important to many users at the moment.”

Dickson counters that broadband Internet applications, which include voice and video, will lead users to become more focused on PC-performance issues. He also says new PC buyers are wary of buying obsolete technology and will flock to systems containing the newer Rambus memory chips, once they become more widely available.

“What happened is that you had a lot of short covering after Intel made its announcement in February, which drove the stock way up into nosebleed territory,” Dickson says. “When a stock doubles or triples in a few weeks and then comes back down, it shakes out the momentum players. So I think we now have a kind of base here.”

Dickson says he thinks the stock will continue to trade in the 150 to 250 range for the next few months and will resume a more-steady upward trajectory as the company approaches the end of its fiscal third quarter on June 30.

“The catalyst I look for next is news of more low-cost Rambus PC models,” Dickson adds.

“I can tell you my outlook for Rambus’s stock: It’s going to go up,” Handy says. “And then it’s going to go down. And then it’s going to go up. And then it’s going to go down. And then it’s going to go up again. We’re no where near the end of the ride.”

On April 12, Rambus reported fiscal second-quarter earnings of 15 cents a share. The results beat First Call Corp.’s consensus earnings estimate by a penny and the year-earlier period’s results by 7 cents.

CC BY 4.0
This work is licensed under a Creative Commons Attribution 4.0 International License.