Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Next installment reviewing Rambus reexams:
http://nicdagreek.typepad.com/my_weblog/2008/08/issues-56-when.html
Da Greek
More on Ware
http://nicdagreek.typepad.com/my_weblog/2008/08/ware-part-ii.html
Da Greek
On Ware ruling: FYI
http://nicdagreek.typepad.com/my_weblog/2008/07/but-first-a-bri.html
Da Greek
next installment is here:
http://nicdagreek.typepad.com/my_weblog/2008/07/one-clock-to-ru.html
Da Greek
FYI, next installment of reexam review
http://nicdagreek.typepad.com/my_weblog/2008/07/and-the-beat...
this chapter even has a happy ending....
Da Greek
FYI, next installment of reexam review
http://nicdagreek.typepad.com/my_weblog/2008/07/and-the-beat-go.html
Da Greek
The next installment, for your reading ... (dis)pleasure?
http://nicdagreek.typepad.com/my_weblog/2008/07/and-now-on-with.html
Da Greek
For those interested, I am going to post analyses for the Rambus reexaminations on this site, since it is easier to manage and focus the content:
http://nicdagreek.typepad.com/my_weblog/2008/07/ok-now-thing...
Da Greek
According to some lawyers who are now just reporting, the S CT looks like they are going to eviscerate the obviousness standard after hearing the KSR case. This does not bode well for patent owners.
Disclaimer: I have not heard or seen the transcripts, this is just the most recent news.
Da Greek
I posted this first on MF, and based on the positive response I am posting it here too:
Greetings folks,
Thought I would share with you some interesting legal developments, post ebay, on how the patent litigation world is coping with the lack of injunctions awardable against infringers. Some very creative lawyers in the ED of Texas have come up with a new theory that is being adopted by the Judges in that jurisdiction, and it goes like this: after a finding of infringement liability, the patent owner "opts out" of the injunction scenario and INSTEAD asks that the Court impose a compulsory license on all products going forward at the rate awarded by the jury, for the entire duration of the patent's life.
For those who can access PACER I recommend you look at the recent pleadings in the FINISAR v DIRECTV case.
Now this is a radical shift from the days when Courts only awarded past damages. What we see now is the patent owner saying, ok, I give up my injunction, so the infringer is free to continue infringing because the patent owner voluntarily agrees that he/she may do so; but at the same time they insist that the Judge must award ongoing DAMAGEs NOW for all uses/sales post verdict.
The beauty of this is that it removes a lot of the sting of the "no" injunction rule now for non-manufacturers after ebay. The infringer knows that if a big royalty rate is awarded, he/she is stuck with that rate, because the Court will impose it going forward on all future activities.
I raise this because in this particular scenario it bodes well for Rambus given the reasonable jury verdict figure in the Hynix case. They can easily tell Whyte that they opt out of an injunction, sla Hynix has a compulsory license at the 4.25% rate. Up to this point Hynix has believed that they could avoid such rate going forward after the appeal is heard and resolve. Now they could get stuck with it post appeal (if it holds up that is on other fronts).
If I were a betting man I would predict we such filing from Rambus at the appropriate time in this litigation.
The other interesting question is whether the Hynix verdict will be affected by the FTC ruling. Typically an infringement award is based on liability under a separate statute and can't be changed based on some extraneous factor such as what the FTC rules is a maximum contract rate that can be imposed. With that said the jury verdict itself was based on "contracts" which the FTC may now say were illegal, so query whether Whyte keeps or tosses the verdict after final remedy is pronounced. It will be an interesting 6 months, for sure.
I am entirely out of the stock now after this recent runup, I'm less shy now about cashing out gains as they come along. I may come back in if there is a significant hit on the FTC remedy. Right now this is more just intellectual/professional interest.
Da Greek
Perhaps NicdaGreek or one of the other board lawyers will post an educated opinion if this is good or bad news for Rambus, but expect the market won't like this news, regardless. God help Rambus if Sharon Prost is on the assigned panel.
It really is a mixed bag, so its hard to say whether it will turn out well or not. On the one hand Rambus gets to appeal the merits of Payne's decision (both IFX and Samsung really) so that gives them the shot they need to get out of this mess in VA. On the other hand you would prefer to have the Hynix case also of record since Payne's is so ugly and will likely contaminate the reviewing panel. So the current set of facts and findings is not exactly in Rambus' favor at this stage.
To be honest I can't predict how the rest of the world is going to perceive this at this point. The only real visibility will come when we see what Rambus appeals, and what their papers look like. When we see that we'll be able to make a better judgement. I strongly suspect Whyte will certify the phase I aspects, probably with Rambus' blessings, to get these appeals consolidated.
Da Greek
How long does a reexamine usually take, I wonder? Without studying the '916, in your experience do most patents survive reexamination by the PTO?
Typically about 18 months... but may be a bit longer with inter partes cases...
It is rare for a patent to be completely killed in reexam; the real danger, however, is that if one is forced to amend the claims - in any way - the requesting party can claim "intervening rights" and thus skate paying any damages for past infringement.
This can be a significant chunk of change of course, and it explains why the person waited so long. There is a rationale for waiting till the last minute: it erases a bigger number of damages if the claims are amended.
Whether Rambus will have to amend or not is beyond my vision right now, I don't know the claims or prior art at issue here - yet.
Da Greek
<Request application was filed with the PTO on August 8. Wonder which MM requested the PTO to reexamine Rambus patent 6,426,916, MEMORY DEVICE HAVING A VARIABLE DATA OUTPUT LENGTH AND A PROGRAMMABLE REGISTER??? >
They did not post the data yet, but since it is an inter-partes request, we should know soon who asked for it. My suspicion is that it is someone like Micron, whose case has not advanced very far...
<Anyone know if all the continuation and divisional patents in its history get reexamined as well??>
Just this patent; one has to specifically ask for the other patents to be reexamined. But it tends to telegraph the mindset of the infringer - in terms of which patents they consider most dangerous to their livelihood.
Da Greek
The only silver lining is that maybe the FTC will go away - but at what cost???
Beware the FTC.
I believe they are lying in wait, only dormant because it serves MU's interest for them to not reveal their true intentions until Hynix is over with. Why intervene now and potentially help a MU competitor? They'd rather watch the court proceedings, and if no miracle develops for Hynix, then spring the surprise. At that point, Hynix will still have a judgement against them, and they will still have to defend that case on appeal regardless of what the FTC does.
It serves both them and MU to wait at this point...
Da Greek
Anything you can do to rid me of these gerbils would be much appreciated!
A can of d-Con mixed in a beer ought to do it for you.
Don't look now, but I guess the hypesters missed big on this one...its not even going to hit the +5 I predicted.
Da Greek
Sorry to butt in, but I happen to have kept track of each person's price at the time of the prediction to get a rough idea what you guys were expecting the price to be.
Sorry to break the news Sherlock, but as you freely acknowledge, your data is almost 3 weeks old, and, no, I don't feel the compulsion to report every trade, twitch, or hair on my moles to this Board as some do. But if it help to put the little gerbils running your mind at rest, I already sold the May calls you are referring to as the price moved higher.
Da Greek
So you bought OTM May calls (45s?, 50s?), when the stock was trading in the 30's, all the while you only expected a $5 bump post verdict?
Who said I bought them when the stock was in the 30s? I made the $5 prediction almost a month ago, right?
My decision to buy OTMs was made a few days ago while we were still at 45+; that is now a fading memory.
Da Greek
Win Loss Win probability
Nicdagreek +5 –10 75%
KentP +5 -5 60%
Skeptic +6 -10 72%
So I guess the three FUDsters were the most accurate in predicting the reaction, huh?
Da Greek
PS: If its any consolation, my May OTM calls are now essentially dead and buried to the tune of some 20k unless there is a miracle soon.
Hi Kent
Just wanted to thank you for the nice job you did with the FUD.Yor posts have caused a number of readers to do their homework in order for them to remain LONG on RMBS.
Would you guys knock off the "Kent FUD" routine? Its getting kind of tired,not to mention grossly inaccurate.
Kent has been long in RMBS before most of you could spell it.
The fact that he doesn't kneel down and blindly swallow the Rambo-centric pablum formula a bunch of you choke down makes him a far smarter investor, and last time I checked, that was the point of all of this: to make money, not idolize a stock or make a religion out of it.
For the record this "skeptical" non-Rambus believer investor made a great return on Thursday when the panic set in, because I had invested objectively based on the odds I saw for the trial (75% long, 25% short). The puts got "called" away at a 3X, and it let me buy more May calls.
So next time you start criticizing someone for not idolizing all things Rambus, remember that a lot of them are making good $$ by investing with their head, not their heart.
Da Greek
Also, my understanding of why Karp was referring to the European patent is because, in Europe, patent applications are made public, whereas, here in the U.S., they are not made public until the patent has issued.
That makes eminent sense. L&L's post seems to suggest that he (KARP) referred to "the written description" also, so I'm more convinced that this was just a bit of confusing reporting and he (KARP) was not referring to the PCT claims when he gave his testimony...
Da Greek
That needs to be verified.
I agree; I posted the request on the YMB, and, of course, they were too busy cheering and high fiving themselves to give an answer and didn't want anyone raining on the parade. At the same time, I can't believe Hynix would have allowed that kind of thing to happen, so I suspect there is just confusion in the reporting. Either that, or the Hynix guys just set a new low for incompetency LOL
Da Greek
The Euopean Patent was written before Karp and Steinberg came to Rambus. It shows these claims were not invented by Karp.
You have this slightly wrong.
The European claims were not added until 1999, by, guess who? If you read the earlier versions, they too make no claim to the DLL, dual edge clock, etc. features.
I don't know what Karp was looking at, but it could not have been the claims of the European patent. If he did, and the jury does too, this will be a big mistake. So I'm curious "what" is in the ORIGINAL European version of the SPECIFICATION that Karp couldn't also point out quite easily in the US version. The two should be identical.
Da Greek
Yeah, that was me...I wish. You're obviously as clueless as to my identity as I am to yours. Check out my Yahoo profile. Do any of those guys look like Tim Robbins?
Ok, ok, so the fact that Susan Sarandon wasn't there should have clued me in LOL. But you are fairly tall, are you not, and you were wearing a suit?
That photo is of TDOX, SDHawgman and myself in front of the courthouse after the opening arguments. I'll let you decide which one is me.
I recognize TDOX on the right. Since its your profile I assume you are in the middle; if so, you look a lot like the guy I remember in the elevator lobby.
Well maybe I'll come to the conduct trial (if there is one....) and we can meet for real.
Take care,
Da Greek
BTW, Nic, were you in attendance? If so, any hints on your identity?
I did not get to go to closing unfortunately... but I did see Karp and Steinberg the week before... and if you were the tall guy in the elevator lobby on the 4th floor (kinda Tim Robbins look alike?) I was standing right in front of you with a blue suit on. I think you invited one other guy to go to lunch (don't know him...)
So you came THAT close - maybe the security guards will let you look at the old videos
Da Greek
First off, Furniss is not a 'good' looking guy (not that that makes any difference. Ha.).
Doesn't he remind you of this fellow from the old Clint Eastwood movies?
http://www.cinefania.com/pics/personas/5/5069.jpg
He must be terrifically talented, because he's no Richard Gere.
Da Greek
That's not true. The EPO patent from 10-31-91 (which is supposed to be an exact copy of the original patent), claim 5 states:
The semiconductor memory device of claims 3 or 4, wherein the clock generator circuitry includes a delay locked loop circuit coupled to the clock receiver circuitry to generate the internal clock signal.
Cal,
Just so you know, the original EPO patent is found here:
WO9116680A1: INTEGRATED CIRCUIT I/O USING A HIGH PREFORMANCE BUS INTERFACE
Here is the original set of claims through 5:
1. A memory subsystem comprising two memory devices connected in parallel to a bus, said bus including a plurality of bus lines for carrying substantially all address, data and control information needed by said memory devices, said control information including device-select information, said bus containing substantially fewer bus lines than the number of bits in a single address, and said bus carrying device-select information without the need for separate device- select lines connected directly to individual memory devices.
2. The memory subsystem of claim 1 wherein said bus contains at least 8 bus lines adapted to carry at least 16 address bits and at least 8 data bits.
3. The memory subsystem of claim 1 wherein said bus also includes parallel lines for clock and power.
4. A system comprising a memory subsystem of claim I wherein each bus of said memory subsystem is connected to its own transceiver device, a transceiver bus connecting said transceiver devices, and a means for transferring information between each of said buses of said memory subsystems and said transceiver bus, whereby memory subsystems may be integrated into a larger system having more memory than an individual memory subsystem.
5. The system of claim 4 having a plurality of memory subsystems.
Hence all the challenges concerning DLL as set out in later filings and, of course, the narrow multiplexed bus.
Da Greek
Hey Cal!
I'm not following you. Are you saying Hynix believes DLL wasn't added to the patent claims until the 'DDR patents' were generated?
That's not true. The EPO patent from 10-31-91 (which is supposed to be an exact copy of the original patent), claim 5 states:
The semiconductor memory device of claims 3 or 4, wherein the clock generator circuitry includes a delay locked loop circuit coupled to the clock receiver circuitry to generate the internal clock signal.
You are reading from the later EPO filing which includes amended claims added later. The DLL claims do not appear in the original, I will send you the document if I can find it on one of my computers...
And here's the '916 patent (which is in this case)
Right, the '916 was filed in 1999, with claims written by Steinberg without consulting either inventor.
So the point stands that the whole DLL thing looks like Steinberg's creation, rather than something that the inventors claimed as theirs. In any event, the fact that he did not bother to ever consult with them about it seems incredibly... irresponsible. I daresay you won't find another patent attorney to back up his comment that one does not consult with the inventors when filing continuation applications LOL
In any event, as I said the key thing is that Steinberg came across as credible; it probably does not matter that the jury does not understand the finer implications of what he said. I'm just not sure I would have said it the way he did.
Da Greek
these were continuation filings?
Doesn't matter. Continuation just means same specification; it does not mean same CLAIMS. And claims are how one defines inventorship.
he was in-house at the time, no?
No, he hadn't joined yet.
so the issues of inventorship, and his authority, and of "cost to prosecute the new filing" are diminished/eliminated - no?
No, that is irrelevant.
But judging from your comments, you can see how confused the jury would be, thus proving my point.
Da Greek
Free Report from Courthouse - Warning, Not LTL Approved
A free report from my visit to the Courthouse:
1) Karp was not finished, they had only started with him when the clock struck one and all the mice ran back to their chambers.
2) Steinberg was a hostile witness, so it is no surprise he gave testimony favorable to Rambus.
3) I didn't get to see Weinstein, wondered what he said about damages and a rate? If someone gets that data, please post so we can compare.
All in all I was struck by two things: (a) it is apparent that Hynix wants to show that the "inventions" in suit were really offspring of Steinberg's fertile imagination, and not from any real data from the inventors; (b) I'm not sure how understandable this was to the jurors.
Steinberg did an excellent job defending the patents, no mistake about it. But he did have his rather... weak moments. He pretty much admitted that there is no real description of a "delayed lock loop" in the patent. He had to say that one skilled in the art would "infer" it was there. But the problem is, he is not one skilled in the art, and he ackwnowledged never talking to either Horowitz or Farmwald about it. Ergo, Hynix wants to paint him as some kind of rogue money grubber who, under pressure from Karp's "we must get better patents quickly..." program simply wanted to impress Rambus with how creative he could be.
Ditto for "dual edge clocking" - he went round and round in circles, but there really is no such terminology anywhere in the patent. I think even the jury could figure out he was just making this up... So again he had to say that he thought it was inherently there, and Furniss pointed out that at one time Steinberg said in a deposition that " any EE with a "scintilla of intelligence" could see it there.
Well Hynix points out that then begs the question of why none of the skilled engineers at Rambus, or even the inventors, "saw" such invention for more than 9 years until Steinberg shows up. I suspect they are going to grill Karp, too, on the fact that his list of "new patents" to file in 1998 makes no mention of the DLL concept for example. If it was not apparent to him, Mr. DRAM designer, how was it "apparent" to a lawyer?
But all of this supposes that the jury understood any of this. From a visceral perspective Steinberg came across very well, so a lot of this detail may escape the jury or.... they may simply not care. I see this case much more critically than they do, so don't take my reactions as representative of theirs.
He also made some crazy comment to the effect that he never consults with inventors before filing new continuation cases? He said this to make it seem SOP that he did not talk to F&H. Let me assure this audience: this is either an outright fabrication, or Steinberg enjoys flirting with malpractice. If an attorney files a case, for example, based solely on his impression of what he/she thinks is patentable, and without consulting the inventors, I daresay that person will be in for a rude awakening. The PTO requires, for example, that the attorney confirm the inventorship of the claims as part of the filing; i.e., if there are two inventors named, please make sure that BOTH contributed to the application. Well, if you didn't bother asking either inventor, that seems like a rather signicant omission. That is definitely NOT SOP in the patent prosecution business, I was surprised he would say that. Again, though, I doubt this would even cause a blip on the juror's radar...
So I am left at the end of the day shaking my head, thinking, Steinberg is very creative, clever and charismatic, and that might be enough for him to pull it off... From reading the jurors, it did not appear that they grasped even 10% of what Hynix was trying to do... It sure didn't seem to rouse TDOX too much, the poor guy looks like he hasn't slept in a week, and he seemed to doze off for more than half the testimony.
Da Greek
PS: Danforth showed up late, pops open his Crackberry and starts whirring through his notes in plain view. I was astonished that a lawyer was not a wee bit more careful since I was sitting right next to him and he doesn't know me from Adam.
Nicdagreek +5 –10 ??%
Put me down for 75%
Da Greek
I guess that explains why I'm neither buying or selling right now....
You're making it too hard on yourself and reading too much into it. "Win" is defined by the market. If the Pop is positive, we have a "win".
In this scenario, Rambus could get a "win" in court, yet the share price *could* decline. In any event, the market will be the *true* judges if rmbs wins or not.
I didn't realize you were defining "victory" by the what the market does in response to the verdict. That seems a bit removed from the underlying question of what defines victory. To me a "victory" is translatable into later gains and it is not just a temporary bump, ala the FTC verdict.
Be that as it may, my only point was to note that there are is a lot of gray area in between the binary situation you posited, i.e. of a +12 gain and a -15 drop, because there is so much involved in defining what what is a "victory." I agree that the scenarios you chose are probably representative of a best/worse case. But I don't think we see either; I think we see a hybrid, so I don't see that kind of movement. On the positive side, I see a "semi-win" with average damages on DDR/SDRAM claims that is not going to bump more than 5 points, IMO. A more limited win with small damages and only a few SDRAM claims puts us down 10 IMO because I think the market is looking for more, and something extendable to other litigations. But I don't see a loss on validity, so -15 seems out for me.
Right now I own a nominal number of shares (10% of my portfolio), a bunch of ITM Leaps for 2007/2008 and short term MAY calls. As the trial winds down and the story more crystallized, I might gamble on some April options as well; I'm tracking their price as they should decline as we go forward and I hope to pick them more cheaply tomorrow than today.
Da Greek
My estimate:
Win = +12
Loss = -15
Probability for win = 85%
Define "win":
>500M judgement?
DDR (DLL) infringement?
No claim lost to validity?
etc., etc.
I can see situations where the stock gains 12, but it would have to be something along the lines of the above. Conversely, what is a loss? If Rambus gets less than $100m? Or does not win on DLL?
I can see the stock getting a hit if the verdict is small (less than $50m), the royalty rate nominal, and no separate infringement win on DLL.
So I'm curious what parameters you are using to define victory/defeat.
Da Greek
Had enough of San Jose
Going home.
Hey Cal,
Thanks for the post, it is useful as a benchmark. I guess we can now offially retire all the hyperbole about $1B from the pumpers, since they heard it from the horse's mouth...
BTW did Furniss/Hynix finish cross-examining him?
I wanted to be there but family circumstances prevented it.. sorry we could not hook up, I could have shown you my IEEE card LOL
Good luck,
Da Greek
So Rambus called a Hynix EE (i.e., JJ Lee) so he would testify on these same topics as Murphy but w/o contradicting Murphy?
That's how I see it.
They have to get SOMEONE from the other side to confirm how the Hynix circuits work. Its one thing to use your guy, but it makes a lot more sense to have their fellow confirm the details - after all, it IS their part. If your expert does his job right, you can usually corral the other sides' engineer to give you the answers you want. Its SOP, and I'm sure that was what Stone was doing.
Da Greek
Yo Nic-
Looks like Teece is up tomorrow.
Great report Biker, sounds like things are going well. Since my schedule is so tight I have to figure out I guess if I am watching Teece on direct or Teece on cross. I suspect Stone may not get through with him until lunch time.... so I will play it by ear.
Thanks again for the notes,
Da Greek
The CAS latency register was also covered in great detail. We’ll see how Hynix responds, as there’s been lots of discussion on this topic.
Thanks for the great report. Of all the people writing about this, I value your evaluation the most since the bulk of the other posters tend to be very selective or biased. I'm keenly interested in how you think the expert holds up on cross when his time comes.
Nic- don't know when Teece is up. I'll ask.
Super, I appreciate the heads up.
Da Greek
Cal,
If you get a sense of when Teece will go on, let me know.
Da Greek
Trial Schedule - Experts
Does anyone happen to have the witness list/schedule,with some kind of indication of the order of appearance? I'd really like to see Teece testify, but I can't afford too much time, so I'd like to peg any attendance to his testimony.
TIA
Da Greek
Something that could either be very very good or could be bad. That's as cryptic as it gets. Mark mentions cheating - sounds like a JEDEC reference was made. Perhaps Hynix is attempting to create a mistrial to create further delay...
See this:
http://www.bloomberg.com/apps/news?pid=10000080&sid=aGETh0VX1m24&refer=asia
From the press release sounds like Hynix wanted to smear Rambus as ineffective in licensing because rates are too high. Now Rambus wants to introduce evidence saying licensing was inhibited due to collusion.
So my guess is Rambus is complaining that they were barred originaly from using this evidence, and now want to use it based on Hynix's opening statement. Should be entertaining.
Da Greek
but you are correcting my thinking by reminding that it charged differently for SDRAM and DDR.
Right?
correct, although I am not sure again what the Hynix rate was; it might have been as low as 2.5%
Da Greek