Documents Reveal Rambus’ Secret Scheme; New Evidence Re-Ignites JEDEC Debate
By Steven Fyffe, Electronic News
Feb 16, 2001 --- Documents uncovered in the ongoing legal battle between Rambus Inc. and several DRAM makers could bolster claims that Rambus deliberately broke the rules of the Joint Electron Device Engineering Council (JEDEC) as part of a secret plan to take over the global DRAM intellectual property business.
The documents include internal memos and e-mails, Rambus’ 1992 business plan and handwritten notes. Hyundai’s legal team successfully had the documents’ confidential status removed, but they remained sealed until Electronic News filed a motion with the U.S. District Court of San Jose to release the documents. A judge did so today.
Some highlights:
From a Rambus business plan dated June 12, 1992:
“Finally, we believe that Sync DRAMs infringe on some claims in our filed patents; and that there are additional claims we can file for our patents that cover features of Sync DRAMs. Then we will be in a position to request patent licensing (fees and royalties) from any manufacturer of Sync DRAMs. Our action plan is to determine the exact claims and file the additional claims by the end of Q3/92. Then to advise Sync DRAM manufacturers in Q4/92.”
DRAM makers say they were not informed of Rambus’ plans until after 1997.
Email from Richard Crisp, then a senior executive at Rambus, dated Wednesday, May 24, 1995. (Crisp describes some of the technology presented by the SyncLink consortium at a JEDEC meeting):
“…As far as intellectual property issues go here are a few ideas: 1) DRAM on a packet oriented bus 2) DRAM with low swing signaling 3) DRAM with a two wire initialization system 4) DRAM with programmable access latency 5) DRAM with on-chip address space decoding.
“I think it makes sense to review our current issued patents and see what we have that may work against them. If it is something really key, then we may want to mention it to Hyundai in our attempts to get the negotiation under way again. If it is not a really key issue, such as the initialization issue, then I think it makes no sense to alert them to a potential problem they can easily work around.
“We may want to walk into the next JEDEC meeting and simply provide a list of patent numbers which have issued and say ‘we are not lawyers, we will pass no judgement of infringement or non-infringement, but here are our issued patent numbers, you decide for yourselves what does and does not infringe.’ On the other hand we may not want to make it easy for all to figure out what we have, especially if nothing looks really strong. If we have a really strong one that has issued that is key to the operation of the SLDRAM, then we may want to play that card, but again with the above suggested disclaimer.”
Rambus patented programmable latency, one of its key SDRAM patents.
Email from Richard Crisp, Dec. 5, 1995. (He discusses JEDEC’s rules for disclosure.):
“…The things we should not do are to not speak up when we know that there is a patent issue, to intentionally propose something as a standard and quietly have a patent in our back pocket we are keeping secret that is required to implement the standard and then stick it to them later (as Wang and Seeq did). I am unaware of us doing any of this or of any plans to do this.”
Hyundai, Micron (nyse: MU) and Infineon (nyse: IFX) claim Rambus (nasdaq: RMBS) stole ideas from JEDEC meetings and quietly attached them to pending patents it filed in 1990. They say Rambus broke JEDEC’s rules and kept the patents secret until after they had been granted in 1997 when it turned around and used them to blackmail the industry into paying royalties on SDRAM.
“(The documents) clearly show that Rambus’ business objective was to manipulate the JEDEC standards to claim patent coverage,” states Hyundai’s motion to have the documents unsealed. “Indeed, one document expressly states that Rambus’ business objective ‘depends on getting a standard, which depends on our patents.’
“They do show that Rambus was well aware of JEDEC rules requiring patent disclosure, and of the JEDEC discussions. This is highly relevant to Hyundai’s claims here and in Germany that Rambus violated JEDEC rules to obtain the broad patent coverage by secretly amending its patent applications to cover ideas and technology discussed at JEDEC without disclosing this to the JEDEC members.”
Others put it more bluntly.
“This is stealing other people’s ideas,” said Farhad Tabrizi, vice president of worldwide memory marketing at Korea-based Hyundai. He has been chairman of various JEDEC committees, representing Hitachi and later Hyundai.
“This is very unethical and against any kind of human decency. The whole idea of open standards is sharing ideas in public for the benefit of the public. Now people are afraid of other companies filing patents and charging the inventors themselves. These documents show that this problem exists.”
Los Altos, Calif.-based Rambus has a different take.
“We feel that they have infringed our IP, and that we deserve to be compensated for the use of our patents,” said Avo Kanadjian, vice president of worldwide marketing at Rambus.
“What we are saying is that even before entering into the very first JEDEC meeting, Rambus shared its inventions and its concepts with a number of the industry players under NDA (non-disclosure agreements),” Kanadjian said. “It’s ludicrous to say we came up with this idea during JEDEC sessions.”
But because Rambus’ SDRAM patents had not yet been granted, it did not reveal them to companies that had signed NDAs, sources said. The only patents those companies were aware of covered Rambus’ own direct Rambus DRAM (RDRAM).
The JEDEC issue has emerged as the key to the DRAM makers’ case against Rambus. Lawyers from both sides have dubbed Boise, Idaho-based Micron’s anti-trust action against Rambus “the JEDEC trial”.
“I must confess I find interesting Rambus’ statement…‘This is just a patent case’,” said Micron attorney Matthew D. Powers from Weil, Gotshal & Manges LLP, Menlo Park, Calif., in a recent conference call with Rambus’ lawyers and the court.
“You couldn’t try the patent issues without the JEDEC issues,” Powers said. “We’re concerned that because Rambus wants so desperately to avoid a trial on the JEDEC issue, that it will continue to delay.”
JEDEC wants to stay neutral, but the documents could raise some troubling issues, said John Kelly, president of JEDEC.
“It’s fair to say that evidence like that is very troubling,” he said. “I have heard from guys in the industry that they regard this as the smoking gun, but you really don’t know until you hear testimony from the person involved.”
The JEDEC process is on trial, alongside Rambus and the DRAM makers, Kelly said.
“It’s a big case for Rambus and it competitors. But it’s a big case for us too, because it really does focus on the standard-setting process,” he said. “The entire process is grounded on trust.
“It is like an athletic competition or a card game or anything else where you have to assume that people are playing by the rules. If one side decides to violate the rules and does it in a way that is difficult to detect, it undermines everyone’s confidence in the rules and in the game,” Kelly said.
The high profile of the Rambus case has monopolized media attention, but other JEDEC-related cases are brewing. Fountain Valley Calif.-based Kingston Technology Co. Inc. accused Sun Microsystems Inc. (nasdaq: SUNW), based in Palo Alto, Calif., of similar abuses of the JEDEC standard-setting process after Sun launched a memory module patent suit against Kingston.
“I don’t think that Sun and Rambus are the only two companies struggling with this,” said John Sutherland, a fellow at Kingston. “I think you will find in the coming months that there are several companies that have been playing games at JEDEC and patenting things that will become industry standards so they can make some pretty good money out there in the market place.”
If the courts side with the DRAM makers’ version of events at JEDEC, Rambus could face near ruin overnight. Rambus is heavily dependent on royalty revenues, with 80 percent of its earnings come from royalties, while contract revenues generate about 20 percent.
Rambus’ stock valuation would also take a massive hit. Financial services company Morgan Stanley Dean Witter’s “strong buy” recommendation on Rambus stock is based on the assumption that Rambus will win. If Rambus loses, all bets are off.
“Our investment thesis on RMBS is predicated its ability to successfully defend its SDRAM, DDR SDRAM, and associated controller intellectual property (IP),” read the most recent report from Mark Edelstone, semiconductor analyst at Morgan Stanley Dean Witter.
“Our strong buy rating on the stock assumes that Rambus will prevail in the upcoming trials with Hyundai, Infineon, and Micron Technology. However, investors should be aware that we believe there is risk inherent in any legal proceedings, and a defeat could potentially negate our investment thesis.”
A verdict in Rambus’ favor would be just as disastrous for the DRAM makers, Tabrizi said.
“God forbid, should Rambus win, DRAM companies will have to look at making memories other than SDRAM and DDR,” he said.
electronicnews.com
I'll go with Edelstone.
jack |
