Rambus Inc (RMBS) 2005 Q1 法說會逐字稿

Good day, everyone, and welcome to this Rambus first-quarter 2005 earnings conference call. Today's call is being recorded. At this time, I would like to turn the call over to Mr. Bob Eulau. Please go ahead, sir.

Thank you, operator, and welcome to our conference call. We are pleased to provide you with information on our results for the first quarter that we just announced. My name is Bob Eulau, and I am Rambus' Chief Financial Officer. With me today are Harold Hughes, our CEO, and John Danforth, our General Counsel. In the first part of the call, Harold will discuss our business and first-quarter accomplishments. Then I will provide a summary and analysis of the Company's recent financial results. Next, John will provide an update on litigation. Afterwards, Harold, John and I will be available for your questions.

The press release for the results discussed here today has been filed with the SEC on Form 8-K. If you want a copy of the release, please visit our website at www.Rambus.com, on the investor relations page under financial releases. A replay of this conference call will be available for the next week at 888-203-1112. You can hear the replay by dialing the toll-free number and then entering ID number 5446227 when you hear the prompt. In addition, we are simultaneously webcasting this call, and a replay can be accessed on our website beginning at 5:00 Pacific time today.

Before we begin, I need to advise you that the discussion today will contain forward-looking statements regarding our financial prospects, pending litigation and demand for our products, among other things. These statements are subject to risks and uncertainties which are more fully described in the press release and other documents that we file with the SEC, including our 8-K's, 10-Q's and 10-K's, and these statements may differ materially from our actual results.

So now, let me introduce Harold Hughes, our Chief Executive Officer.

Thanks, Bob, and good afternoon, everyone. I am happy to announce that we delivered record revenue in the quarter, closing out Q1 at 39.6 million. That's up 22% over first quarter last year, and up 3% sequentially. I will let Bob provide greater details on the financials later in the call.

In addition to strong topline performance, we made progress on many fronts during the quarter. The first area I would like to highlight is our continued leadership in technology innovation, clearly the undisputed core of the Company. We recently demonstrated our DDR2 memory controller technology, running at 800 MHz, as part of the Denali MemCon Show last month. This is the first time that anyone had publicly shown the DDR2 memory controller running at that speed. And while we continue to be at the forefront pushing memory interfaces to unprecedented speeds, we're also making innovations that deliver unmatched performances on other parts of the computing platform. In this corner, we unveiled a new technology for DRAMs called micro-threading. Applying our micro-threading technology to a DRAM core enables a significant increase in efficiency, resulting in up to four times greater performance, when compared to traditional DRAM applications such as 3-D graphics and advanced video imaging.

Naturally, we have filed patents of our micro-threading innovations, which, in conjunction with other research efforts, brings our overall patent portfolio, both issued and pending patents, to 789. The innovations presented in this broad portfolio have worked their way into higher-volume shipments than many might imagine. In fact, in this last quarter, the industry passed a milestone of over 0.5 billion RDRAM devices shipped, a tremendous accomplishment for any semiconductor technology or device. RDRAM during this time on the market has seen wide acceptance in applications such as game consoles, personal computers, high-definition television's and even in network routers. While RDRAM continues to go forward, we have a successor, XDR DRAM, the world's fastest DRAM technology, ramping up for high-volume production.

During this quarter, Samsung announced that it had begun volume production of 256 MB XDR DRAM devices. In addition, Toshiba and Alpida announced that they are now sampling the next-generation higher density, the 0.5MB version of XDR. Just to provide some local color on the performance of XDR, Toshiba announced that its devices are operating at speeds of 6.4 GHz. That is roughly eight times faster than best-in-class traditional DRAM performance.

We're also seeing momentum in areas of logic interfaces. Here, our innovations in the area of high-speed signaling are realizing significant adoption, as we help customers enable their next-generation products. To date, we have signed nearly 50 licenses through our logic interface products, half of which are PCI Express. As PCI Express becomes the mainstream solution for personal computers, the PCI Express shipments continue to ramp, we will realize the benefit of this focus. Further, PCI generation one will be followed by generation two and others, and as data rates increase, our value proposition grows commensurately. We stand ready to help our customers today, and we will be ready to help them in the future.

The next area I would like to touch is our progress that we have made in patent licensing. Earlier today, we announced that we had signed an agreement with NEC Electronics, whereby they receive access to DDR2 memory controller patents as well as some future innovations. NEC is one of the contracts that expires this year, and with this new agreement with NEC Electronics, we are anxious to expand our efforts helping them bring advanced products to market. NEC Electronics joins Infineon as our patent licensees signed thus far this year.

Meanwhile, we have a number of other patent license agreements which are due to expire in 2005. Needless to say, we're in active negotiations with these customers, and we will keep you updated as these agreements are brought to fruition. I'm optimistic that we will be successful. We believe the Infineon agreement that we signed a few weeks ago will contribute very positively in advancing our other patent license negotiations, but I just want to reiterate a point I made during the conference call announcing the Infineon deal. The terms of the agreement with Infineon reflect unique circumstances, and should not be extrapolated to apply to other negotiations.

Continuing with customer news, in February, along with Sony, Toshiba and IBM, we unveiled details about the revolutionary Cell processor. We believe we play an integral role in that overall relationship, and we appreciate the trust Sony and its partners have placed in Rambus to make the Cell processor a reality. We have designed all the memory and processor bus interfaces on the Cell, providing an unprecedented aggregate I/O bandwidth of 100 Gbps, a number unimaginable not that many years ago. The Cell processor uses both our XDR Memory Interface technology, as well as our Flex I/O processor bus technology. We look forward with great anticipation to our customers' future announcements of products taking advantage of the tremendous computing power of Cell.

And with that, I will turn things over to John, who I suspect has a few things to say.

I made a mistake. I will turn it back to Bob, who will talk about the financials.

Thanks, Harold. Let me share some of the highlights of our financial results of the first quarter of 2005 with you. Our first-quarter revenue increased 22% over the first quarter of 2004 and, as Harold mentioned, this is a new record for us, at $39.6 million. Our operating income and net income decreased 43% and 47%, respectively, when compared to a strong first quarter of 2004. Cash flow generated from operating activities remained solid, at 7.4 million. Total cash and marketable securities were up 224 million from December, mostly as a result of the convertible debt offering that we concluded during the quarter. I will discuss the balance sheet and the convertible offering again shortly.

First, I'll give you additional information on revenue. Total revenue for the quarter was 39.6 million, up 22% over the first quarter of last year and up 3% on a sequential basis. Total royalties for the quarter were $33 million, up 20% over the first quarter of last year and up 1% sequentially.

I want to address why these results are better than the guidance we gave three weeks ago. When we issued our most recent guidance, we were using a royalty report from one of our larger licensees that proved to be incorrect. The report was corrected by the licensee, who promptly paid royalties of about 1.6 million higher than we had expected at the time we last updated you. This is frustrating, but a reflection of the reality we face in relying on our licensees' royalty reports. You will notice that the accounts receivable were also a little higher than normal. This reflects the 1.6 million that we received in the first week of April associated with that one customer. We perform royalty audits with our significant customers, but there will never be 100% assurance that our customers are reporting their royalty obligation to us correctly. We intend to continue to perform audits periodically with major licensees.

SDRAM and DDR royalties from memory devices and controllers were up 42% over the first quarter of last year and up 3% sequentially. If you exclude the Intel cross-license agreement, the revenue from SDRAM and DDR memory and controller royalties comprised 52% of our revenue for the first quarter of 2005.

I'd like to make a couple comments on the license renewals that we need to make this year. We announced this morning that we had signed a patent license renewal with NEC Electronics. This renews one of three major contracts that expired as of March 31st. We are in very active negotiations with other licensees on renewing their contracts. We cannot predict when and if these renewals will be completed, but under the contracts that expired at the end of Q1, payment is still contractually required in the second quarter. There's another significant contract that expires on June 30th, and payment for that contract is expected through the third quarter.

If you look at last year's royalties, which totaled $120 million, 66% of the dollar value of those contracts is under contract through 2005. The amounts they pay may fluctuate, but I mention this so that you get a better picture of the challenge that we face this year. It's a good signed that Alpida and NEC Electronics have renewed, and we remain optimistic that we will meet or exceed last year's level.

Moving on to contract revenues -- contract revenues were $6.6 million this quarter, up 30% over the first quarter of last year and up 9% from last quarter quarter. The increase in contract revenues in the first quarter of 2005 over the first quarter of last year reflects the continued progress we are making on XDR, Flex I/O and RaSer contracts.

With regard to costs and expenses, total costs and expenses for the first quarter were $34.7 million. This was up 45% from the first quarter a year ago or $10.8 million, and up $5.8 million from the previous quarter. Cost of litigation was up $7 million, compared to the first quarter of 2004, and up $3.6 million on a sequential basis. This increase reflects, among other things, the conclusion of litigation activities with Infineon and preparation for the Hynix and Micron trials, which John will discuss in a moment.

Operating expenses, excluding costs of litigation, were up 20% when compared to the first quarter of 2004, and up 10% compared to last quarter. The increase in operating expenses, excluding costs of litigation, over the first quarter of last year was largely attributable to increased investment in the logic interface business in the engineering area and an increase in compensation expense in marketing, general and admin areas. The increase in compensation expense was associated with hiring in several departments and the expense associated with the restricted stock grant to our new CEO. The increase in compensation expense also largely explains the increase in operating expense, excluding costs of litigation, over the previous quarter.

As previously mentioned, we officially opened our design center in India this quarter, and plan to increase our investment in engineering talent and infrastructure associated with the Bangalore operation. We view the ongoing investment in the center as an important step in lowering our overall cost structure over the long run.

Net income for the first quarter was $4.4 million or 11% of revenue, compared to 8.3 million in the first quarter last year and 10.5 million in the previous quarter. This continued profitability led to solid operating cash flow, which was $7.4 million for the quarter. Our cash, cash equivalents, short- and long-term marketable securities ended the first quarter at $460 million versus 236 million as of December. Our cash, cash equivalents, short- and long-term marketable securities were up $224 million from last quarter, primarily due to the convertible bond offering and partially offset by the simultaneous $75 million stock repurchase.

I would like to make a few comments on the private placement of the convertible bond offering. We have previously been constrained in our comments, due to the disclosure rules regarding discussing the offering while it is taking place. We believe the cash from the bond helps us strategically, since companies with cash in the semiconductor industry can execute their strategy without compromise. In our case, this means we can execute our litigation and engineering strategies for many years without compromise. The other reason we did this deal was a result of the terms we were able to achieve. Our understanding at the time we did the deal was that the terms were the best terms of any midcap tech offering in the last seven years.

Now, I will give you some thoughts on what to expect in the second quarter of 2005. This guidance obviously has a lot of uncertainty associated with it, but it reflects our reasonable best estimate at this point in time, and our actual results could differ materially from what I am about to review. We estimate our revenues to be in the range of $38 to $42 million, and that our operating costs and expenses will be in the range of 34 to (technical difficulty) million dollars. We expect to continue hiring in the next quarter, including increased investment in sales and marketing, our Bangalore design center and additional US engineering.

Litigation spending is always the most difficult to predict, because we do not control the timelines and requests from the courts, nor do we control the actions that our adversaries may take which cause us to incur additional expense in any particular quarter. Evidence that is produced in hearings can also cause us to change our plans. Based upon this, we estimate that our litigation expenses will be in the range of between $8 and $11 million, but this range is very dependent on the activity levels of our ongoing litigation.

We are estimating that net interest and other income will be between $2.5 and $3.5 million. And finally, our tax rate is likely to be in the range of 36% to 38% next quarter.

So, now, John will update you on litigation.

Thank you, Bob. Good afternoon. I want to focus my remarks today on two sets of issues -- first, the ramifications that we see, as we see them, of the Infineon settlement; and, second, the path ahead in our remaining cases, particularly the Hynix case, where we have uncovered possible new evidence.

First, let me address the Infineon settlement. The Infineon settlement has now caused us to terminate an amendment to our Samsung license, and that amendment has been in place since 2001. That license now reverts to its original terms and needs to be renegotiated this year, in any event. Beyond that, in assessing the legal and practical impact of the Infineon settlement, I think it's important to stress that the Infineon case in Virginia had been ongoing for almost five years, that that case had gone badly for us almost from the start, and that that case had, we believe, been in the way of our ability to resolve many other matters. You will recall that two sets of rulings in Infineon -- a fraud verdict in May 2001 and adverse claim constructions earlier that year -- although both later conclusively reversed by the federal circuit, had led to significant compromises with certain licensees, including Samsung. And those rulings created years of delay in our other licensing and litigation efforts. For example, they pushed back the Micron and Hynix trials by several years. We hope to avoid similar delay now.

Apart from being the first Rambus case to go to trial, the Infineon case was also, in our view, unique in many other ways, that in Virginia twice precluded us from being able to litigate our patent claims on their merits. Some of the unique aspects of the Infineon case have been identified by me in the past. I have, on a couple of occasions, discussed the track record we had in that case. And some of the unique aspects of the Infineon case are also identified in the collateral estoppel brief we filed this week in the Hynix matter in the northern district of California.

Because of these factors, we believe the Infineon settlement does not, from either a practical or legal perspective, create any form of template for other cases or licensees. There are other, purely procedural, reasons why, as our brief this week in the Hynix case makes clear, the preliminary rulings of the Infineon trial judge ought not to govern the outcome of any other cases now pending. The issue of whether those rulings should govern, which involves what lawyers call collateral estoppel, is set for hearing in San Jose in the Hynix case on April 22nd.

Moving ahead to the path in front of us for our other cases, I can report as follows. The Hynix case had previously had a schedule in place that consisted of the April 22nd hearing on collateral estoppel, followed by a May 9th trial on the Hynix defenses of unclean hands and spoliation, followed in June by a trial of Rambus' patent claims, followed in October by a trial of Hynix's various counter-claims.

This past Monday, at a hearing we requested with the court because of newly-discovered evidence, the May and June dates in the Hynix schedule were taken off the calendar. The April and October dates remain in place, and at the April 22nd hearing coming up next week, the parties expect to discuss new dates for the unclean hands and patent trials.

I want to anticipate and, to the extent I can, try to give preliminary answers to a number of questions raised by this newly-discovered potential evidence and by the Hynix trial schedule changes. Because these tapes are newly-discovered, and because we don't know what they contained or all of the circumstances that surround them, I will not speculate on what significance there may be, if any, to this newly-discovered evidence. I will say that while many of the backup tapes we found are blank, others do contain data, including email backups, from the mid-1990s, and we are now therefore devoting very significant resources to trying to restore these, to retrieve the data from them, and to review them for information responsive to discovery in our various cases. We also believe that we understand the circumstances that led the these tapes not being included in earlier document searches and productions, but I am not prepared to discuss details at this point until we have assured ourselves that we understand this as fully as possible.

As to the impact of these tapes on the Hynix and other trial dates, we are hoping to be in a position to assess this prior to the April 22nd hearing. At present, we think a delay longer than the 45 days suggested by Hynix may be required. That is why we believe it is appropriate that the Hynix court proceed as it has, and that the Hynix file schedule remain in flux until we have more information, which we hope will be by April 22nd.

Another recent event may also bear on timing in the Hynix case, and that is the successful recent intervention of the US Department of Justice in our case to prevent us from taking discovery as to DRAM price-fixing or communications between the DRAM industry and the US Department of Justice. Currently, that limit on our discovery in Hynix -- that is, in the Hynix patent case, as sought by the DOJ to protect its ongoing criminal investigation of Hynix and as ordered last month by the discovery master in the Hynix case -- has no time limit. While we hope the time impact on our case will, in fact, be relatively short, we may appeal at least the time period of the stay; we can take an appeal to the trial judge. Also, limits obtained by the DOJ on similar price-fixing discovery by other parties in other cases are, to my understanding, subject to deadlines and re-review which fall this July. And there are signs that Hynix may be -- I say this is speculation, but Hynix may now be contemplating a settlement of the DOJ's price-fixing claims, since Hynix recently announced that it has increased its reserves for price-fixing, fines and/or damages to approximately $350 million.

So, putting all this together, we hope the DOJ requested limit on our discovery will be relatively short in duration and will, in any event, not become the gating item on when we can complete the Hynix patent case in San Jose.

Moving briefly to our other cases, we have seen efforts by Micron to try to align that case with the Hynix and Infineon cases by seeking discovery of the same privileged documents the Hynix and Infineon reviewed relating to our document retention, and seeking to hold an early hearing on alleged spoliation issues. We have resisted Micron's efforts in this regard. We think the evidentiary hearing on spoliation requested by Micron is premature, and we are suggesting that such a hearing, if required, should follow in time the hearing in the Hynix case and ought not, in any event, be set until there is more visibility as to the newly-discovered Rambus backup tapes.

Currently, the Micron patent case continues to have a trial date in late February 2006. We anticipate that the parties and the court may want to assess this date and the existing pre-trial schedule now, in light of developments in the Hynix case, particularly the backup tapes we have found.

In closing, I want to touch on two other, more recently filed cases -- the DDR2 case we filed with San Jose last quarter and the antitrust case on file in San Francisco Superior Court. We have no trial date, of course, yet in the DDR2 case, and discovery has not yet begun. In the antitrust case, we had our first status conference last quarter, after a long and successful fight initiated by the other side to keep venue of the case out of San Francisco. We prevailed, and the venue is going to be in San Francisco. At that hearing, the San Francisco Superior Court trial judge indicated that he would at the next status conference -- which is set for April 22nd -- make rulings on a protective order, and the discovery on our antitrust claims would begin immediately thereafter. In that discovery, we hope and expect to get some or all of the documents produced by the DRAM companies in connection with other pending DRAM price-fixing cases.

At present, those are my comments. Thank you very much. I'll hand the mike back to Bob.

Thanks, John. We are now going to take questions, and also respond to some questions that have been sent to us during the quarter from shareholders. We couldn't address every question. We've tried to address a number of the questions in the text of the material we just went through. And then we will answer a couple of questions as we go through here.

(OPERATOR INSTRUCTIONS). Gary Mobley, B. Riley & Co.

Your royalty revenues grew nicely year over year and 1% sequentially, and I think it was better than you expected, and you explained that. But could you also help us understand the disconnect between your royalty trends in the overall DRAM market and what some of the incremental contributors may be there, with respect to RaSer and XDR?

We are very pleased with the royalty number for the quarter, and what has happened over time is our royalties have become more stable than the DRAM industry, because we are not solely dependent on the (technical difficulty) because of the memory controller business. In addition to that, from those that we are getting royalties in the DRAM business, several of the deals are on a fixed, per-quarter amount. So we don't expect to see the same level of volatility that the DRAM industry sees.

And any RaSer royalties in the quarter?

There were RaSer royalties. Once again, they were really pretty small, and not at a level that is very material yet.

And XDR potentially generating royalties next quarter? Is that a fair statement?

We really don't know precisely what our customer plans are. We are hopeful that we will begin to see volume production soon.

On your last conference call in January, you were fairly conservative on your expectations for contract revenues. I think you highlighted your contribution on Cell might be concluding. Is that still your stance? And why the better-than-expected contract revenues for the just-completed quarter?

So a couple of questions there. I think, first of all, in the quarter we just completed, the engineering team did a great job, and they got a lot of work done on the contracts that we did not fully anticipate at the beginning of the quarter. So that's a good effort on their part. Going forward, we definitely, based on the contracts we have at this point in time, expect contract revenue to go down. And we said when we originally signed the deals with Sony and Toshiba that we would be recognizing the revenue on those contracts over (technical difficulty) the last quarter (technical difficulty).

Daniel Amir, WR Hambrecht.

Good quarter. I guess I have a couple questions. One, maybe you can comment a bit on -- two of (ph) your products have been on the market traction (ph) of the XDR. I guess you commented a bit on some of the traction you are getting, but maybe if you can provide a bit more clarity what you are seeing out there for the year? And also comment a bit on the serial link business. And I have a follow-up question.

I'll take a stab at those two. First of all, on XDR, you have seen announcements from some of our licensees that they are sampling and moving into production. And we really don't know their production plans, so it's very difficult for us to give more color on what is going on than that. We are pleased with the progress. We know that the products are in good shape; we have worked very closely with our licensees. But I really can't give you much in the way of a forecast there.

In terms -- and what was your specific question with respect to the serial link business?

A bit about what you're seeing in the marketplace there, a bit on visibility, revenue, et cetera?

Well, in the serial link space, PCI Express continues to show good traction, and we have got a number of deals we find related to PCI Express. Overall, we find in excess of 40 licenses with various customers for different implementations of serial links. So in terms of customer traction, we are pretty pleased. In terms of royalties, we frankly wish we were a little further along at this point, but we are hopeful that the royalties will come, in time, as we continue to sign these deals.

Now, the other question is just to clarify the issue of the agreements that have now -- I guess you have two customers that now, on the agreements ended here on March 31st and one agreement that is ending in June. How does the outlook, then -- they continue paying contract revenues based on the current production, at current rates, but basically it's being renegotiated on any new terms like DDR2, for instance?

I'll take that, since I made that comment. The main thing I was trying to do is give you some more visibility on what's going on this year. We had said for quite some time that this was a big year in terms of renewals, and that we are very pleased with the renewal with NEC Electronics that we announced this morning. There are a couple of other contracts that are quite material that concluded as of March 31st.

Now, under the existing contract, they still owe us payment in the second quarter, and we fully expect to receive that. And the teams are in very active negotiations with those two companies, as well as the others that are up for renewal this year. So we are well aware of the risks, but we are also pretty optimistic that we will get those deals done.

(OPERATOR INSTRUCTIONS). Mike Crawford, Barrington Partners.

Bob, further to the license renewals, I believe your top five royalty licensees back in 2001 were Intel, Hitachi, Samsung, Toshiba and NEC. So is it Hitachi, Samsung and Toshiba that you are trying to get renewals for now? Are those the three companies that you mentioned, referred to?

I don't want to get into specifics on which customers and what stage. We will definitely be promptly announcing license agreements as we sign them going forward. As I mentioned, there are two other fairly significant customers that expired March 31st, and we are in discussions with them.

And could you also confirm what was said about the Samsung license after the Infineon settlement? I'm not sure if I heard that correctly.

Let me make very clear what I said. We had an amendment we entered into with Samsung in 2001. That amendment was cancelable or terminable by us upon the happening of certain conditions. The Infineon settlement satisfied those conditions, so we have terminated the amendment. Therefore, the contract, as it originally stands or as it originally was drafted, is now in force. Now, that contract has a very short period of time left to it, so we will have to renegotiate it, anyway; but that is the current state of play.

So that is the contract that is in force until June 30th?

Correct.

And also, John, can you talk a little bit more about these tapes that were discovered? These were Hynix's backup tapes?

No. I can tell you they are not Hynix's backup tapes. These are tapes which are backup tapes from the Rambus email system, we think. And we have been able to restore only a small number of them, so I really don't want to go into further detail until we have a very high level of confidence that we can speak definitively to this. But it's the Rambus -- not the Hynix -- email system. And these, we think, include -- among the data that is there appear to be at least some backups of emails from the mid-1990s.

And regarding the Hynix discovery that is on hold because of the DOJ, did the special master -- this is the stuff pertaining to the ADT and JRA groups?

Actually, no. This deals with our efforts to get documents showing that the DRAM industry colluded on pricing. And that is relevant to some portions of the Hynix case that is currently set for trial this year. The DOJ intervened to stop us from getting two categories of documents. One was communications by Hynix with the DOJ, and the second was a broader category of documents which would show price-fixing by Hynix with other DRAM manufacturers.

And so has there been a ruling on the other discovery that at one point was authorized and was then put on hold into the -- I think it's the JRA group?

The JRA discovery is currently -- we have been given a copy of the JRA agreement. And frankly, there have been other things going on; I haven't checked the current status. But I believe that we will hear soon what the status is of the other JRA documents in the Hynix court. And I apologize; I don't have the most recent information on that.

Gary Mobley.

I'm just trying to flesh out a little more detail on the NEC license agreement. Could you give us the scope of the agreement previously, and what is incremental about the new agreement?

I'm reluctant to go extensively into the details of this agreement. It covers memory-related products.

And this is for some of their SoC designs?

It covers controllers and, if they decided to make it, DRAM.

But that is in Alpida now, right?

The centerpiece of the NEC Electronic license was DDR2 Memory Interfaces.

John, how does the Tessera amended suit against the memory IDMs -- I guess, essentially trying to coincide with your antitrust suit against the memory IDMs -- how does that impact your dealings in that case?

Well, it's an interesting development. On the one hand, I assume it is going to create some level of overhead for me in responding to discovery requests relating to these issues. On the other hand, because they filed in a jurisdiction that moves very quickly, I guess there's a possibility here that Tessera will try some of these issues before we do. I hope that's responsive to your question.

Yes, it is; thank you. One last question --

Let me add one last detail. Conceivably, Tessera will come up with a ruling -- which, of course, we are not a party to that case, right? But one of our opponents is, Micron. So Tessera is pursuing these claims against Micron, and theoretically there could be fact findings against Micron that Tessera obtains that are to the benefit of Rambus down the road.

And Harold, if you can talk about what do you think the biggest catalyst will be to get the licensees that are up for renewal to re-sign? What's your greatest point of leverage with those guys?

The power of our patents; the extent to which we can work with them to produce products which allow them to be successful, so that they look at us as much as an engineering partner as they do just a patent licensing company; the extent to which the new capabilities that we have announced recently will allow them to create new products around those pieces of technology and, by so doing, get a jump on non-licensed DRAM manufacturers. That's the main thrust of our negotiations and discussions.

Keep up the good work.

Trying hard.

Okay. I'm going to move on to, actually, one shareholder question, I think. The other question was really a series of questions regarding the tapes, and I think we have said about all we can on the backup tapes at this point.

So, Harold, the question came in -- is it a reasonable expectation that mainstream PCs will need XDR?

Let me answer it slightly differently. Getting on the Intel roadmap, as it were, is one of my top priorities. Actually, after executing on our legal agenda, that is my number-one priority. It's an area, obviously, where a lot of water has gone under the bridge, and it relates to many of the things that are currently under court actions. We have relationships with Intel through the fact that we are cross-licenses. We have relationships with other parts of the PC platform, such as graphics companies. And, obviously, we have licensed some of the larger DRAM companies on XDR. And turning that into a proposal that is attractive to Intel and PC OEMs is our job over the next series of quarters. It won't be easy, but it's a commitment that we are undertaking.

Great. So thanks, Harold. I'd like thank everyone for their questions today and your ongoing interest in Rambus. We are very pleased with the start of 2005, and we look forward to the rest of the year.

This does conclude today's conference. We thank you for your participation, and you may now disconnect your lines.