Slashdot is powered by your submissions, so send in your scoop


Forgot your password?

Intel Submits Patent Covering Itanium Instructions 167

chris.bitmead writes: "Rather than submit garden-variety claims to the U.S. Patent and Trademark Office (USPTO), Intel is trying to patent the functions carried out by specific instructions. In doing so, the company appears to be, in effect, trying to patent the IA-64 instruction set itself." Is this an attempt to block out even reasonable competition, or is this just "business as usual" as at least one voice in the story says?
This discussion has been archived. No new comments can be posted.

Intel Submits Patent Covering Itanium Instructions

Comments Filter:
  • Really? If so what exactly are they charging it on, and how do they count it?
  • > IA64 blows. Nobody uses it nor will they ever. Time to grow up, intel.

    Nobody uses it? Well since it hasn't really been released yet that is kinda of understandable.

    Also have you actually looked at the instruction set? They have some really novel stuff. If it is new then go for it, patent it.

    However like a previous poster said, the patent system would be great if it was 3-5 years patents, not 20 year patents.

  • by RedWizzard ( 192002 ) on Sunday October 29, 2000 @11:12PM (#665639)
    Unfortunately modular design is very difficult when it comes to supporting multiple CPU families. The problem is, of course, that you need to swap the north bridge (what Intel now call the MCH) when you swap the CPU type because the bus protocols will be different. There has been some speculation that you might be able to drop an Athlon into a Alpha motherboard (and vice versa), as they use the same CPU bus protocol, but I've never heard it being done.

    One interesting design was that of the PIOS One, which put the north bridge and the CPU(s) on a daughterboard that plugged into a PCI bus. Unfortunately it never got passed the prototype stage but it would have made for an easy way of changing CPU families and even changing the number of processors (up to four).

  • This doesn't look like patenting the instruction set at all. That would be utterly anticompetitive, as it wouldn't allow competition by manufacturers wanting to make their own chips with compatible instructions sets. It wouldn't even allow simulators to be created by software tools companies, and arguably even compiler writers would be caught by it. I don't think they'd be so blinkered to attempt that.

    No, this looks like merely a patent on particular aspects of the implementation of certain instructions. As long as the implementation details aren't generic catch-all ones that every competitor would have to implement identically, this seems relatively innocuous as patents go.
  • Wow, the ever insightful comments of an AC!! I love the attention you're giving me, so keep it up! If you didn't find it interesting, why would you keep bothering to reply? Is it *your* personal quest with me?? Awwwww yeah!!!
  • by Anne Marie ( 239347 ) on Monday October 30, 2000 @06:43AM (#665642)
    Then there was the owners' secretary, who - to quote the Partner on the job - "liked the cut of my jib" - and hinted about her upcoming weekend in Palm Springs.

    She's just doing what it takes to survive, since she's just been abandoned in her existing relationship with her bosses; her entire world has just come crashing down, and she's desperately flailing about to try to attach herself to another providing male. And you, as the white knight who's ridden in on your white horse to save her and what's left of the company from this fate, are the perfect opportunity.

    If I may hazard a guess, she's probably in her late thirties, not so old that she's given in to societal pressures and renounced her status as "spinster" by marrying the first dweeb with a paycheck who comes along, but not so young that she's holding out indefinitely, as if somehow her hand won't be forced by the same society that grinds us all up and spits us out and won't take no for an answer.
  • While they're at it, maybe they can patent faulty division. I remember that innovation of theirs back from the Pentium.

    Then if anyone does that to us again, Intel can sue them for us...

  • I'm the only one, eh? This guy [] claims he beat me to it. I must admit, your trolling was pretty good with that one, but not good enough!!! There have been a lot of ACs posting about Siggy's new account, so how about trying another angle on your weak trolls.... haha!!!
  • Does anyone else remember when IBM tried basically this same thing with MicroChannel Architecture, and other companies came out with EISA and made it an open standard. MCA died the death. Will the same thing happen here?
  • never claimed I wanted karma, fool. And I'm not a cheap whore, I'm an educated virgin!
  • Just because Intel is patenting IA64 doesn't mean that it's the end of the world. 64-bit processing is still in its infancy in the consumer world (either vaporware or prototype, but still not released). If IA64 really takes the cake and beats out AMD's tweaking of x86, then maybe Intel will start dropping prices and using their heads when designing chipsets.

    Of course, there's been transitions like this before (Windows 3.1, VESA, Win32, DirectX, Linux, SDRAM, OSS, etc.). This transition is like the previous ones; it requires patience and judgment (though a fat wallet certainly helps).

    Personally, I like how Intel is patenting IA64; that way, there won't be any sleazy IA64 imitations out there, as was the case with the K6, IDT WinChip, and Cyrix's entire line of CPUs (quote from the Quake manual: "Computers without an FPU will not run Quake... at all. This includes the now infamous "Pentium-class performance, for less money" knock-off chip [Cyrix 5x86]). Yes, this might be an assembly/C/C++ programmers nightmare, but isn't that why programmers are always learning new things? No? Well they should be; maybe that's why amateur programming stinks so much.

  • The federal government can only be sued if it gives you permission to sue it. I think that's in the constitution. So I guess that you could ask the USPO for permission to sue them? Or do you need to ask someone else? (I don't believe that that is specified.)

    Caution: Now approaching the (technological) singularity.
  • Actually, Microsoft seems to like having multiple architectures running the Windows NT architecture. They've had it on:
    • Intel iAPX32
    • Fairchild/Intergraph Clipper
    • Motorola/IBM/Apple PowerPC
    • MIPS Rx000
    • Digital/Compaq Alpha
    and have announced support for Intel iAPX64 when it ships. So,it looks like the problem is that nobody buys other architectures since none of these sold well enough to keep in development.
  • Will you give up your $90k job for $35k to work for USPTO?

    The solution to this is to axe the $200-300k (or whatever it is) that US Senators bring in per year, after they retire and put that into things like raising salaries on USPTO employees and maybe something like, I dunno, teachers.

    - Scott
    Scott Stevenson
  • Y'know, as a software developer rather than a hardware developer, I'm much more concerned with the instruction set rather than how easy it is to implement caching hardware. RISC is nice for the hardware people since it makes it easier to compensate for lousy RAM access speed but it makes software harder to do.

    Now, don't interpret this as saying that the PentiumX is a good CISC instrution set, it isn't, but why do we not have ANY choices besides having to reinvent 80% of the instruction set in software every time?

  • True, but lets be honest. Most Linux people (including /.ers) run Linux on standard Intel, "Designed for Windows" hardware. Any advantage of portability to other architectures is not seen in practice since nobody bothers with any other options.
  • by harryseldon ( 29164 ) on Sunday October 29, 2000 @11:12PM (#665653)
    Intel is hardly the first to patent aspects of an instruction set. Nobody ever cloned the VAX, because essential aspects of the instruction set were patented. The IBM 370 arch was widely cloned because it was not protected. There are intel patents on the x86 arch and microarch, but (apparently) not sufficent to prevent cloning. MIPS has patents around a few instructions but one embedded CPU company (Lexra?) got around it by not implementing those load instructions. Aspects of the Alpha arch are patented IIRC. No alpha clones, sorry. In short - looking for 'good guys'? keep looking. No modern processor instruction sets are unprotected.

    not speaking for my employer. whoever that may be.

  • Other companies (hopefully) should forced not to copy intel, and to come up with some new stuff.

    Hmm, patents used to protect IP and force other competitors to come up with own ideas!
    Could it be this is actually a Slashdot story cheering the advantages of patent law?

    Oh my god!!

  • This post is purely based on Intel POV and is in no way reflects the views of this author.

    For Intel, with AMD breathing down their necks, filing more patents is the smart thing to do. Anything you can do to slow down the other guy will enhance shareholders values, that is what this is all about. It is after all their R&D money and their researchers. As we saw time and again the USPTO is dumb enough to allow this sort of thing, so why not?

    Remember, all corporations are out to make money, anything else are just unforeseen side effects.


  • One has to wonder what would have happened if Intel had patented the extended instruction set of the 80386... or the first Pentium chips. No clones? Would AMD or Transmeta ever existed as CPU manufacturers? This sort of thing scares me, as it indicates to me that we may miss out on something great (think of AMD today) in the future.
  • by pjrc ( 134994 ) <> on Sunday October 29, 2000 @11:21PM (#665657) Homepage Journal

    A computational system in which an operational code (opcode) consisting of a sequence of numerical data instructs an Arithmetic Logic Unit (ALU) to perform an operation on two operands, storing the resulting arithematic or logical result into either of the operands.

    What is claimed is:

    1. A computational system comprising an ALU (Arithematic Logic Unit), c a operational code decoder, memory bus interface, and microcoded control logic, wherein,

    Arithematic Logic Unit further comprises circuitry to perform mathematic operations of addition, subtraction, increment, decrement, multiply, divide, and logic functions of AND, OR, XOR, right and left bit shift;

    2. Operational code decoder comprises circuitry that extracts encoded information to direct the activity of the Arithematic Logic Unit (claim 1);

    3. Memory bus interface that transfers the operands required by the ALU (claim 1) and instruction opcodes needed by the decoder (claim 2);

    4. Microcoded Control Logic which sequences the timing and produces control signals with the proper timing to direct the activities of the ALU (claim 1), Opcode Decoder (claim 2), and Memory Bus Interface (claim 3).

    Description Of The Invention

    The present invention related to the operation of a computational device, used to execute computational tasks, wherein the computational task to be performed by be programmed by creating a list of operational codes.




    maybe someone else wants to continue this... it's late and I'm getting tired...

  • "Can't use your instruction set?!? OK, we'll just download another..."
  • by trims ( 10010 ) on Sunday October 29, 2000 @11:36PM (#665659) Homepage

    After reading this article, and looking at one of the patent mentioned [], I'm going to hazzard a reasonable guess at exactly what's going on here:

    1. They aren't patenting IA-64 codes. That is, the aren't trying to patent JMP for jump, FOO1A for super-duper multimedia instruction, etc. I don't think they're fool enough to believe that anything like that could ever get patented, or for that matter, copyrighted or trademarked.
    2. It is unclear if they are trying to protect functionality or implimentation. From the patent application itself, it appears as though they are only trying to protect their implimentation of how to do interrupt context returns, not the concept of interrupt context returns itself. The article, however, sounds like they're trying for the latter.

    The problem with this kind of patenting is that the "concept" is closely tied to the "implimentation". That is, the concept may be so narrowly circumscribed that any implimentation is an 'infringing' one.

    Also unclear in this whole mess is how a software trap would fit in. Suppose Intel was granted the "broader" patent which covered not just the specific transistor layout of the interrupt return handler, but the more general case of returning interrupt context for IA64. Does this preclude software implimentations of that IA64 instruction (which would be particularly relevant to code-morphers like Transmeta, but even to AMD et al which do translation to microcode)?

    I'm by far no Patent Lawyer. If the scope is the narrower one, I see no problem, and indeed, is well within the goal of patents. I'm alot let sanguine if the patent covers more than the circuit design, however.


  • I don't have any prior art, the author does. And I don't have much money, but Andy Tanenbaum might...
  • Ok, I agree that any probems with the patent office stem from the politics behind the patent office. Goes for any government agency. To tell you the truth, shame and ostracism is probably as far as I'd be prepared to go in the fight against the information fascists. I just hope harder heads than mine are taking action on other fronts. I agree that all "intellectual property" is a commons, but I hate calling it IP. To me IP implies that it's natural state is to have one owner, and "stealing" it is a crime. I think that "stealing" it is ok and monopilising is the real crime. I think it differs from, say, a pasture, in that everyone can glut themselves on knowledge without taking any of it away from the rest of us. All you take away is monopoly power, that is, money.
  • I don't know the specifics, but the mobile phone co.s thought it'd be cheaper.. which I don't think it was... look it up!

    all misspellings were intentional.

  • When I'm ready to move up to the 64 bit processors, I'm already banking on AMD's 64 bit solution

    So, when do you suppose ordinary computer users, or for that matter, geeky computer users like myself, are going to be moving up to 64 bits? The jump from 16 bits to 32 bits is obvious -- lots of useful programs use more than the 64KB that 16 bits directly addresses, and the work-arounds to get to more memory on a 16 bit CPU were a pain. 32 bits gave us 4GB to address directly. I don't see myself needing more than 4GB of directly addressable memory space soon. 64 bits gives you something like a bazillion bytes.

    I know that high-end servers sometimes utilize a massive amount of memory -- and clearly that is the market for 64 bit CPUs now. Will the technology filter down to the user-level anytime soon? If so, why?


  • What? Why would you go on about microprogrammability in connection to IA-64? This is an important issue for high-end RISC processors since when, exactly?

    In case you missed every conceivable opportunity to find any information on IA-64, I'd fill in the novelty as follows. Among general-purpose CPUs that hit the mainstream (please remember this bit before flaming me about Multiflow or your favorite DSP or graphics chip), the first implementations of IA-64 will be:

    First VLIW-type chip, with at least some potential of avoiding the standard pitfalls of VLIW.

    First CPU with extensive support for control and data speculation.

    First CPU with register rotation for overhead-free software pipelining of loops.

    These may or may not be good ideas. I suspect that some of the stuff that is in the IA-64 architecture will turn out to be a bad idea. Not so much because it's inherently stupid; more because some of the features are going to be hard for compiler writers to exploit, and may become underutilized.

    However, no-one who knows anything about computer architecture would deny that the IA-64 is novel.
  • I know there were some Sparc clones, notable is Ross Semiconductor. I think Fujitsu did as well.

    Alphas had up to two cloners. One _was_ Samsung, but they are OEM to Compaq as well now. Mitsubishi was in the 21164PC alliance but it seems they backed out before fabbing them. Both cloners were licenced.

    The way Intel supposedly had cloners in the beginning is that to be allowed to be used in military and certain government projects, your products had to be made by more than one company, so Intel licenced out their mask designs. They backed out of the government and therefore the mask licencing market at around the 386 days, so many companies had "486" chips that were poor performers at best and ended up being souped up 386 chips with bigger cache. To keep up since then everyone had to make their own P5 and up designs.
  • OK. Let's look at the scenario without patent protections.

    Companies A & B are in the same industry. Both have $100,000,000 budgeted for a new product.

    Company A spends $99,900,000 developing a product (paying programmers and engineers) and spends the remaining $100,000 on marketing.

    Company B spends $ 100,000 making a copy of Company A's product and spends their remaining $99,900,000 on marketing their copy.

    Guess which one succeeds in the market?

    Would you rather see those salaries go to marketing people or programmers/engineers?

  • Nobody ever cloned the VAX

  • This patenting of specific functions smacks of Microsoft tactics

    Actually it smacks of Rambus tactics, which Intel has recently been decrying. Pot meet kettle?

    Whatever you may say about Microsoft, abusive patenting doesn't seem to have been among their sins.
  • "Our customer base sees this type of litigation as business-as-usual," said Charley Cheng, chief executive officer and president of Lexra.

    Oh, Brave New World, that has such people in it.

    Simple evolution should not be patentable, period. Evolution of a concept in obvious ways is not novel.

  • Windsor, Ontario. 10.30.2008
    A recent discovery by a research team from the University of Windsor in Ontario Canada has unlocked one of the secrets to the universe. A recent finding which is the first to quantify what had been previously known as "Dark Matter" has caused a stir amongst Venture Capitalists and Academics alike.

    Academics are relieved that a previously misunderstood phenomenon has been resolved. Dr. S.Banndyd head of the Cosmology department at the University Of Windsor is proud of the team lead by Researcher Dr. B. Smith, "This marks a great achievement by the team, and they should be commended. The fundamental truth, now finally understood is marks a boon to human knowledge. The mystery of Dark Matter and gravity is finally revealed. We are ecstatic. But we really owe our discovery to 500 years of public knowledge and contributory research." Dr. Smith is quoted as saying "We knew this to exist, the interaction between DarkMatter and the rest of space, but had simply not been documented before. But we have finally put the facts on paper in a quantitatif manner." he later added "We are very proud - and will be going out for some WalkerVille Ale this eve."

    The Head of the Corporatist Whoring Department Mr. F. Mehard is estatic. Citing the case law established in the United States in the earliest years of this century, where corporations were able to patent basic action and fundamental ideals. He feels this will be a boon in license revenue for the institution.

    "We have spoken with an array of Venture Capitalists and lawyers, and are sure that a new startup will be founded to commercialize this information." said Mr F. Mehard. "The new corporation will likely be called Gravity Inc., we feel that our patent-filling-slave^H^H^H^H associate will process the request online this morning - we should have our patent by mid-day. This 'Dark-Matter'(TM) ® © which we have described plays such a fundamental part of "RealyGrabbyStuff(TM) ® ©." ReallyGrabbyStuff(TM) ® © is the new trade name chosen for Gravity, and Mr. F. Mehard feels his product will be a hit saying "We fully expect that our Team Of Rabid Lawyers(TM) ® © will have no problem extort^H^H^H^H^H^H closing license deals with the residents of the planet. Without our "ReallyGrabbyStuff"(TM) ® © every person on the planet would be lost to drift in space - and as such, they owe the privilege being Flaeterning(TM) ® © (the expression used when describing the application of "ReallyGrabbyStuff(TM) ® ©") to the new Gravity Inc. Corporation. We were the first to document this behavior completely, and as such are entitled to owning all application of said technology".

    Mr. F. Mehard and the Team Of Rabbid Lawyers(TM) ® © then left he University aboard their Anti-ReallyGrabbyStuff(TM) ® © vehicle, bystanders were noticed as saying "boy are we lucky - maybe now we wont have to live in the sewers and eat rats." the departing dignitaries were then seen tossing small pieces of food(TM) ® © and other(TM) ® © stuff down at the hungry academics and students below.

    Dont like this vision of the future? Then take control of your country from the bastards! Tell your friends/neighbours/relatives/coworkers to:
  • They figure, they better do it before someone else does. Of course with all the convenient secret patent 'licenses', intel will naturally be using this exclusively to sue AMD or anyone else trying to compete with them. Like I said, business as usual.
  • For the record, IBM also had a patent on the ISA bus, as well such base standards like VGA, only that they were available for much more favorable licencing terms than MCA.
  • an attempt to block out even reasonable competition AND is just "business as usual"
  • They have some really novel stuff.

    Compared to what? Other 32-bit architectures? Other 64-bit microcomputer achitectures? Or mainframe architectures, which happen to be 100% microprogrammable? It doesn't seem that novel to me.

  • by Mr Z ( 6791 ) on Sunday October 29, 2000 @11:43PM (#665675) Homepage Journal
    Not that I'm saying that it isn't possible that Intel is doing this, but the fact that they suddenly submitted a bunch of patents hardly constitutes evidence.

    Also, as I understand, it's pretty typical for patents on a CPU to be filed all in a burst around the same time the CPU's info's being rolled out to market. Part of the reason for this is that the patent disclosures themselves sit around in the pipeline, gradually making their way to the USPTO. Then the marketing-side of the company decides to do a Release to Market of some more details, so there's a sudden rush to flush the pipeline so that the company doesn't forego any patent protection on those patented ideas that may be presented in the RTM.

    At least, that's how it looks like it works here for the patents I was involved with on TI's TMS320C6400 CPU. I won't comment further on the content of those patent applications, or the purpose behind them other than to say I think all the semiconductor companies play the same game here.

    So, don't just single out Intel, 'kay? And put your conspiracy theories away. This is just business as usual, and its purpose is to give the originating company an advantage and a defensible barrier against direct competition by cloning. It just so happens that cloning is more important in Intel's world than many other worlds, so people get hypersensitive about it.

    Program Intellivision! []
  • Andy Grove quote!

    Must be afraid of being like ZEROX.

  • I think you're pretty much on the money.

    There is real scope for a massive industry split. Remember when IBM tried to reclaim the PC market with the PS/2 and "micro-channel architecture", a non-ISA bus which they had patented? (1987). The world of the PC just walked away from them and did its own thing, EISA and so on.

    This could be the luckiest break that AMD will ever get.

  • by Anonymous Coward
    Funny thing is that ther IS a 64bit Windows 2000 coming out soon, and acording to the Intel site they've been working in cooperation with Microsoft to create it more or less specifically for IA-64. Try doing your research rather than just being the typical Pro-Linux anti-MS type.
  • ... is now secure. If Intel uses IP to try to prevent cloning of IA-64, the cloners will have that much more incentive to squeeze more and more out of x86.

    Economically, they must - the patents won't be licensed, so there is no other alternative but bankruptcy. I predicut x86's lifetime will be lengthened significantly by this development.

    Anyone want to start the FISF (Free Instruction Set Foundation)?

  • Heh. There'd better be more to it than that, or my Foundations of CompSci textbook is prior art.
  • What is Intel really trying to patent here? Not ideas, that's for sure. They're trying to patent low-level implementation details like how certain instructions impact data flow throughout the IC.

    That is EXACTLY what you are supposed to patent. You are not suposed to patent "ideas". You are supposed to patent specific implementations. If you are correct (I haven't read the patent, not being fluent in legalese), then there is absolutely nothing wrong with this patent.
  • It's no different than Transmeta's patenting, but I know, I know... Intel bad - Transmeta good.
  • AMD, TransMeta, VIA (Cyrix) are teaching Intel not to rest on their laurels. Intel's lesson taken to heart is: Patent everything and if they want to imitate it then license it. But AMD has been pretty good at reverse engineering to get around this. It would be interesting to see if the patents are worded better to give reverse engineering less wiggle room. Keeping an eye to the past as well as to the future.

    In other Intel news, The Reg has this hoomerous piece [] on Intel views on the P4 release schedule.

    "Hello, Intel? Please connect me with Bud and Lou."


  • Not the same motherboard, but rather differernt motherboards without a lot of modification between them. Thus, to make it easy to make different boards to support different chips... not to make 1 board to support them all.
  • Now there's a voice of insight. Sad but too true.
  • Ahem,... it's [not] usual for hardware companies to patent their instruction-sets. Intel, AMD, MIPs, et al - they all do it as a matter of course in their business operations.

    Yes and no. The distinction has to be made between patenting the instructions themselves and patenting the implementation of the instructions. Not everyone is so terribly protective of their instruction sets. For example, for USD 99, you can obtain an unlimited license to produce SPARC processors, and to use the name SPARC in association with them. At the same time, it would probably cost you 99 BILLION dollars to get the chip masks for the UltraSPARC-IIIs. The instruction set is "out there," but the implementation is not.

    The question I have to ask here, though, is - Does anyone really care about ia64? It's not a good architecture, and it's not a good instruction set. Like everything else Intel does, it'll be a technical dud that earns commercial success through heavy marketing and lots of exclusive contracts. That is, assuming they ever actually mass-produce the things, which seems in doubt as they're already 2 years late and counting.

    AMD's better Hammer architecture

    This isn't a comparison I'd like to have to make. On the one side, you have Intel finally trying to cast off the shackles of 30 years' worth of backwards-compatibility - a good thing - and still managing to botch things horribly. On the other side, you have AMD trying to extend an ISA that has been extended far, far too many times already, an ISA that wasn't very good to begin with. If you ask me, both comapnies have the wrong idea. I for one will be sticking with MIPS and SPARC until they finally die. Here's to hoping that will never happen.

  • No, the real solution isn't for technically qualified people to give up their current jobs ($90K or not) to work for the patent office -- it's for the patent office to allow qualified people to review the pattent applications. They can do this by hiring these people (but as you point out, that's expensive), *OR* we can find another way, by allowing qualified people to review and comment on applications.

    Or, instead, we can allow the situation to continue to deteriorate until Congress steps in and changes the patent rules. That might be good... or it might be very, very bad. We just don't know.

    -- Michael Chermside

  • why ? simple. ripping DVDs and other media. notice you cant create files over 2GB on 32 bit platforms. large files like DVD video cant be ripped easily without a 64 bit chip. and thats where it all will come down.
  • most of the stuff that intel has included in their IA-64 instruction set has been done before. VLIW was done during the 80's, there are ARM processors that have predicated instructions and just about every current processor has deep pipelines. the unique thing about intel is that they decided to put it all together. the bizarre thing is that they seem to have lost the plot. hp is recommending that it's customers don't go with the first generation of the Itanium chip (too slow compared with PA-RISC). this patent "should" be blown out of the water...
  • Remember, all corporations are out to make money, anything else are just unforeseen side effects.

    Corporations would do well to remember that customers are just out to get good value for their money, and whether the corporation makes a profit or not is just an unforseen side effect. Intel has been churning out vapor in the processor field for a few months now. When they do turn out an actual product it's a disaster. They're falling farther and farther behind AMD. It would be unfortunate for them if there was a fork in the processor architecture and they found themselves standing on the wrong side of it. Because at the moment, the company with the higher-quality product stands the better chance of running away with the market. In my ever-so-humble opinion.

  • Ah, but INTERCAL has prior art, with its bitwise, unary AND .

    Program Intellivision! []
  • Someone post the instruction set on Napster!

    All generalizations are false.

  • Chaos rules man! (;(left handed emoticon)

    This will eventually cause harm to people and industry. When something finally hits people in the wallets then the government will take notice. Maybe this is that one last staw that will break the camels back. Maybe this will push congress into doing something about technology patents.

    Or maybe even better Intel will get the patents. It will refuse to liscense them to anyone. Every other major chip vendor in the world will get together and create a pc based on a sane architecture. Linux will get ported over very quickly to the new improved architecture along with BSD, BeOs, and other alternatives. Windows being tied to the 80x86 architecture will take much longer to port. People seeing the vast improvements good hardware that is open standard can bring to desktop software will buy the new hardware and of course install linux, bsd,(insert operating system of choice) since windows is delayed for the next decade or so.

    OK happy halloween. I was just kidding none of that will ever happen. Intel will probably get its patent and any other patent it wants, it will try to keep from liscensing it but will eventually give in because of the threat of monopoly charges.

    ----I promise never to write a responce drunk again I swear!! :)
  • Funny, I was just reading the Gnu Manifesto for the first time in many years and I came across a wonderful description of free market competition as a race where the winner is rewarded for the best product, fastest time etc. and this is a benefit to society. The only problem being that if allowed, the participants may forget about the whole point of the race and discover other winning strategies, such as punching other players which serve their goals (winning) but not society's (faster runners/processors etc)

    I think this sort of development quite similar and I'm not sure I see where society benefits from Intel patenting simple instructions.

    I know all the standard arguments i.e.: by locking down everything that can be considered IP they can make more money and thus make better products for society. But the experience of the last 20 years or so would seem to show that one company cannot provide all that our society/ecomony/community needs, even though they are quite capable of *controlling* all those. Our needs require a competition *and* co-operation i.e. standards & interoperability.

    I would like to end on a positive note and urge all the americans reading here to use whatever means you can to force a re-evaluation of intellectual property's place in your very important economy. I know a lot of people think their vote doesn't matter, or that a vote for third party candidates is somehow wasted - I believe very strongly that this is not the case. The vote is your only guaranteed input into the running of your state - use it! And if you don't like the choices, start working now for better ones in election years to come!

    ... A Canadian who is happy to have a number of interesting electoral possibilities worthy of my support who actually get mentioned once or twice in the mainstream press. :)

  • Is it just me or do companies now days forget what got them where they are? Everywhere I look I see companies patent everything under the sun, and most of these companies were successful because because they were able to copy and improve in some way what someone else was doing. It just seems to me like they are biting themselves in the arse if an effort to play alone. Apparently they slept throught economics and the benefits of competition.
  • by Frac ( 27516 ) on Monday October 30, 2000 @12:12AM (#665697)
    now when I write IA-64 assembly code, I can just look it up in the patent database. Soon the patent database will be the authorative resource for assembly programming!
  • Unfortunately, patent reviewers who don't grant patents aren't patent reviewers for long. Even if I was to become a patent reviewer, I would be out of there in a week.
  • Ooops ... There's a big bunch of treaties to make patents valid worldwide

    Some days it really pisses me off that there's no "misinformative" option in moderating.

    If you are aware of such a treaty, feel free and provide such a citation. If you heard this from your mother's friend's sister's secretary, do a little fact-checking first.

    There is a World Intellectual Property Organization [] which does accept patent applications. However, it does not grant worldwide patents. It merely expedites applying for patents in several countries (or regional organizations, see below), but it's still up to each individual country (or regional organization) whether or not to grant the patent.

    There are three regional organizations I know of which grant patents which are applicable across multiple countries: the European Patent Office [], the African Regional Industrial Property Organization [] (covering much of English-speaking Africa), and the Organisation Africaine de la Propriete Intellectuelle [] (covering much of French-speaking Africa). However, each of these organizations have their own offices for evaluating patents. So you can't just go to the Estonian Patent Office, have your application examined by them, and get a European patent. A European patent application is examined by EPO itself.

  • The patent system is enshrined in the constitution, theoretically the highest law of the land. Anti-trust laws are not (unless you believe that that's "regulating interstate commerce"). So patents trump anti-trust.
  • by OnanTheBarbarian ( 245959 ) on Monday October 30, 2000 @12:29AM (#665707)
    Firstly, I would suggest that anyone who hasn't read the IA-64 architecture book, or at least a decent summary of the contents, should turn the volume down a few notches. Thank you.

    This article is sort of silly. "In effect, trying to patent the instruction set itself" is a fairly vague notion; in fact, what Intel is doing is patenting some of their software techniques (expressed usually in small groups of instructions) for prefetching and control/data speculation. Right or wrong, this happens all the time. If some company has a nifty new caching algorithm, they will patent the idea; not the gatelist and implementation.

    For example, if you could implement a IA-64 clone by (say) ignoring all prefetch instructions, and just fetching the data when it was needed (effectively turning the chk instructions into the actual loads, for those who are aware of this stuff - you could do it with binary translation). While this may not be a very good idea, it wouldn't infringe their prefetching patent, even if you used the same instruction mnemonics and produced a chip that could run the same binaries.

    Personally, I think these patent are potentially disturbing, but put it in perspective with common practice. Read the back of Microprocessor Report sometimes; there are lists upon lists of patents being granted for techniques in exactly the same fashion as above.

    As for the patents, I haven't read them, but I suspect that they'll have a tough time with them. IA-64 didn't spring out of nowhere, and a lot of the ideas that went into it follow a fairly predictable (no pun intended) path of development in academia and industry. A fairly stacatto summary of these paths can be found at Historical background for HP/Intel EPIC and IA-64 [] - if you don't already know something about computer architecture, don't expect to be illumined. The point is, Intel (or more accurately Idea or whatever the Intel/HP collab. is called) hasn't necessarily added that much to prior art here, so the patent may be too broad and subject to either legal attack, or too narrow and easily worked around.

    Oh, and to the people cheering on the failure of IA-64, I beg to differ. Some of us write compilers and binary optimizers and code generators, and the death of the x86 architecture would make us very, very happy. The fact that the first IA-64 is going to be a dog isn't really that suprising - it's a huge engineering task and the first chip was always going to be a reference chip more than a serious performance model.

  • One has to wonder what would have happened if Intel had patented the extended instruction set of the 80386... or the first Pentium chips. No clones?

    IIRC, Intel needed to allow others to create clones in order to qualify for government procurement, which requires more than one source exist for technology. Of course, it's quite logical for them to have supposed that a sizeable chunk of their market would be for the government, after all, how many people in the late 70's knew that computers would be in daily household use by now?
  • Patents have a finite lifetime, as do copyrights (though they keep getting extended). However there's no way that software can go into the public domain just because it is not patented - typically the lifetime for patents is much less than that for copyrights (e.g. 20 years vs 50).
  • In a brave move by processor company Intel, they have now copyrighted the process of moving bits and section of memory to registers where they can be modified/used. When other companies claimed "Prior Art!" Intel referred them to their 8086 processor line.

    When asked why they weighted this long to file this copyright, they answered "Well, we got a few lawyers on lone from Rambus and Amazon so when they suggested it we said -- Why Not!"
  • It's obvious: by patenting the instruction set of IA64, Intel will prevent _any_ competition in that field. And in the hardware world, the best way to kill a system is to prevent the existence of clones (see IBM and PS/2 patents ==> no more PS/2). This is thus clearly a move from Intel to kill the IA64 platform before it only exists. Anyway, that's not so bad, since there are so many other good processor types (see ARM, for example).
  • > I for one will be sticking with MIPS and SPARC until they finally die. Here's to hoping that will never happen.

    Cast my votes with Alpha AXP and POWER architectures...

  • Around about 20 years ago, most computers used in the home had 8k or 16k running on a chip with a 16 bit address bus, which could address 64k maximum. By the mid 80's, up to around 256k was common, but bank switching techniques were required to access it. At the same time, the PC clone started becoming common for home use.

    At the present time, for home use, we've got home systems usually with 128Mb or 256Mb, and the 4Gb address space.

    Many people are having problems hitting the 4Gb limit on servers and other high memory usage systems, I predict that bank switching solutions will again come in vogue for these applications, and within 5 to 7 years, the home user will have the same problems.

  • Evolution would be dumping the intel instruction set and architecture for superior technology. They're just patenting an implementation, which is ass silly, but not truly stiffling to the industry.

  • I beg to differ. Your comment that people should understand the issues involved before commenting is reasonable, but your assertion that "IA-64 patents are legitimate" is ridiculous.

    What is Intel really trying to patent here? Not ideas, that's for sure. They're trying to patent low-level implementation details like how certain instructions impact data flow throughout the IC. Such "inventions" belong in the public domain, and only give ammunition to people who want to abolish intelllectual property altogether.

    The real purpose here is clear for anyone to see: to set up a legal minefield for AMD and anyone else who wants to clone the IA-64. Hopefully the people in USPTO will think twice before rubber-stamping this particular load of crap.

    As to the death of the x86 architecture: it will probably happen at around the same time we switch from fossil fuels and start using the metric system. It will certainly be very chilly in hell. ;)

    Haven't you heard? Old standards never die, they just get extended, extended, and extended some more until you can hardly recognize them.... and they're a real mess. :)

  • My first reaction to this kind of articles is always: "I'm safe here in Europe from the problems with the U.S. patents office"

    Naturaly, the next tought that pops-up is: " Ooops ... There's a big bunch of treaties to make patents valid worldwide

    That get's me worried!!!

    When, and if, the problems with the U.S. patents office are sorted out, the problem will move elsewere - if patents are global in reach, companies will simply try to patent things (anything) in the places in which they are more likely to be granted so as to increase the number of worldwide patents they have. Competition between "patent heavens" (places were patents are easely granted), will drive restrictions on the granting of patents to zero.

    I can just see the great patents of the future:

    • "A method to obtain the number of elements of the union of two groups from the number of elements from each inicial group" (sum)
    • "A method of maintaining a human body in the surface of a liquid without auxiliary equipment" (swiming)
    • ...

    Excuse me, i have to go and throwup now ...

  • The original IBM MVS OS (designed for OS/360, the base for OS/370, OS/390, and now OS/Z) was not patented, so it entered the public domain. IBM later fixed this with later versions (IBM leads the world in the number of patents filed, but it used to compete with AT&T for this distinction).

    I know this partly because I was called in to secure the records of a certain collection agency in SoCal several years ago. They ran the public domain (old!) MVS on end-of-life IBM hardware (for which I think they didn't have maintenance, as that cost more than the hardware was worth).

    In case you didn't catch the drift of "secure the records," the company was in receivership, and it was the sort of situation where someone was going to go to prison (misappropriation of customers' moneys). The owners were on the lam, and the poor guy who was Operator/Systems-Programmer gamely helped me to capture their files (on the old 9-track reel tapes, using bare TSO commands). [I do hope he found a good next job - he was good but saying he knew MVS 0.9 wasn't leading-edge!]

    Then there was the owners' secretary, who - to quote the Partner on the job - "liked the cut of my jib" - and hinted about her upcoming weekend in Palm Springs.

    Ahem,... it's now usual for hardware companies to patent their instruction-sets. Intel, AMD, MIPs, et al - they all do it as a matter of course in their business operations.

    However, the real issue is _which_ instruction set will gain favor with developers. Will it be the (lame, IMHO) Intel Itanium, or AMD's better Hammer architecture?

  • Is this an attempt to block out even reasonable competition, or is this just "business as usual" as at least one voice in the story says?

    Why can't it be both? Both an attempt to block out even reasonable competition and and business as usual.

    I don't mean business as usual, in the Microsoft sense of being in the actual business of eliminating competition, but I mean "as usual" in a sense of this is the kind of thing that big corporations do?
  • And where are you going to find the money to sue Intel?

    Unfortunately, in the US today prior art does not stop you getting a patent. Instead, you have to file a lawsuit challenging the patent first, then bring the prior art as evidence, and hope the judge agrees with you.

  • by segmond ( 34052 ) on Monday October 30, 2000 @05:23AM (#665750)
    Stop blaming the USPTO for everything. It is our problem, we are the cause, yes you and me. Listen very carefully, USPTO doesn't just provide patent for computer technology. The provide patents for every subject you can possible think of in this world. Awarding a patent to anything requires intimate knowledge with that field which enables you to know if the patent is legitimate or not. Therefore as you can see, for the patent office to honor only real and non stupid patents in the computer field. They need real computer professionals, people who intimately understand computers, follow the field, and keep in touch with it. Now back in the days, It was fun working in the patent office because you get to see so much "new?" things and such. But today, it all boils down to money. Will you give up your $90k job for $35k to work for USPTO? And how many slashdotters will do so? Thus as you can see, the problem is me and you. We are greedy, and this is just part of the things we get in return for our "greed."

  • They has a resonable time limit on patents 3-5 years seems to a decent limit. The real problem is that alot of things that are being patented are things that if was ask by any decent programmer they would have came up with the same solution. Well I guess the other /.ers are going to have with this one.
  • by matman ( 71405 ) on Sunday October 29, 2000 @10:38PM (#665752)
    Well, this is sort of interesting. In a way, it will force motherboard and software manufacturers to build in more modular fassions, so that supporting different chips wont be so hard. What this means, is that hopefully new chips will be able to be more liberal in their design, hopefully speeding advancements. Other companies (hopefully) should forced not to copy intel, and to come up with some new stuff. Now, if mobo manufacturers dont start modularizing to be able to support different sorts of cpus, it may be hard to use non intel chips, even if they exist, because the other hardware that they would need to co-exist with wouldnt exist.
  • There was the KFKI TPA-11/580 [] VAX clone.
  • by 91degrees ( 207121 ) on Sunday October 29, 2000 @10:42PM (#665757) Journal
    They're a big corporation. They'd be stupid to do anything else. The real problem is with the USPTO.

    As soon as anyone gets a patent they immediately receive a virtual monopoly on that product. It lasts for a lot longer than the product could practically be useful for. It needs to be sorted out, but by whom? The US Government ain't gonna do it. They need the corporate contributions.
  • by gorilla ( 36491 ) on Monday October 30, 2000 @12:00PM (#665758)
    The fact that the USPO doesn't pay the market rate, or close to it, for professionals is the USPO's problem, not ours. If this means that they have a 25 year backlog on patents that expire 17 years after the start of the process, then I'm sure that the companies asking for patents will get congress to provide the USPO with enough money to hire qualified inspectors. Unfortunatly, the USPO has decided that they will instead approve all patents, as rapidly as possible.
  • by GrandCow ( 229565 ) on Sunday October 29, 2000 @10:43PM (#665761)
    So let Intel patent their poor-performing functions for the IA-64. How many people will really care? personally I'm not giving it more than a casual thought. When I'm ready to move up to the 64 bit processors, I'm already banking on AMD's 64 bit solution, which is also incredibly fast with 32 bit processes.

    This patenting of specific functions smacks of Microsoft tactics, but oh well. This might have worked back with the 16 and 32 bit processors, but since AMD is developing a completely different architecture, the big competition is still going to be there.

  • 64-bit processing is still in its infancy in the consumer world (either vaporware or prototype, but still not released).

    Hello?!? Alpha? SPARC?

    Yes, this might be an assembly/C/C++ programmers nightmare, but isn't that why programmers are always learning new things?

    Can you explain what you mean here? I don't follow.

  • Have you ever read this kind of patent? Since my last post, I headed over to the patent database, and sure enough, it's another one of those dull and windy patents about a specific prediction technique. After skimming one of the patents mentioned in the article, I had to skip "LOADRS instruction and asynchronous context switch," lest I go into a coma.

    Yes, they are patenting ideas, albeit mindnumbingly specific ones (something for which we should be grateful; the specificity and dullnes s of the patent application is a bit hint at how little new stuff is introduced with one of these patents). I don't think that they _should_ be able to do this. I didn't say that I thought Intel having these patents is good, I said that they were possibly legitimate, in the sense of all the other CPU architecture patents out there. Personally, I think most of these patents are utter drivel; a mountain of prior art with a tiny cairn of original work perched on top of it.

    I don't know what you mean by "how certain instructions impact data flow throughout the IC". Ummn, don't all instructions affect data flow throughout the IC? This is so vague as to be meaningless. Can you give an example of how you think such a patent could restrict a whole class of other implementations?

    Despite all this, I'm mostly in agreement with you here. I do think they are laying a legal minefield for other IA-64 implementers. I wouldn't call it cloning, as one clones a chip, not an instruction set. I am pessimistic about the USPTO doing anything; most of those other patents at the back of MPR went through without any problems that I heard of. The fun starts later, when two deep-pocketed companies get into it.
  • by ishrat ( 235467 ) on Monday October 30, 2000 @02:02AM (#665770) Homepage
    BountyQuest is a Web site that rewards people $10,000 and up for information that challenges patents. Rea d more [] on this.
  • the real issue is _which_ instruction set will gain favor with developers. Will it be the (lame, IMHO) Intel Itanium, or AMD's better Hammer architecture?

    Maybe the real issue is, which instruction set will gain favor with customers. I've been around the block enough times now to see that technical superiority means absolutely nothing as to whether something will gain favor in the marketplace. Let's see... MS-DOS/p-System, VHS/Beta, 8088/68000, IBM-PC/Macintosh, USB/Firewire, MS Windows/(insert any other OS here), etc.

    [Aside: the original IBM PC was offered with two operating systems: MS-DOS and the p-System. The p-System was superior in many ways.]

    As to whether it will be the Intel Itanium, or AMD's Hammer architecture, maybe it could be both? Wouldn't it be nice if we didn't have to have this kind of standardization? Maybe we could have standard memory, busses, card slots, usb, and other interfaces to ensure component interchangability, but maybe we no longer need to standardize on a particular instruction set? Wouldn't this free up resources to do real innovation, rather than expending effort to come up with innovative ways to make the ancient 8088 instruction set faster? After all, all modern OSes are architecture neutral. Apple developers have now been shipping "fat binaries" (i.e. compiled for two different architectures) for years. [This naturally begs the issue of doing the final code generation on the target OS.]
  • Let's try and get this patent pushed through as quickly as possibly


    Because at the rate Intel is going, it will expire before working chipw actually hits the market

    BTW, How long have SGI and Sun been using 64bit Risc?

  • What I don't understand was at some point it wasn't legal to patent ISA's. See IBM versus Amdahl as the example. I thought this got litigated on during the 70's and IBM lost? But then, you would think patenting algorithms, i.e. natural laws was illegal. Guess that's changed somewhere along the line.

  • It's amazing what some reporters will build a story out of. Anyone else notice the total lack of this guy to commit to anything?

    ...more than 20 new patents "suggests" Intel Corp. is expanding...

    Some experts "wonder" whether Intel...

    ...the company "appears to be", in effect, trying to patent...

    Not that I'm saying that it isn't possible that Intel is doing this, but the fact that they suddenly submitted a bunch of patents hardly constitutes evidence. Not to mention the fact that the reporter hasn't even gotten anyone to go on the record to claim authorship of this pondering. So basically its all just suppositions. Certainly always possible, and knowing computer corp. strategies even likely, but shoddy journalism is still shoddy journalism.

  • by sabre ( 79070 ) on Sunday October 29, 2000 @10:48PM (#665782) Homepage
    This seems pretty strange to me, lets consider the situation that Intel is in:

    1. Merced/Itanium has been delayed countless times.
    2. Performance predictions (at least for the first generation or IA64) have been scaled back to the point that it appears that IA32 will be more performant in the same timespace
    3. AMD is going forward with their own 64 bit chip, the sledgehammer. This has the advantage that it will (probably) have a much smaller die and use much less radical design techniques.
    4. IA64 is so tied to compiler technology (that isn't good enough right now) that performance will be a huge problem at least for the near future.
    5. IA64 is, in general, in a state of massive hemmorhaging. (as with most of Intel's near future plans, but that's another story)
    Given all of this, I'm inclined to believe that their patents (which I'm sure they will get... :( aren't worth the paper they are printed on.

    "Bring it on Intel"


  • by KillerBob ( 217953 ) on Sunday October 29, 2000 @11:00PM (#665786)
    I don't really mind Intel doing that. I also don't think Intel will actually enforce such a patent:

    They're reliant on Microsoft to stay in business. While people don't really have a choice about Microsoft (don't. You know, and I know, that there are always better options to Microsoft, but do you expect Joe Idiot to install Linux?), they do have a choice about Intel. As more and more people are shying away from Intel in favour of IDT WinChip, Cyrix, AMD, and other giants, Intel would only shoot itself in the foot by enforcing such a move.

    Microsoft would not be very friendly to having to write new versions of all their operating systems, each coded to a different architechture. They aren't very happy with having an Alpha and an x86 version of Windows NT. How do you think they'll react to an Intel, and x86, AND an Alpha version of Whistler?
  • by RedWizzard ( 192002 ) on Sunday October 29, 2000 @11:00PM (#665787)
    My first thought was that it's ironic that Intel are trying patents on instruction sets now, given that no one has expressed any interest in cloning IA64 at all. AMD have their own 64 bit architecture after all, and it looks more acceptable for desktop use than IA64. This [] is an excellent overview of AMD's plans.
    Of course with all the convenient secret patent 'licenses', intel will naturally be using this exclusively to sue AMD or anyone else trying to compete with them.
    Until Rambus appeared these sorts of things were almost always resolved via patent license trading. AMD have a fair portfolio of their own, I doubt they'd be afraid of anything Intel are likely to do.

    Nope, my guess (without having checked the details of the patent) is that this is an attempt by Intel to get some leverage over Transmeta (or anyone else) incase they want to simulate the instructions in software. I'm sure Intel would love to get their hands on some of Transmeta's patents.

  • by gunner800 ( 142959 ) on Sunday October 29, 2000 @11:01PM (#665788) Homepage
    I would like to patent the idea of my phone number causing my phone to ring.

    Sure, people have been using phone numbers for similar effects for a long time, but this particular number has always been mine.

    My mom is not a Karma whore!

  • by jovlinger ( 55075 ) on Monday October 30, 2000 @05:56AM (#665792) Homepage
    This patenting of specific functions smacks of Microsoft tactics

    I'm suprised. Really. I had long understood that the industry had agreed that specifications are not patentable. Implementations, yes, but not specs. This is why it is legal to create plug-in replacements for things like windows, libc, whatever; because their APIs are unprotectable. Of course, you are under no obligation to publish any or all (c.f. intel undocumented instructions) of these APIs.

    Surely the instructionset of a CPU is its API? Now if intel were patenting nifty tricks to implement these instructions, all would be well, but I just don't see how anyone could think it possible to patent an API (MIPS case nonwithstanding -- key here is the subclaim "and method for same", I think).

    So I'll throw this out for discussion: what nifty implementation tricks do you see as patentable?
    Negative example: the tomasulo algorithm, which is a big ball of hair implementationwise but effectively just renaming.
    Positive example: outof order commit with exception coherence.
  • is an attempt by Intel to get some leverage over Transmeta (or anyone else) incase they want to simulate the instructions in software.

    Transmeta's Code Morphing technology is designed to emulate CISC architectures. Itanium is a VLIW (explicitly scheduled RISC) machine.

    I'm sure Intel would love to get their hands on some of Transmeta's patents.

    Transmeta's most obvious patent that I can see is a "commit/rollback" structure for registers. However, prior art for this dates all the way back to the Z80 processor, which had two parallel sets of registers.

  • Hey, I submitted to the IOCCC, and am waiting patiently for them to finish
  • Don't blame the murderer. He's too psychopathic to do anything else. The real problem is with law enforcement.

    No, it's not a perfect analogy, but it gets the point across. For every company that abuses the patent office, there is another that does not. Just because a system allows you to exploit it doesn't make it your obligation to do so, nor does it remove any moral obligations to behave responsibly.

    And who do you think payed for the patent office to get to this state? Do you really think they decided they were just to strict in their criteria for handing out patents? Or that incompentence magically appeared in the last twenty years, where as they had been doing a great job beforehand?
    Bush's assertion: there ought to be limits to freedom
  • There was a report on a while back that it's indeed possible to run two Athlon CPUs on an Alpha dual-slot motherboard.

    - A.P.

    * CmdrTaco is an idiot.

  • by peter303 ( 12292 )
    They've tried to get out of that rut several times
    in the past. Chips since the Pentium essential
    emulate x86 on a RISCish substrate. Emulation
    will continue as long as the software is around.

"The Avis WIZARD decides if you get to drive a car. Your head won't touch the pillow of a Sheraton unless their computer says it's okay." -- Arthur Miller