Slashdot Mirror
Court Decision Favors Rambus
RoscoeP writes "This story from News.com: "A federal judge has overturned two counts of fraud against chip designer Rambus...". At least Rambus can't pursue litigation against Infineon for SDRAM though." See our previous stories about Rambus for far-too-much background.
It's nice to see a judge standing up against large companies, to stop them from throwing lawsuits around against smaller companies
I'm not positive...but I believe Infineon is a larger company than Rambus.
"Sure it gives us geeks with our open source a bit of a chuckle, but it does make at least a little business sense to not screw over your 3rd party developers, especially in a world where a good number of your sales are driven by the fact that most 3rd parties still develop exclusively for your OS. "
.. )
That is a problem with Linux.
Nobody gives a damn about end users, and I don't only mean real end users but also developers.
Half of binary software that came with RH 6.0 will not work with 7.0..
What kind of bullshit is that? It almost seems like RH is bend on pissing off commercial developers ( who end up protecting themselves using static links, which results in tons of duplicated static code
Well, not yet. This is what they are fighting about. But if Rambus prevails, this is what will happen. Let's just hope they die a horrible death.
___
If you think big enough, you'll never have to do it.
No, it does not.
99% of Win32 software created 5 years ago will work on Windows ME.
You are the one who doesn't know what he is taking about.
IANAL, but will go out on a limb to say that the job of US courts is to interpret according to US laws, not laws of other countries, or the EU. US courts do not fall into line behind the decisions of other EU courts, unless the judge is senile.
A feeling of having made the same mistake before: Deja Foobar
its more about getting liscensing for DDR
Licensing for Dance Dance Revolution? They'll have to ask Konami about that.
Oh, you meant "Deutsche Democratic Republic"? Sorry, West Germany bought them out way back in October 1990, even before the Internet had a World Wide Web.
Oh, you meant "double pumped SDRAM".
Will I retire or break 10K?
Clearly this is false since both VIA and SiS are coming out with P4/DDR chipsets this month (and Intel with a P4/SDR one). The only thing delaying the DDR version of Brookdale is validating it (this is Intel's first DDR chipset while VIA and SiS have more experience with it).
1. Kick RAMBUS out of JEDEC, and make all other member sign agreements not to divulge anything to anyone at RAMBUS. 2. Have JEDEC hold all patents involved with ram. So no JEDEC members have to pay royalties. 3. Change rules so that ALL patents and patent applications must be signed over to JEDEC.
...and they're the only ones.
Another Rambus story. Am I the only one who just couldn't give a rats ass anymore?
Casual Games/Downloads
Check out The Inquirer for more information. They said the court makes no difference to the patent infringement case.
--
Error 500: Internal sig error
So your job as a network architect is to actually spend hours running debug version of kernel looking for some small bug ?
How do you know you haven't introduced much bigger bug somewhere down the line ?
Are you kernel developer?
Are you suggesting all companies hire part-time kernel developers for occasions like that?
Correct me if I'm wrong, but if you have a chipset (and for that matter, a CPU) specially designed around the RDRAM, you get something like double the effective memory speed.
Of course, that's assuming you can get the chipset and CPU working right, which isn't easy. I think this is one of the reasons Intel is backing out of their support for RDRAM.
Your Servant, B. Baggins
Infineon and Rambus have been battling in European courts over the same issues, and the EU judge had cleared Rambus of any fraud charges, so it seemed like it was only a matter of time before the US courts fell in line with that decision.
Maskirovka
Call me a cynic, and I'll take it as a complement.
Doesn't work that way, thankfully. Their patents can still be held illegitimate. Only the fraud charge was overturned, and fraud is a much more serious issue (and much harder to prove) than simple filing of an invalid patent. Fraud requires knowledge that the patent was false, as well as intent to obtain money thereby, as well as goodness knows how many other additional legal hurdles, whereas patent invalidity simply requires that the court determines that the patents were granted erroneously, with no question of intent or motive.
As an analogy to the courtroom scenario you describe, suppose I am in posession of some stolen property. They can't prove I stole it. Do I therefore get to keep it? No. Regardless of whether I actually comitted a crime in obtaining it, it still doesn't legitimately belong to me, and must be returned to the owner.
"Never let your sense of morals prevent you from doing what is right" -Salvor Hardin
http://eet.com/story/OEG20010810S0081 Basically, Rambus is in the same boat they were before. Don't know how the assessment of the ruling came out, but if you ain't a hardware guy, don't fsck with it (pardon the pun...).
how many times can they go back and forth.. and does the first decision hold any more weight than subsequent rulings.. appeals.. ?
What! Do I look like a people person?
I'll say "Intel sucks, RDRAM bites, DDR rocks and AMD RULES"
I'm on a tbird right now with ddr and I agree that RDRAM bites, but there is a via chipset that allows you to use ddr with a pentium 3. DDR rocks but it isn't amds only.
How does he know that he did not introduced new bugs that will cost his company hours of downtime three weeks from now?
BTW.
Is it really efficient to pay networking people for debugging kernels (which is a big job on its own)?
Why do you think there is a market for precompiled and packaged OS in the first place?
Erik
"You," Bite me.
"Each and every one of you." Bite me.
Considering their aim is to be an Intellectual Property innovator and owner, they'd better be. Las time I looked they were pumping 50% of their revenues into legal wrangling. This must be the reality of being and IP company, beg, buy or steal then defend or bluster, and hope like heck your oppenents aren't A) Many B) Richer C) Smart
A feeling of having made the same mistake before: Deja Foobar
Infineon, a spinoff of German electical giant Siemens, _is_ much larger than Rambus, particularly resource-wise.
A feeling of having made the same mistake before: Deja Foobar
When I first read the headline, I saw
"Court Decision Flavors Rambus"
mmmm, Rambus
Aside - if it had been a DMCA violation, we would have been executing^H^H^H^H^H^H^H^H^Hprosecuting employees by now.
In my mind, at least, it's going to be very hard to maintain that their patents are bogus if they are not fraudulent. I don't know much about how our appeals system works (never had to use it, thank gods) but I expect that the higher court to whom they appeal cannot effectively re-open the investigation of fraud without Infineon also appealing.
That means the next case is (probably) going to go like this:
These your patents? Yup. This their RAM? Yup. Pay 'em. Case closed.
AFAIK there is no viable way for the court to assert anything else without a _major_ break from precedent.
Remember, though. IANAL. I really hope I'm wrong.
a few weeks/months waiting time? :P..
unless you're compiling on a VERY slow machine..
Microsoft and others (including Redhat) are ALWAYS slow to bring out patches.. why? Well first, they have to find and apply the patches themselves (and in Microsoft's case, they have to find the problem first, whereas Linux-based companies can just apply the patch that someone else has created).. but what takes more time, these companies have to be relatively sure that applying the patch wont accidentally break anything else! Applying the patch may well fix a bug, but if 90% of the 3rd party software for whom this particular bug was a problem all created various workarounds, you may end up breaking all that extra software (again, this tends to hit Microsoft harder, since a Linux vendor would probably have just submitted a patch themselves if they happened to run across the bug)..
I've heard similar reasoning to excuse at least a large number (probably not all) of the reported 65,000 bugs in Win2k when it was released. Sure it gives us geeks with our open source a bit of a chuckle, but it does make at least a little business sense to not screw over your 3rd party developers, especially in a world where a good number of your sales are driven by the fact that most 3rd parties still develop exclusively for your OS.
That's where you're slightly wrong. One thing (there are others, I believe) that set RDRAM slightly apart from normal DRAM is delayed gate technology. Rambus was the first to incorporate this into their DRAM, which is what makes RDRAM move faster (even though it seems like it shouldn't, move data faster by putting a "stop light" to tell data to stop(!!!) and go?! WTF?!). This was unique to RDRAM only; in fact, it was a fundamental patent. When Infineon incorporated this into SDRAM without paying royalties, that's when Rambus got pissed.
Geoff Tate would leap from the bushes, run down the street and fax the design to the US patent office, then sue unless he was paid a stipend.
Lucent would attempt to make the cans and string and lose a large fortune, requiring massive layoffs at their lemonade stand.
Steve Case would be adamant that the string have a huge granny knot in it to be more effective
Bill Gates would say it has to be 4 inch braided nylon rope with a detour through his house
Larry Ellison would hold his breath and turn purple unless the cans were imported from Japan
Thomas Penfield Jackson would call Gates a big stupid jerk and get dragged home by his mother and grounded for a week
A feeling of having made the same mistake before: Deja Foobar
Yes, but fraud, too has different interpretations, depending where you are. If you ever want an education in "rights" then travel to verious countries, or just research, and find out how differently "rights" are observed and protected there. Burden of evidence, innocent until proven guilty, etc. varies widely. In some countries what Rambus did would be viewed, as with Clinton's sexual antics, as par for the course.
A feeling of having made the same mistake before: Deja Foobar
YES.
Not having to pay RamBus 3.5% will make the prices fall that much faster.
Exceeding the recommended torque is not recommended.
They believe what they are doing is protecting their inventions. I mean, to them, SDRAM is just an extention of RDRAM (incorporates such things like Delay Gate technology), which is why they should extend their claims to SDRAM. Infineon then claims no, they invented it first, and shouldn't have to pay for anything. At the end, it will just rely on prior art, and in that case, Rambus seriously believes that it can win (at least, for DDR DRAM). This is what I hear from a bunch of Rambus employees.
There you go ...
You got a patch for known problem and you applied it.
The only difference from commercial products is that it was source patch as opposed to binary one.
I fail to see any advantages here. I mean what is so superior about performing recompile instead of simply downloading patched kernel?
Many of what you describe are were covered in the Magna Carta, which the US Constitution was modeled after. "Life, Liberty, Pursuit of happiness" has been interpreted to no end to cover many of these issues. Rambus attempts to prevent Infineon from selling products in each country, or in the EU by battling within Italian, German or EU guidelines in Europe. How that plays on that side of the pond has no relation in the US, even if both were US companies (which Infineon isn't.)
A feeling of having made the same mistake before: Deja Foobar
All in all, it's still a really bad thing for Rambus, as they've got to be running so low on revenues, from slumping sales in the IT market, that they should be considering out of court settlement with the manufacturers. Although such a concession might be too late.
Yet there's still the matter of Rambus vs Infineon in a german court, where a technology expert was appointed recently, and the italian procedings which looked like a win for Rambus, but the judge put on hold. Wait and see.
A feeling of having made the same mistake before: Deja Foobar
99% of Win32 software created 5 years ago will work on Windows ME. You are the one who doesn't know what he is taking about.
And when Microsoft tells you that your software, which won't run on windows ME, is too obscure to bother with, you will understand the power of Open Source.
Reboot macht Frei.
This could set an unfortunate precedent vis à vis Microsoft's .NET initiative. The opposite outcome may have deterred M$ from claiming patent rights after the standards-setting process.
I'll say "Intel sucks, RDRAM bites, DDR rocks and AMD RULES!" Go READ the datasheets for the P4 and the Athlon and come back and tell me which is better. But if you can't understand those datasheets, shut up, you're not entitled to an opinion. Hersay does NOT count!
I have, on several occasions, detailed how AMD has licensed, cloned, or otherwise second-sourced every chip Intel's ever built, usually with better yields - which means better speed, better reliability, and LOWER PRICES for what are essentially the same chips.
That's not just the CPUs, either. That's UVEPROM, EEPROM, peripheral chips, and Intel's entire line.
But Intel has extravagant multimedia ads which lie out their ass, as well as, of course, dancers in bunny suits, so they get brand recognition (sorta like Micro$oft) and AMD must suck, right?
WRONG! AMD has people in bunny suits, too, only they're busy burning silicon instead of dancing on your TV. This savings allows them to bring you their product at lower prices.
And RamBus? RamBus should just cash it in, retire, and get out of the market before someone else sues the remaining snot out of them.
Exceeding the recommended torque is not recommended.
512 MB 800 mhz Rambus RDRAM - $380
256 MB PC133 Micron SDRAM - $25
512 MB PC133 Micron SDRAM - $77
Remind me why I need this stuff? If you want to see the future of RAMBUS, reference "microchannel" expansion slots. It seems you can only strongarm an entire marketplace if you make software....
Seen any BadMarketing lately?
Here's a link to the actual docket http://www.rambusite.com/RambusVsInfineon/Docket.h tm .
It probably would be nice to have someone comment on what was exactly denied in the ruling.
Right; but that's a -patent- issue, not a -fraud- issue.
They're still in business? I was under the impression that the Infineon (sp) judgement was crippling...hm.
El riesgo vive siempre!
His job is to keep the network running.
His tools consist of whatever software is provided by his employer, whatever tools he brings to the job, and his wits and skills.
If chasing down a kernel bug is what it takes to keep the network running, then that's what it takes. I would hope and expect that he didn't do this in a vacuum, and consulted with kernel developers somewhere in the process.
Being able to say, "I tweaked the kernel and fixed the problem." isn't the point. Nor is saying, "It's Microsoft's problem, and a fix is due tomorrow/next week/next month."
The point is that the network is up, and because the Source was available, he could apply all of his wits and skills to the task. Without the Source, he's working with fewer resources.
The living have better things to do than to continue hating the dead.
Edith Keeler Must Die
20 January 2017: the End of an Error.
If you'll look at the financial statements of companies like Toshiba that make these chips, I think you'll find that their profit margins are none too fat. 3.5% is a big bite, and one that the manufacturers can ill afford.
If I were Micron, I'd tell RamBus to go fsck themselves. I doubt that they'll get any more manufacturers to cave in to their claims. At best, they'll win hard-fought court battles for any royalties at all, battles which will leave them financailly drained. MUCH more likely is that they'll fight another fraud suit and _lose_, at which point the RAM industry will laugh at them even harder than they are now.
Exceeding the recommended torque is not recommended.
Every time you purchase memory, you put a little cash back in Rambus' pocket. Or am I completly off my rocker?
Memory of all types continues to get faster and cheaper at an astonishing rate. Is this decision really going to have any effect on that?
But even if they're not criminally liable, the amount of bad publicity they've recieved in the last little while can't be anything but bad for busines. What with the fraud cases themselves as well as Intel dropping support, added to the cost difference between Rambus and SDRAM, they've distanced themselves from the majority of the geek community, something that other notable companies can afford to do, but they can't. When SDRAM is less than CDN$35 for 128mb, It's just impossible to sell on truse alone when you have no more trust.
Now it's just sit back and wait for more bad news for Rambus.
- Relativistic? That's barely Newtonian!
I used to trust the media to tell me truth
But now I see the payoffs everywhere I look
Who do you trust when everyone's a crook?
Revolution calling!
Revolution calling!
Revolution calling you!
Personally, I'm glad Rambus is dying the slow, steady death; grasping onto it's customers as life preservers is a bad side effect, but it won't keep the afloat forever. Why does hasn't Intel made more progress in introducing a DDR chipset? Do they really think this company should be allowed to live? I realize they've got too much Rambus product, but is there some sort of contract preventing Intel from releasing DDR chipsets? Rambust needs a proper burial, and I think Intel needs to be the one nailing the coffin shut.
SELECT * FROM USERS WHERE A_WINNER = "YUO";
Ya, getting the source is such a pain in the ass. Do any of you know of any warez sites where I can download the source?
http://saveie6.com/
Um...between the above and the subject line you're using, I hope you didn't get the impression that I am some sort of Intel fanboy.
Oh, no way! My comment was aimed towards the described induhvidual, and mostly at the general "Intel Inside" mentality. While I can't track him down and flame him directly, this is as good a forum as any to rant in. I hope _you_ don't get the impression that I'm less than proud to be an AMD customer.
I guess I _did_ kinda go off, didn't I. It's not as if anybody blasted you or the others who were calling the RDRAM fans zealots. I should probably go over to the fool and foam at the mouth there for a while...
Exceeding the recommended torque is not recommended.
I'd be happy to see 208 hours uptime on my windows2k advanced server, which doesn't even want to boot if you have terminal services set to start automatically.
However the many linux servers I have only get rebooted when I do a hardware upgrade or replace the UPS.
Umm... it's April Fools' Day, right?
Right?
(panic attack sets in)
First, I don't know if the patch that I got from the person responsible for that particular portion of the kernel introduced new bugs. Second, neither do users of other OSs know if patches/service packs add new bugs. Finally, I didn't go through the code itself to find the bug. It just so happened that the kernel panic I got pointed me to line 155 og highmem.c (BTW it reads "BUG();". Obviously, someone knew that his particular thing 'could' happen. It requires a very specific combination software and hardware to exercise the bug (SMP and >900MB).
Wherever you go, there I am...
Rambus is home in the courtroom.
Just up on Yahoo news: Shareholder suit filed in US District Court of Northern California on charges of fraud in representing their patents on SDRAM.
There's some serious blood in the water here. Could this be the beginning of the end?
A feeling of having made the same mistake before: Deja Foobar
Now for once and for all, since it's only a sig: (and for some strange reason people just don't get it)
I always thought it was more of an "I'm a geek and like girls in general." I think we can all follow the logic from there and most of us can relate to it somewhere back there.
Now back to your regularly scheduled RAMBUS-related-discussion. Sorry to have wasted your time.
- Relativistic? That's barely Newtonian!
"How does he know that he did not introduced new bugs that will cost his company hours of downtime
three weeks from now?"
He doesn't but then again he wouldn't know that if he was installing a binary only patch from MS either. You need to do testing with any patch from any vendor and you allways need to know what you're going to do if there are unforseen problems. At least this way he knows exactly what was done and can put things back if necessary very quickly.
Boy, what a difference spin makes. Check this article out:
http://biz.yahoo.com/st/010810/28625.html
Seems pretty much pro-Rambus.
Now look at this one:
http://biz.yahoo.com/bw/010810/100178.html
Basically a reprint of Infineon's press release. Boy, and you thought the patent system was confusing! So what's the real analysis?
I have a system or two running SMP and >900MB physical RAM, and I've never seen that... OH WAIT I'm running Windows 2000. With an uptime of 208 days. Sorry that uptime isn't longer, but the server was built 208 days ago.
"Rambus revealed during the Infineon trial that it charges royalties of 3.5 percent on DDR SDRAM and 0.75 percent on SDRAM. "
Seems a little extreme, eh? What a waste of time patents can be...
We dance to all the wrong songs.
--Refused.
But the court granted Infineon's request to prohibit Rambus from pursuing further litigation regarding Infineon's SDRAM memory and ordered Rambus to pay $7.1 million in Infineon's legal fees.
OUT
It's nice to see a judge standing up against large companies, to stop them from throwing lawsuits around against smaller companies who cant afford great legal defenses. Maybe this will (*cough ADOBE cough*) stop (*cough ADOBE cough*) from suing smaller groups over trivial (or perhaps nonsense) items.
Mooniacs for iOS and Android
"That's the kind of guy I am: the kind of guy who needs a girl who hates guys like me."
-Too Much Joy "Magic"
Seriously, no mention is made as to the judges' reasoning behind this. Was there a bit of data that the jury wasn't privy to, or is Judge Robert E. Payne just a dick? And what's with Geoff Tate's comment? "...the record has been set straight on DDR SDRAM." Ummm, what? For all we know those counts of fraud were overturned for shits and giggles. Anyone have other sources they could share? Perhaps something that actually tells me something? Need input.
Democracy is two wolves and a lamb voting on what to have for lunch. Liberty is a well-armed lamb contesting the vote.
Yes, but we're not talking about the finer points of patent law in this case; we're talking about fraud, which is a pretty much the same thing in both courts. Also, this case has international scope, because Infineon is a European company. These two court cases aren't happening in separate vacuum-sealed spaces, either.
So it is quite likely that, upon reading the EU judge's decision, that Robert E. Payne applied similar logic to the case over here.
My point is that the laws are the same. Fraud, murder, theft -- it doesn't matter if you're in Liverpool, Rome, Tijuana, Montreal, or Dallas; they have the same definition. Reparations are different, and the processes for making judgment are different, but we don't disagree on what it is the way we do on, say, patents.
If Payne did so then Infineon has a clear path to appeal, as US courts should only apply by US law. Now, US law could be part of a treaty with the EU, but I haven't read such as mentioned anywhere.
A feeling of having made the same mistake before: Deja Foobar
Well? What's on that line? Spill the beans dammit :)
UPDATE: Via is also making a chipset for the p4 that allows you to use ddr. Here's linkage.