Message Font: Serif | Sans-Serif
 
No. of Recommendations: 35
Great thread! I hope you guys don't mind me putting in my 2 cents. This is my first post to your board, but I might be able to clarify some things.

***WARNING: LONG POST (but it contains just about everything I know about Rambus)***

Rambus has spent the last 10 years developing a proprietary technology that allows chips to communicate with one another at very high speeds. The first, and most obvious, application of this technology is to RAM. Specifically, they have a technology that lets you get data from the RAM chip to the processor at very high speeds. The technology is scalable in both width and depth (more bits/channels, or more memory). It is also scalable to much higher clocking rates. Currently, there are plans for memory that can transfer at 6.4 GB per second, though current technology is limited to 1.6 GB per second.

Rambus holds a lot of patents and has a lot of lawyers. They are an IP-only company; they live and die by their intellectual property. I would assume that they have done a better job to protect it than most corporations would. Because they have to give everything they know to manufacturers, everything they deem important must be protected by patents. They have a lot of patents, and many more on the way.

What is an IP-only company? Rambus does not manufacture anything. Instead, they sell the rights to manufacture or use their IP in other products. The going rate is currently up to 2.5% of revenues on RAM chips and 3% to 5% on controller chips. "Preferred" customers get discounts, so we generally assume about 1.5% on the RAM chips. Since Rambus doesn't make anything, they just research and develop (with their partners) and cash the checks. This is an untested model, but many people have noticed similarities in Rambus and Qualcomm, especially since Qualcomm dropped their handset manufacturing.

Why the jump in price now? Up until a month ago, RDRAM could only be found in game machines. Intel was late with its control chips, and it looked like there were viable alternatives to RDRAM about to come onto the market (DDR and DDR2). Well, Intel got its act in gear, the first RDRAM modules are being shipped, and DDR appears to infringe on up to 4 patents held by Rambus. And Rambus took a very long time to develop its case against Hitachi before announcing the suit, and they are talking tough and not backing down.

Rambus believes it will account for 10% of DRAM sales this year, 30% next year, and 60% the following year. These are huge numbers. This puts them in the $1B revenue range in 3 years. According to the Fool, they had $43M in revenue in 1999. That is a 20-fold increase in 3 years. Rambus is also currently engaged in adapting its technologies to solve problems in communications. That is potentially a huge market.

Rambus again has the blessing of Intel (though I am not sure they ever really lost it). In fact, Intel has bent over backwards to bring DRAM manufacturers on board. Intel has invested a lot of its time and its money, and I think they have a sizable stake in Rambus, both directly and indirectly through the supply chain.

Rambus technology will scale to higher frequencies as fabs get smaller. This is terrifically important. The current "open" memory architectures are reaching the end of their usefulness. It is becoming more and more difficult to scale them up to ever faster and wider processors. By tweaking the designs, the manufacturers may be able to go one or maybe two more generations, but it is getting more and more expensive all the time.

What does this mean? Well, it means an architecture change is inevitable, no matter what the chip manufacturers would like. They cannot stick with the same architecture or minor variations of it for much longer. Since RDRAM is now on the market, those who go through the expense of upgrading now will see rewards down the road.

What about the high price of RDRAM? Rambus claims that the cost to manufacture RDRAM is 40% higher than for other RAM types. This is supposed to drop to about 10% by sometime next year. How does this justify a 4 to 10 times cost differential that we are currently seeing? Most likely it is just the RAM manufacturers trying to line their pockets. They can't make the stuff fast enough right now to meet the demand.

So why the heck is Rambus stock so darn expensive? Is it a short squeeze, speculation, or what?

That is the really big question. Let's do the math.

Rambus earned 34 cents in 1999, is expected to earn 62 cents in 2000 and $1.84 in 2001. Taken as percentages, (I will be approximate here) that is 100% and 200% growth for the next two years. 2002 will likely be another 100% year, for about $3.60 a share in earnings. That gives us a more reasonable 55 P/E looking 3 years out.

Revenues will likely grow even faster than earnings. For that, you take the estimated market share, multiply by the size of the market, and factor in the assumed 1.5% royalties (could average to as much as 2%). 60% * 100B * 1.5% = 900M. That is something like 20 times what they made in revenues in 1999.

All this is assuming that only the RDRAM market is a factor. This is likely to be wrong. There are other markets for the chip-to-chip communication technology, and some of them don't involve RAM at all. And they are very big, but also very speculative at this point. Still, 5 years out, the picture still looks very good for Rambus.

So is Rambus worth its current valuation? Yep. Maybe even more. It looks high until you factor in the incredible growth that this company is about to see. Even after the run-up, I believe there is at least a 10-bagger in this company, and it could be a 25 bagger given a little more time.

What are the risks? We expect extreme growth, and without it, the stock will tank. RDRAM prices must come down to within 10% of other alternatives before any competing technologies have time to penetrate the market. Rambus must settle with Hitachi or win the case, and the outcome must favor Rambus in a big way. In the long-term view, Rambus must continue to expand its horizons with R&D. If it is a one-trick-pony, the stock will eventually fall hard.

On the bright side, things are looking very good right now on all fronts. I thought the Rambus/Hitachi suit was a good thing and took the drop in price as additional buying opportunity.

So what about the "short squeeze" that people keep referring to? We don't know yet if that is what happened. I suspect that there was a lot of short churn. As the stock rose, the old shorts got out and new shorts got in. Some people look at P/E and forget to do the math and the research on the company!

A lot of shorts lost their shirts in a very short time. Say that three times fast. But the stock is at an all time high, and it looks very expensive to anyone who does not look long and hard at the fundamentals. Taking a look at the volumes from the 15th on, it looks like there may have been a short squeeze on the 15th, but the volume declined sharply after that. Yet the stock price continued to rise dramatically.

Bottom line: the big squeeze may not have occurred yet. The stock could be driven, artificially, to $400 or more - soon - if the short positions in Rambus are still high.

If it goes to $400, I will probably hold on to my shares. I am expecting $1400 to $2000 a share by 2005.

If you read this all the way through, feel free to pick at my assumptions. But please be constructive and civil. I am here to learn, not to flame.

JasonX


Print the post  

Announcements

What was Your Dumbest Investment?
Share it with us -- and learn from others' stories of flubs.
When Life Gives You Lemons
We all have had hardships and made poor decisions. The important thing is how we respond and grow. Read the story of a Fool who started from nothing, and looks to gain everything.
Contact Us
Contact Customer Service and other Fool departments here.
Work for Fools?
Winner of the Washingtonian great places to work, and Glassdoor #1 Company to Work For 2015! Have access to all of TMF's online and email products for FREE, and be paid for your contributions to TMF! Click the link and start your Fool career.