Cor - at first I thought you were full of crap with that number but when I look at the most recent ......

Hi Will,

Happens frequently, mostly in the morning

This quarter is probably going to be the worst as the Infineon royalty has not yet kicked in, the serdes income is still small and there is no noticeable XDR production yet.

It is disconcerting to see that management (Hughes the financial engineer) has apparently decided to mask it with a phony piece of profit generated through the buyback. Would be better to simply face it and say, "look we are going to have zero profit this quarter due to xx, yy and zz" but the outlook is a, b and c".(and to be a bit more specific on for example the serdes ramping, it's 4 years plus now that we have been waiting for those)

On:
They do have 1.7 per share in net cash so as things stand now...

I would use the "shareholder equity", which is at $ 280 million approx. count in 10% dilution for the convert which is not counted yet because it would be "antidilutive" and you get $ 2.50 a share.

Does this strike anyone else as unseemly?

Yes, I do. And many do. It is all right when things are going great, but the ratio of exercises to earnings does matter.
http://www.secform4.com/insider/showhistory.php?cik=917273

Cheers,
Cor

Did you have a SELL? I don't think so. You stated RMBS will have a 400% INCREASE in earnings in 18 months. I believe you listed RMBS as a hold.-Cal

Maybe our readers could read better than you did. But then they had the benefit of the full text
(which you are not entitled to)

Anyway I know what most of them did and it was to sell or substantially reduce their positions and one even shorted. (which we did not recommend, but it wasn't a HOLD as you suggest)
Those who were not in but had our May report which definitely was a BUY have been holding off.

Nuf said about this, you're not going to believe it anyway.

Cheers
Cor

Late call by him, but he was responding to an appeal for an opinion at TMF today, in fairness.

True enough, Threejack.

In that same thead after having been asked by Gotta I also responded and lifted a piece of the SemiReview veil.
http://boards.fool.com/Message.asp?mid=22946049

Obviously without some view on settlements (or a win), we will see the 12-month trailing eps go down in the coming months. On the basis of current earnings, one can defend a stock price of approx. $ 5.

We will need something more than accounting tricks (I am referring to the buyback of part of the onvert, which apparently they have the attention of reporting as a profit, whereas one can calculate that in fact it was a loss transaction) to reverse the course of the stock price.

I am still firm about the target we set initially, although I am getting unsure about the time scale on it.

As you know, we (semireview.com) wrote on June 7th, with the stock price at $ 15.67:(amongst others)

"Due to this uncertainty and the delays associated with it, we do not think it is the right time to buy Rambus. We would even suggest buying Put options for existing positions to protect against the likely intermediate downside.

At the end of the day, no matter how good this development may be for the value of Rambus' IP long-term, the short term story here is that Rambus just picked a fight with one of their biggest, and possibly most important customers, and there will be repercussions."

Sadly enough, we were right. We are 30% down in stock price from that time. I don't think the slide has stopped yet. I also have stopped believing in "catching a falling knife". If there is a massively positive development legally (for example in the Hynix case), there will still be time to get in with maybe a loss of the first coupla dollars of a rise.
======================================================
That is what we wrote and we probably saved out subscribers quite a bit of money writing that in our update, even though we are firm on the further out future.

(It also answers Cal's question some months ago how we could not have a BUY on it given the other things we wrote

One has to read those things in context and in toto.



Cheers
Cor

IDF Summary

by Brian Dipert (I know this guy from insightful articles about memory trade-offs)

http://www.edn.com/blog/400000040/post/1130001313.html?nid=2434

IDF: Reading Between the Lines

Aug 23 2005 10:12PM | Permalink | Email this | Comments (2) |
Blog This! using: Blogger.com | LiveJournal |

It's been a busy day here at the Intel Developer Forum in foggy San Francisco. When I left Sacramento late afternoon Monday it was nearly 100 degrees and the sun was blazing bright....two hours later, after my train-plus-bus ride here, it was 35 degrees cooler. I don't care if Mark Twain never did say 'The coldest winter I ever spent was a summer in San Francisco", it's still an apt quote (although he clearly never visited Northern Indiana in late January). But I digress...

As predicted beforehand, Intel CEO Paul Otellini rolled out three new (i.e. above and beyond the roadmap unveiled at the last IDF) power-tuned 65 nm-based processors at his keynote this morning. Here are a few tidbits:

1. The new CPUs are Woodcrest (server), Conroe (desktop), and Merom (mobile), based on a common micro-instruction core which combines the best of the NetBurst (Pentium 4) and Mobile (Pentium M) predecessors, and all touting EM64T 64-bit support.
2. Silicon for all three products is running very well. Otellini's presentation ran on a Merom-based notebook.Conroe was shown running 64-bit Fedora Linux, and Woodcrest ran Windows Server 2003.
3. Conroe will deliver a 5x performance-per-watt improvement over Northwood (the initial iteration of the Pentium 4).
4. Dempsey, Presler and Yonah, introduced at the Spring 2005 IDF, are also running and on 65nm. And two more single-core 65nm products are also running.
5. By Q3 2006, 65nm CPU shipments will exceed 90nm shipments. Also, by Q3 2006, dual-core CPUs will outship single-core CPUs. Over the next 18 months, 60 million dual-core processors will ship.
6. Looking beyond dual-core, over 10 quad-core and beyond projects are underway.

Otellini's presentation, which also covered a broad range of other topics, was followed by a more indepth press-only briefing given by Stephen L. Smith (VP of the Digital Enterprise Group) and David Perlmutter (VP and General Manager of the Mobility Group). I'll begin by providing links to a few of Smith and Perlmutter's foils:

* Revised CPU roadmap (note the new 65 nm products, and the four-plus core chips at the tail end of the decade)
* Single-to-multi-core transition forecast
* Intel's micro-architecture history
* Intel's next-generation architecture
* Intel's next-generation architecture (foil 2)
* Intel's next-generation architecture (foil 3) (particularly note the shared L2 cache, the L1-to-L1 cache direct communication link, and the dramatically shortened pipeline compared to the 20-stage pipeline in the first-generation 'Northwood' Pentium 4 and the 31-stage pipeline in today's 'Prescott' Pentium 4)
* One micro-architecture, three platform spins (note the common die between platforms, likely differentiated by clock speed and operating voltages)
* Platform timeline

A few bombshells from the presentation; all three of the new CPUs will be single-die dual-core configurations (unlike Presler, which will implement a dual-die, single package approach to 'dual core'). And none of these initial implementations of the new micro-architecture will support HyperThreading, although Perlmutter didn't rule out a resurrection of HT in the future.

Later that afternoon, Ronny Korner (the CPU validation manager for the Mobile Microprocessor Group) gave an in-depth presentation on Yonah, Intel's 1H '06 65nm, dual-core followon to Dothon (today's Pentium M). Again, here are a few particularly tasty foils:

* High-level specifications
* Architecture block diagram
* Multimedia-tailored enhancements over Dothan
* Yonah's Smart Cache
* Increasing performance while simultaneously reducing power versus Dothan
* The Sossaman spin of Yonah for blade and rack servers

Much speculation pre-show centered on whether or not the power-thrifty and clock-efficient Pentium M would be the foundation for Intel's next micro-architecture. I strongly suspected it would, and based on what I heard today I'm confident I was right. Simplistically, if you take Yonah, add a Pentium 4-like front-side bus, Virtualization Technology and other minor enhancements, and vary the L2 cache size.....you have Woodcrest, Conroe, and Merom.

Or at least that's what I think. Your thoughts? For more IDF coverage, see the usual suspects; AnandTech, Ars Technica, ExtremeTech, The Inquirer, The Register, etc...even MacWorld's here (for the first time)!

This article has many links embedded to foils etc, so worth reading for who wants the nitty gritty.

Cheers
Cor

I think they do have a booth.

Booth comes with the silver sponsor thing.

So they definitely have a booth.

Cheers
Cor

PS I am the only one here who has ever visited a IDF?

We're going to have plenty of information about FB-DIMMs for you coming over the rest of the week. In the meantime, it looks as though DDR3 is only going to be on graphics cards for the foreseeable future.

All depends what "foreseeable" means

DDR3 is already on the AMD roadmaps for example for 2007.

http://epscontest.com/presentations/05q2_analyst-day.htm?slide=60&a


Cheers
Cor

PS Elixe called my attention to that series of slides on TMF.
And I can never remember how to post those things here as a graphic.

Hey Nxt

Did you get my emails with the attachments?

Cheers
Cor

Threejack: Geez, went right by me. I am getting old.

I am getting old too, but better to get old than to die young

Cheers
Cor

nxt,

on:
At least I could see XDR in the most recent memory roadmap press release...albeit from May 2005.

Right, also it is shown in two pages (I believe 3 and 18 w/o looking) of the cmos memory catalog:
http://www.semicon.toshiba.co.jp/eng/prd/memory/doc/pdf/bce0004f.pdf

So if they really cancelled the XDR product they would have taken it out of there imo. Anyway why do that when the initial license fee has been paid already, one can just hold production until there is demand, maybe run a few test wafers every now and then.

The flash market is really hotting up, so there will be some switch of capacity from dram to flash and both sides prices will go up. Good quarters for the MMs in view.(2 probably)
see for example:
http://www.eetimes.com/showArticle.jhtml?articleID=169400638

Apple Computer Inc. plans to buy as much as 40 percent of the NAND flash output of Samsung Electronics Co. in the second half of this year

Cheers
Cor

He has a free account and will not be able to see your PM message here.

Did not realise that, Threejack, thanks.

Well if he post an email address I can sned it to him, or alternatively he can mail me at anybody@chipstocks.net

Cheers
Cor

Miserable corporate governance?

(also a repost from TMF (one of mine): http://boards.fool.com/Message.asp?mid=22909712)

Yahoo profiles give as a new feature an indication of the quality of Corporate Governance. The numbers given for Rambus are terrible.

I wonder what it is? The exec. compensation, options scam or what?
I have not found the formula by which it is computed, but anyway this is what is given at:
http://finance.yahoo.com/q/pr?s=rmbs

Rambus Inc's Corporate Governance Quotient (CGQ®) as of 1-Aug-05 is better than 2.9% of Russell 3000 companies and 1.2% of Semiconductors & Semiconductor Equipment companies.

Those are horrible percentiles.

What yahoo says about this is:
Corporate Governance Quotient (CGQ®), a corporate governance rating system provided by Institutional Shareholder Services (ISS) on over 7,500 companies worldwide, evaluates the strengths, deficiencies and risks of a company's corporate governance practices and board of directors.

CGQ uses a comprehensive set of objectives and consistently applied criteria for each of the companies rated. The database features corporate governance rankings on more than 7,500 companies worldwide, and includes underlying data points for up to 61 corporate governance variables, categorized under eight areas of focus: 1) board of directors, 2) audit, 3) charter and bylaw provisions, 4) anti-takeover provisions, 5) executive and director compensation, 6) progressive practices, 7) ownership, 8) director education. Some of the variables are reviewed together under the premise that corporate governance is enhanced when selected combinations of these variables are adopted. To facilitate analytical comparison, each company is scored individually, relative to its market index and its industry group.

The data provided in CGQ is derived primarily from public
disclosure documents, press releases and corporate websites.
It then goes through a rigorous verification process by ISS'
senior corporate governance analysts and is put in to the CGQ database.

Two CGQ ratings are generated on each company. Each company's CGQ is compared with other companies in the same index and industry group.

* All scores are relative (percentile basis)
o CGQ Index Score: compare to relevant market index including: S&P 500, Mid-Cap 400, Small-Cap 600, Russell 3000, ISS Small-Cap, S&P/TSX
o CGQ Industry Score: compare to industry peer group using 24 Standard S&P "GICS" groups

Users can access CGQ data on Yahoo! Finance's Company Profile pages.

If I were Rambus management I would look into how that rating can be improved. Maybe some more education

So I was getting a little curious and decided to call up some more out of my "chipstocks universe". This is what came out:

(first column is rel. to Russell 300 or S&P500, second column in industry peer group: "Semiconductors & Semiconductor Equipment companies".

 
 ADI		 7.4	34.5 
 AMD		97.4   100.0 
 CREE		88.7	95.1 
 INTC		78.0	94.5 
 MIPS		71.3	72.6 
 MU		67.7	92.1 
 NSM		46.1	87.2 
 RMBS		 2.9	 1.2 
 TXN		37.7	84.8 
 XLNX		29.7	81.7

Something strange at Toshiba re XDR.

Hi guys,

I am reposting this info from TMF, where Elixe noticed this:
(http://boards.fool.com/Message.asp?mid=22911338)
quoting Elixe:

Toshiba XDR missing

It's looking more and more that Toshiba has dropped XDR from it's memory portfolio, at least for now.

All data and specifications for XDR have disappeared from their web sites.

With Samsung and Elpida sampling chips, few customer commitments to XDR, a late market of relatively small demand generated by the PS3; Toshiba may have made a business decision to drop it for now.

If true, this has to be of concern to Sony and Kutaragi. Without strong demand, I believe Samsung and Elpida will charge high prices for product. As I recall, Samsung initially sold RDRAM for >$40.00. And without demand, the price learning curve will be very shallow.

JMOs, though.
==================================================
I feel that Toshiba's priorities are now very much with increasing production capacity of flash memory which is having a great season (Toshiba partners with Sandisk).

But the disappearance of the datasheet off the site is strange although the mos memory product catalog still shows XDR memory.

Cheers
Cor

"TMF board has been too OT for my tastes lately."

Hey Pruf,

I am the King of On topic posts on the TMF Rambus board
But you are right the OT threads are longest...

Cheers
Cor

WOW, interesting stuff, Threejack....

having seen the Intel entry, I also had a look at the AMD entry (much less to spend of course):
http://publicintegrity.org/lobby/profile.aspx?act=clients&year=2003&cl=L000054

Thanks for the links.

Cheers
Cor

Short interest is indeed up, Threejack.

see:
http://tinyurl.com/bqtt3

"Rambus, Inc. - Common Stock 12,273,374 10.52 1,400,574 8.76"

I believe that includes the 4 million shares which were shorted by the convert holders earlier this year. If so, effective short position is at most 8+ million shares.

Another way of looking at it, is that about 11 million shares can be had by converting the converts. We don't really know how many of those have already been shorted in a trading mode of hedge funds (covered by the converts).

Cheers
Cor

Also, agree it is an amazing coincidence a fellow named L.F. Payne heads up MWC and shares the same last name as His Nibs. Common name, Payne, but an amazing coincidence nonetheless.

Hi Threejack,

Apparently somebody on yahoo called the guy and asked him (whether he was family of R.P.). He replied " not that I know of", which I think is kindy funny:)

Cheers
Cor

Ah yes, Intergraph (INGR), held it for a bit more than a year for a double at that time, and some calls. Good memories:)
Didn't have time to look at the market, was sailing on the Med for about a year then. But knew it would happen in the end (this was after infringement had been established and Intel kept pestering INGR)

Cheers
Cor

Thanks, Threejack, I can probably get the latest Riley report.

Cheers
Cor

Threejack, do we know who the two analysts are in that yahoo ae?

One has an estimate of 2 cents and the other of 5 cents. I believe they could well be both too optimistic.

Not just the missing Samsung revenue but there may be more not yet renewed licenses, although they may kick in later. rdram will have been slow (one of the big users, CRAY, was not doing at all well) and to the extent that memory prices matter they were down about 30% versus the previous quarter.

I have to check the exact wording in the Q&A of the CC to hear what revenue Eulau estimated for qtr3.

Cheers
Cor

Hey Threejack

I have been inputting backlog into my spreadsheets including Hynix financials.

I honestly don't know how they did it. Being such a one-product (almost) company in a commodity market with duties against them in two continents, they made a profit in every quarter of 2004. For example IFX did not manage that from a much easier base.

(it helped a bit having that fab in the States so they could use that for US consumption and bypass the duty, which is fair game).

Anyway, I 'll keep tracking them a bit closer.

Cheers
Cor

PS two more market days to go before earnings , the clock ticks

Hi Threejack,

I see that you also have some paranoid traits (like me:) , but one learns as a Rambus investor

re:
Maybe the Hyundai chaebol traded the stock up to achieve a better equity for debt deal in its latest creditor shuffle; this round the locals took the loans.

Hynix wasn't really in receivership, that deal was made quite a while ago (about june 2002) with the creditors after the whole theater about the potential MU takeover (karate chopping Koreans in the streets, many). The point I was trying to make is that those banks (a.o. Korean Exchange Bank) are part of the chaebol, it's all one big clan.

But on the cash, they are spending it fast, going to build a fab in China. (no more in the States

Cheers
Cor

Hey Threejack,

I was not referring to Hynix being tough in litigation (not my thing, legal nitwit, remember); I meant it relative to doing business under adverse conditions (high debt, duties around 40%+ in EU and USA, low dram prices). They really did well in their business, that stock price did not shoot up without some background.

Also they sold a major division and managed to make a profit. What happened a few years ago (documented on my site of course:) is that a lot of debt was changed into stock, so the banks (members of the chaebol club) became part owners.

If by Chaebols you mean cabals, not sure who you are referring to; the new creditors are not a cabal.
Look here to see what I meant, one has to understand the concept of chaebls to udnerstand Korean business:

http://www2.sjsu.edu/faculty/watkins/chaebol.htm

Cheers
Cor

Hi Threejack and Paul,

Hynix is the best performing semiconductor company over the past two years, one year, six months ... of the top 30 semi companies as far as the stock price goes.

see for some details:
www.chipstocks.net/hynix/hynix-top.php

compare with others and you will see.

The story fails to mention that they also divested a whole division (non memory) via Citybank last year. (and made some good profits despite EU and USA rulings against them and very high duties) They are a lot tougher than I ever expected, or let's say the Chaebols won't let one of their own die.


Cheers
Cor

Samsung replies available from litigation update on Rambus site:

1) re Rambus vs. Samsung:
http://tinyurl.com/9jx8g

(pdf, 26 pages)

2) re Rambus vs. Hynix, Inotera, Samsung ... (DDR2 etc)
http://tinyurl.com/7m3yj
(pdf, 31 pages)

Should keep eveybody happy, eh busy, for a few days.


Cheers
Cor

Hi Newcombe, Proximity Communication:

Yes we discussed this last year July on TMF (don't know whether you have access, trial is free, I believe).

The thread is at:
http://boards.fool.com/Message.asp?mid=21082683&sort=whole

I wrote inside that thread:
Here is a Reuters link:
http://research.sun.com/async/RelatedSites/Versions/SunMicrotoPresent2.html


snip:
By placing the chips edge to edge, directly touching, so data can flow freely, Sun has taken out the need for the tiny wires, pads and solder points that now connect chips on printed circuit boards that help make up computer systems, Sun said.

The breakthrough could mean sending data among chips up to 100 times faster than current top transmission rates on traditional semiconductor-chip interconnects, Sun said.

It would also solve one of the oldest challenges in the chip industry: the bottlenecks that crop up when chips -- which are getting ever faster -- are connected to one another.

"It's faster, cheaper, and uses less power," John Gustafson, principal investigator for Sun's high productivity computing systems," told Reuters.

Sun already holds seven patents on the new design and will seek to capitalize on them commercially, a Sun spokesman said.

another Sun link:
http://research.sun.com/async/RelatedSites/Versions/Sun_Shows.html
snip:
Gustafson believes proximity communication will make today's circuit boards "as obsolete as the vacuum tube," but that day seems far in the future, according to Richard Fichera, a research fellow with Cambridge, Massachusetts analyst firm Forrester Research Inc. "Until every piece of technology out there gets integrated with this, it's not the end of the circuit board," he said. "It's a very interesting way to connect between high-density chips."
"In three to five years, we could see this in a rollout of products from Sun, and perhaps some other selected partners, but as a de facto standard? No way," he said.

Here is a related patent application from Sun:
http://tinyurl.com/47tlj
snip:
[0005] The present invention relates to semiconductor devices, including multi-chip semiconductor devices, and methods of coupling semiconductor devices. In a particular embodiment, the semiconductor device is a multi-chip semiconductor that comprises a first semiconductor device and a second semiconductor device. The first semiconductor device has a first surface. The first surface contains a first ridge alignment member and a second ridge alignment member, the first and second ridge alignment members forming a receiving area between the first and second ridge alignment members. The second semiconductor device has a second surface, the second surface containing a third ridge alignment member, the second semiconductor device positioned such that at least a portion of the third ridge alignment member is located within the receiving area of the first semiconductor device.

Here is the paper presented by the Sun fellows.
http://research.sun.com/async/Publications/KPDisclosed/sml2003-0241/sml2003-0241.pdf

Proximity Communication

This paper reports results from wireless chip to chip
communication experiments. Sixteen bit words pass from
one chip to another in parallel without detectable error at
1.35 billion data items per second for a total data rate of
21.6 Gigabits per second. The experiment transmits pseudo
random patterns between chips built in 350nm CMOS
technology. Chips touch face-to-face to communicate. .....

This paper is truly interesting stuff and only 4 pages to read.

(see especially figure 8)

Anyway if Sun gets the contract from DARPA, they will build a computer with this in 2006. If not they will continue this research anyway.

Cheers
Cor

Very interesting stuff, will probably be tried out by sun first, I have not yet checked if they did get that Darpa contract.

What does it take to get a BUY?

You keep trying for some freebies, hey Cal

Frankly it takes a LOT, if we define a BUY rec as a recommendation of buying the stock NOW for today's price. Such recommendations are a moving target and subject to the situation at hand.

We don't have a SELL rec on Rambus either;)

Cheers
Cor

Rambus Investor Relations are nonresponsive to even simple information requests.

Hi Threejack,

I believe Rambus IR is not inside Rambus, but is an external firm handling it.

The C2C insider newsletter was mostly technical marketing (and some outrageous predictions were made once in it, so probably the person to talk to would be the VP of Marketing.

Cheers
Cor

It's hard to believe cor would be hyping RMBS. Ha. We've come full circle.

I am not. We don't even have a BUY rec on Rambus right now.

Subscribers have the benefit of the full argument, the pros and cons, the upsides and the downsides, the risks...
(and not only about Rambus)

I am not going to get drawn out here paragraph by paragraph on the 20 pages we have already written about Rambus.

Have a nice weekend

Cheers
Cor

Would that be a best case scenario, or guaranteed? Ha. What's your probability of reaching $35 in 18 months? My prediction of $50 to 75 in 3 years would be a best case scenario.

Hmm, 35 in 1.5 years would probably lead to 50-75 in 4-5 years

Hope you can provide the probability analysis before I take off. Ha.

I think I missed your subscription payment, must check paypal, LOL.

Have a great trip. What's the plan? Biking, rafting?

Cheers
Cor

Hi Nordicprince:

on:
.however, it seems to me that samsung's agreement to pay royalties constitutes an implicit agreement that the rmbs' patents are valid and enforceable...especially in the absence of any verbiage to the contrary

We can unfortunately now know exactly what is in there. But it is very well possible that Samsung has this worded such that they are covered (given that there was litigation about the patents already going on then).

They could say: we are going to pay, but we don't ackonwledge patents unless proven in court in the current cases or something like that. The fact that they got a better deal later without significant opposition from Rambus indicates such.

We'll just have to wait how it all pans out. The wait can get long in this case. First case management conference in October.

Cheers
Cor

WSJ on addition of Samsung to civil AT suit:

http://online.wsj.com/article/0,,SB111888690873761173,00.html

Rambus Adds Samsung to Lawsuit
June 15, 2005 9:58 p.m.

Rambus Inc. said it added South Korean electronics company Samsung Electronics Co. to its existing antitrust lawsuit against DRAM manufacturers Hynix Semiconductor Inc. and Micron Technologies Inc. In a filing with the Securities and Exchange Commission Wednesday, Rambus, which makes technology used in memory chips, said it took the step after a series of other developments, which included South Korean Hynix's plea agreement with the U.S. Department of Justice. The filing said Hynix promised cooperation in the DOJ's investigation of a possible DRAM industry conspiracy to fix RDRAM prices and otherwise eliminate competition in the 1999-2002 time period.

The dogs of war are loose:(

Cheers
Cor

and we AGREED that rmbs had the patents...

That part is not necessarily in the agreement between Samsung and Rambus. In fact, I am willing to bet it isn't.

And that is the only reason why Samsung could during the VA trials get the payments reduced to fixed payments. No way could they have achieved that otherwise.


Cheers
Cor

Cal, a very good summary of the technical situation.

Rambus is not stagnant; they keep innovating. They have turned their business from a one dimensional RDRAM business to a true chip interface business. Their PCI-E designs are top of the line. FlexPhase is great. They have the fastest parallel bus in the industry. XDR is the fastest DRAM. RaSer X technology is the fastest backplane serial link in the industry. Rambus is active in building design centers around the world as well as providing design seminars to in the US. They are aggressive in providing top notch turn-key designs.

As you say, the expectations of $ 1000 share price (if one ever had those) need to be toned down to something realistic.

We at SemiReview.com have a 18 month target of $ 35, which we have not changed after June 6. (we have issued a 2 page update about the Samsung situation and possible short term repurcussions though)

Such a target should be enough to keep Rambus in a diversified tech portfolio, what does it matter if the view is for a double or a tenfold increase? This is only the case for people who just bought one stock (and I kow some did) in which case you want the one with the highest reasonable target. The people who are most disappointed are the ones who had too much of the protfolio in Rambus, were margined, used shorter term options or LEAPS; that is what is giving a "snapback" effect.

The first page of our Rambus report of last month is available free here:
http://semireview.com/reports/SemiReview-RMBS-Page1of16.pdf

We have a special offer on the first year of the subscription here: ($ 400 instead of $ 600)
http://www.semireview.com/special-offer.htm

Earlier reports which are free are at:
http://semireview.com/reports/tsra-BUY050211.pdf

http://semireview.com/reports/SIPI-Report.pdf

Cheers
Cor

Hi Threejack, I agree that TMF piece was shallow at best and negative and bashing at worst.

I do not quite understand why they do this. These are amateur writers, who get paid a fee per story by TMF. But they are put in the publicly available area of TMF, so one must assume that they are meant to advertise TMF features, not so much the message boards but the various newsletters.

Different from the normal usance this time was that TMFTwitty did not put a link to the story on the Rambus message board.

Anyway it produced the to be expected angry reactions and we'll see what TMF has to say about that

Cheers
Cor

The new Samsung filing can be found at my website. Twobytebus scanned it and I put it up. TMFers have had access to it for a day now and I can have a bit more traffic, but not of the Yahoo kind.
Those who are interested can download it from there.

So please do not repost these links:

http://www.chipstocks.net/rmbs/samsung-complaint1.jpg

until

http://www.chipstocks.net/rmbs/samsung-complaint10.jpg

(page 9 has the "prayer for relief", invalidation of patents)

Cheers
Cor

re: Just to be sporting, I will predict Infineon exits the DRAM business by 2010.

Very sporting, Threejack, to predict something which happens in five years time

Please correct me if I'm wrong, but seem to recall it wasn't that many months ago (less than a year, maybe) this was its intention before it decided to pursue DRAM market share.

Yes, you have your wish, I will correct you
They never said that. They said (already about 2 years ago) they wanted to become less dependent on the commodity dram market. To achieve this by:
1) growing the rest of the business faster
2) concentrating on higher value dram (low latency, cell dram, GDDR etc)
3) outsourcing part of production so that they don't get hurt so bad in the downcycles.

By all signs they are doing this. They achieve a higher asp for dram than Micron for example. Growing the rest of the business is more of a problem, although in some sectors they are doing well. (wireline division is a problem child)

In any event, consolidation will occur and one or more of the Thieves will be out of the DRAM business.

Yes, consolidation will continue, it has been going on for quite a few years and Infineon was one of those growing in it. Hard to predict now if Hynix or Micron or Samsung will be out of it; for Hynix and Micron it would be not out of the dram business, but out of business. Samsung won't happen in any case. Infineon is the least likely as they have a nice cost structure now


Cheers
Cor

On the other hand, it would be poetic justice if Infineon went under after negotiating a lifetime license to Rambus Memory IP.

Hmm, dram is only about 40% of Infineon's business.

I predict it does not go down, but picks up some market share from the battling "amigoes" ..

Cheers
Cor

Nor a leak out of Samsung

Agreed, but Samsung does not have a history (habit) of putting the dirty linnen on the washing line outside as Rambus does

Cheers
Cor

Look at those dates, Threejack, 3 days ago and not a peep out of Rambus...

Cheers
Cor

This helps to explain why Infineon decided to settle with Rambus. Perhaps Mr. Hughes got as much as he could expect under the circumstances given Infineon's current financial difficulties. Assuming Rambus meets the hurdle for the contingent payment, wonder if Infineon will be able to pay off the remaining $100MM?

Hey Threejack

These stories about how IFX would not be able to pay turn up from time to time. I have always (for about 3-4 years now) on TMF said this is poppycock, if anybody can pay it is Infineon.

The story is rather tainted. (well reporters gotto eat

One thing Schumacher accomplished (by whatever means) was to propel Infineon into the top five semi companies. (from nr 6 to nr 4 in a few years, after some formidable ones like Intel and Samsung and Texas Instruments). This is no mean feat.

Also note from the article;
Infineon is not fighting for survival - the company has a solid balance sheet -

Cheers
Cor