Forgot your password?
typodupeerror
Technology

Transmeta to Release Processor in January? 152

Posted by Roblimo
from the one-can-only-wonder dept.
Scipius writes "German tech-mag c't reports that Transmeta's new processor will likely be released on the 19th of January 2000. It also reveals the apparent code name: Crusoe." The article's in German, of course. But we'll take a juicy Transmeta rumor - and that's all this is - in any language. Babelfish time!
This discussion has been archived. No new comments can be posted.

Transmeta to Release Processor in January?

Comments Filter:
  • I'm really glad to see from progress out of this company. Despite the vapourware rumors, that is. Should be interesting to see how the competition goes. Perhaps a real intel downfall?
  • The rumor kitchen reports that the geheimnisumwobene processor prozessorschmiede Transmeta on the first Comdex day, to which 15 November -- as announced by Transmeta coworker Linus Torvalds already - admits now finally the date for the conception of their long expected processor to give wants.

    *snicker* .. the rumor kitchen. Mahir, head chef of the rumor kitchen, kiss you!

  • I really hope to see some 90's processor technology, no more of Intel's 70's technology. :-)

    Yeah but would still be 10 years out of date. :)
  • c't claims that it is thought for notebooks, due to it's low power consumption. Maybe "Crusoe" is cheap also ;-)
  • by A Big Gnu Thrush (12795) on Wednesday November 10, 1999 @11:41PM (#1543583)
    If Transmeta is a fabless chip manufacturer, then this will just be an announcement of what they plan to build. Not that I'm accusing them of vaprware, but early chip announcements tend to be more like bad sci-fi than actual news. Think of Intel and their gHz processor - cooled by a desk sized freezer beneath the unit.

    Either way, it will be more interesting than the Lucent (might not be a) router announcement.
  • by Bob Ince (79199) <and@[ ]desk.com ['dox' in gap]> on Wednesday November 10, 1999 @11:43PM (#1543584) Homepage
    it isn't quite hard to quess that transmeta's projects have something to do with processors.

    Well, yes, we'd kind of known that for some time. :-)

    The new-news hidden in this article (apart from the codename itself) is that Crusoe is to be aimed at low-power devices like laptops, which is quite a different market to the mega-workstation many people here wanted it to be (perhaps due to dislike for Intel).

    'Course, low-electrical-power doesn't mean low-computing-power. Look at the ARM series, for one. Or c't could be wrong, though they usually aren't. Guess we'll just have to wait and see - to use a phrase already worn out in Transmeta discussion...


    --
    This comment was brought to you by And Clover.
  • Hmmm, I guess I might just have to wait before buying my new laptop.
    By the way, is this going to be an entirely new architecture, or is it compatible with some already-extant standard?
    (And the obvious question) Will Linux run on it? :-)
    (Perhaps that's why they hired Linus, hmm?)

  • kinda like that biz markie song...

    on non-flamer tip transmeta's most recent patent release pretty much let the cat out of the bag. if they deliver i'll be first in line to order!
  • by ylle (38991)
    Transmeta's home page has been updated. It is apparently Y2K compliant. Does that mean we can't expect to see it change before next year?
  • Anyone seen they changed the transmeta website ?
    how long has this 'new' site been up
    their new design really kicks ass!
    never seen a website so well designed, and it works with all browsers and it loads fast too. great! kudos to their webdesigner! :-)

    ---
  • Hey! I didn't see the Y2K-compliance mentioned last time I checked the site. Isn't that progress?!? :)
  • by cureless (35682) on Thursday November 11, 1999 @12:04AM (#1543590)
    If it's a processor then some software is going to run on it? This could explained, as mentioned in an article above, why they hired Linus.

    Think about it. A *new* architecture, with some really fancy new characteristics but no software? I don't think so. On the other hand, if they involve the opensource community . . . BINGO! a real competitor to Wintel.

    Who else would you choose as youre link to the community but Linus. He's the head of development. He can make sure everything runs on this "Crusoe".
    I'm not directly tied into kernel development, but you sometimes wonder why some patches don't get included . . . :-)

    But in the end, if it's a processor, Linux will run on it. How else can you have a top-secret processor? Who can they really trust as their OS of choice? Windoze? MacOS? BE? . . .
    You never know, kernel 2.4 might as well be ready to run on "Crusoe"

    cl

    --
    Reply . . . let's get it over with
    --
  • In English: The rumor kitchen reports that the geheimnisumwobene processor prozessorschmiede Transmeta on the first Comdex day, to which 15 November -- as announced by Transmeta coworker Linus Torvalds already - admits now finally the date for the conception of their long expected processor to give wants. But already beforehand the message penetrated for c't editorship that was to be introduced the processor circulating under the name " Crusoe " on 19 January 2000 (by the way one Wednesday, no " Friday ").

    Crusoe is to direct owing to its very low current consumption primarily at the Notebook market (s/c't)
  • Can you imagine the slashdot effect thier web server(s) will see once they have some real content? They'd better be ready for one bumpy ride...
  • So, why is it going to be called Crusoe ??

    There must be a reason for this name?!

    It's years since I read the story but involved some bloke shipwrecked, washed up on an island, and meeting a native he called 'Friday'.

    Hmm...

    Or are they just playing with us?!!

    BTW: www.transmeta.com [transmeta.com] has changed! (And no, www.transmeta.com/crusoe.html and www.transmeta.com/crusoe/ dont exist!

  • by Sir_Winston (107378) on Thursday November 11, 1999 @12:11AM (#1543595)
    The fact is that if the Transmeta CPU architecture is anything like what's in the Transmeta patents, and if they can at least come up with a few engineering samples, if will mean a radical shift in our ideas about processor design. As it stands, the instruction set is what defines a CPU--CISC, RISC, x86, HP-UX, etc., are all involved in defining the processors which use these instruction sets, but Transmeta changes this. The Transmeta idea as expressed in their patents would create a category above this--no longer is it 'an x86 processor', it's 'a processor running x86 instructions'. This is a radical idea, and a radical paradigm shift--we should all hope it comes to pass. If it's a great and practicable design, it shouldn't be too difficult for Transmeta to partner up with anyone from AMD to Motorola. This sort of radical advancement--again, if it comes to pass--makes me wonder what the heck Intel and all their capital were doing designing the inflexible Itanium, which executes its native (and sure to be poorly supported except for network/server apps for at least a year or two till prices come down remarkably) instructions with Alpha-killing speed but chokes on anything else including the x86 with which they were supposed to be compatible to some reasonable degree. Just 2cents from a guy who plans on supporting anything but the Itanium (mmmm, legacy games under 64-bit AMD....)
  • by Daffy Duck (17350) on Thursday November 11, 1999 @12:11AM (#1543596) Homepage
    I'm as excited as anyone by the prospect of totally new technology, but I'm also bracing myself for the possiblity of both short and long term disappointment when this mystery product debuts.

    In the short term, it could turn out that the product isn't the fusion-powered anti-gravity time-travel device that all the secrecy has led me to expect.

    In the long term, even a fantastic product could end up going nowhere. I'm thinking particularly of AMD's woes. Not only has Intel (allegedly) managed to convince some major motherboard manufacturers not to ship their Athlon boards, now [cnet.com] they've gotten a major OEM (Gateway) to drop all AMD processors from their product line. And Intel's anti-trust case inexplicably disappeared into thin air.

    Even if Transmeta has the coolest CPU ever, do they stand a chance against Chipzilla? Here's hoping...

  • by Anonymous Coward
    It's not exactly vapourware, since they haven't announced ANYTHING. Real vapourware should be competing in the marketplace preferably years before it becomes a real product, to keep potential competitors from designing something similar.
  • Does this mean the whole "morph host" x86 emulator theory is out the window?
  • by xtra (24662)
    and did you check out the source code:

    There are no secret messages in the source code to this web page.
    There are no tyops in this web page.
  • Reason for the new look website:
    Someone finally worked out how to survive the slashdot effect :-)
    Did anyone else think to look at the source?- worth a look.

  • by Stephen (20676) on Thursday November 11, 1999 @12:34AM (#1543601) Homepage
    Here's a better translation, albeit only based on a bit of high-school German and a small dictionary. E&OE.
    The rumour on the grapevine is that the secretive processor manufacturer Transmeta will finally reveal the date for the introduction of their long-awaited processor on the first day of Comdex, 15th November (as already announced by Transmeta employee Linus Torvalds). But c't has already heard that the new processor will be launched under the name "Crusoe" on 19th January 2000 (a Wednesday, not a "Friday", by the way). Owing to its very low current consumption, Crusoe will be aimed primarily at the notebook market.
  • The comm ent [altavista.com] doesn't specify what are its sources, but it:
    1. Confidently states that Crusoe works by special hardware translating the instructions and then storing them in a huge cache,
    2. Says that because of that, MS-Windows will probably not run on Crusoe. This is because of Windows' habit of altering its code on the fly for reasons of optimization, and
    3. Speculates that Linus was hired because Linux is to be [one of the] first OS[s] to run on Crusoe (ok, so this isn't new :D ).

    It looks like both Intel and Microsoft are facing Interesting Times... :)
  • by BigTed (78942)
    The name Crusoe could be a suggestion at what the developers think of their new chip. Being seperate from the rest of the manufacturers but close enough that it is still reminiscent of what users are comfortable with.

    A new platform that runs native Linux and without all the flaws and inherited legacy hardware in the INTEL architecture, sounds good to me :-)
  • Not really. Those not-so-opensource-products-with-no-source-code might need to run on a x86 achitecture.

    This processor might be able to run non-native (what ever native means) linux binaries, it might even be able to run the whole thing; linux-x86-on-Crusoe, linux-ppc-on-Crusoe, linux-sparc-on-Crusoe, and obviously linux-Crusoe.

    Also, once the plataform is up and ready, sooner or later other stuff (non-linux) might want to be ported, and while it's ported a non-native version can be "emulated" (what ever that means to this processor).

    I don't remember the whole discussion on the patent thing, but did it specifically target x86? What about other architectures, even new architectures? This might give Transmeta the upper hand in a place where 64 seems to split compatibility all over.

    cl

    P.S. I'm making stuff up as I go so don't trust me.

    --
    Reply . . . Let's get it over with.
    --
  • Linus Torvalds already - admits now finally the date for the conception


    Linus is having a baby! I hope his boys can swim.


    "If you can't take the shoddy journalism than stay out of the rumor kitchen."




  • by ZahrGnosis (66741) on Thursday November 11, 1999 @12:58AM (#1543607) Homepage
    This is why Transmeta is absolutely cool. Notwithstanding people that can read German (and I'm sure you're plentiful), we've got a one paragraph article that babelfishes really poorly about some bizarrely unsubstantiated rumors and it's going to be a VERY popular slashdot thread, because, well, because it's Transmeta. (Circular logic... cool).

    As for real content, I'm surprised by even the rumor that the supposed chip would be a notebook chip. Why a notebook? Linus has said recently that Linux is likely to develop towards embedded applications (it really does perform well there). How let down would we be if Transmeta's first chips were low powered, linux-powered embedded app chips? Really think TV/Network Computers or the like...

    Also, if I read the babeled German correctly, they're going to announce the Concept on Jan. 19th. It still could be sometime before we see product (whatever it may be). This should come as no surprise, since TransMeta clearly hasn't employed hundreds of chip-builders lately (someone would have noticed that, I think).

    I'm waiting to be awed by whatever they eventually produce, but for now, it's enough to be in awe of the amazing hype and free-publicity. Amazing, isn't it that doing the exact opposite of Microsoft (by spending NOTHING on advertising) is garnering TransMeta (and thus Linus, and thus Linux) a decent amount of press?

    Keep it up TransMeta!
  • by kzin (3595)
    If Crusoe is really aimed at laptops, wouldn't Linux be a strange choice for its main OS?
    Now I know we all like to think Linux is better at everything, but right now it's still a server OS that's rather hard to operate for non-techie users, which makes choosing it as the main OS for laptop a risky vote of confidence.
    Wouldn't MacOS make a much better choice, especially considering its recent revival? (iMac)
    Well maybe there's already a hidden deal with Apple. Or maybe Linus is developing Linux to take advanteges of Cruso's unique features and be the first OS to run everything. That would be nice... :)
  • Transmeta processor " Crusoe " in January expects

    The rumor kitchen reports that the geheimnisumwobene processor prozessorschmiede Transmeta on the first Comdex day, to which 15 November -- as announced by Transmeta coworker Linus Torvalds already - admits now finally the date for the conception of their long expected processor to give wants. But already beforehand the message penetrated for c't editorship that was to be introduced the processor circulating under the name " Crusoe " on 19 January 2000 (by the way one Wednesday, no " Friday "). Crusoe is to direct owing to its very low current consumption primarily at the Notebook market. (as/ c't)

    --

  • I'd have to agree. All of this secrecy is causing quite a large number of people to get really anxious about the possibilities. In all reality we have little reason to belive this "product" will do anything, aside from the patents.

    From the patents we can derive quite a lot about what "it" is supposed to do, however because patents are public information, "it" might not be what transmeta is really doing. Transmeta is being really secretive thus the information in the patents is what transmeta wants us to "know". To take it one step further, transmeta could be working on anything, and the patents could be an attempt to mislead those who copy or stop them, or the patents could be on an auxiliary technology (ie. supporting chipset).

    My friend and I joke about what transmeta is really doing, we have concluded that they are making chips that will power the next generation of washing machines. Don't expect anything from transmeta. I'm not saying they are making vapor, I'm just saying that if you don't expect anything you will be nicely supprised when they make the awesome chip you never expected :)
  • Yeah, but half the reason Linux is hard to operate involves less than stellar hardware compatability. I know Linus has been interested in improving Linux's laptop capability (due to purely selfish reasons, natch) and if this is a new architechure, one would imagine it would require a new, designed-for-Linux bunch of hardware that would be easily supported.
  • by Anonymous Coward on Thursday November 11, 1999 @01:29AM (#1543614)
    Isn't this going to be capable of becoming a JAVA OS.
    If it stores the machine instruction set in a local cache and runs it from there.. Well that seems to be exactly what JAVA would need to have a Java-Chip.
    Considering it's low power consumption it would make an excellent candidate for the jini - appliance environment as well as the PC. Add to that the potential JAVA chip concept and you have the Jini Project from Sun sitting in your lap.
  • by Anonymous Coward
    My IE 5.0 loads the page 5 MINUTES and then CRASHES! There must be an EMAIL WORM on the page! I think I have to install a SERVICE PACK from BILL!
  • According to my sources, that processor-related patents are just side-effects of tests of the new caffeinated beverage they're developing...
  • Well, I don't think my translation is in much better English than the babelfishes one, but... - oh well, I'll have a try. Here it comes:

    The rumour says that the mysterious processor company Transmeta will eventually anounce the release date of their awaited processor on the first day of Comdex - 15th November - as earlier stated by Transmeta-coworker, Linus Torvalds. However, a message got through to the editorial board of c't, stating that the processor - codenamed "Crusoe" - will be presented on January, 19th 2000 (which is, by the way, a Wednesday, and not a Friday). Crusoe, which is supposed to have a low energy uptake, is believed to be aimed at the laptop market.
    Regards,

    Babeluary

  • Just search for transmeta on the uspto.gov web site among the *trademarks*... (BTW: it can be long known from the maniac rumour monger yours truly here [linuxtoday.com] in this [linuxtoday.com] Linux Today article...)
  • by clarkma (32199) on Thursday November 11, 1999 @01:51AM (#1543619)
    So Transmeta are finally going to be ready to *say* something. The funny thing is that their patents are quite revealing about what they're up to - a speeded up version of the self-modifying FPGA technology that has occasionally spawned 'new era' claims. I'm not saying that their chip is just an FPGA, but that the effect is meant to be much the same: a metamicrocode that can be optimised in near-real time by a JIT-like (or is dynamic compiler a better term than JIT?) compiler and scheduler.

    Please though, don't beleive all the speed hype. Remember, it was a year ago or so when 1GHz sounded astonishing, but now it's almost boring for those chiller guys. The thing is going to be *flexible* not *necessarily* fast.

    Curiosity killed the cat, but who ever saw a cat reading a patent application?

    .sig thingy
  • by ebcdic (39948) on Thursday November 11, 1999 @02:03AM (#1543622)
    As has been pointed out elsewhere (comp.arch I think), the patents that are available on the net reflect what Transmeta was doing a couple of years ago, when the patents were filed. So don't be sure that they are central to their current plans.
  • #1 and #2 and the previous patents collide badly:
    the processor should be able to run self-modifying
    code. Naturally, it will be a lot SLOWER because
    of the retranslations... If it is in critical timing loops, then just maybe there is a problem
    but on the whole, I'd think this is a rosy herring.
  • by Anonymous Coward
    Uhm, exactly what does this mean in their trademark registration?:
    --
    Serial Number: 75422458

    Registration Number: (NOT AVAILABLE)

    Trademark (words only): TRANSMETA

    Current Status: An office action making FINAL a refusal to register the mark has been mailed.
    ^^^^^
    Date of Status: 1999-08-05

    Filing Date: 1998-01-23


    CURRENT OWNERS

    1. Transmeta Corporation

    GOODS AND SERVICES

    computers; computer operating systems; computer hardware; computer peripherals; integrated circuits; semiconductors; printed circuit boards; firmware; middleware;
    utility software; and application software for use in connection with computers systems, computer hardware, computer peripherals, integrated circuits, semiconductors, and
    printed circuit boards


    PROSECUTION HISTORY

    1999-08-05 - FINAL REFUSAL MAILED

    1999-05-03 - COMMUNICATION RECEIVED FROM APPLICANT

    1998-11-03 - NON-FINAL ACTION MAILED

    1998-10-20 - ASSIGNED TO EXAMINER

    1998-10-20 - ASSIGNED TO EXAMINER
    --

    Does this mean Transmeta Corp. is up a creek without a paddle on the trademarks or does it mean that they can't get a trademark for vaporware?
  • by Dacta (24628) on Thursday November 11, 1999 @02:16AM (#1543625)

    Had anyone else heard the rumor about the Transmeta chip being low power consumpion before? I sure hadn't, and to me, it doesn't mesh well with the idea that it can run multiple instruction sets.

    Surely this would require a large amount of memory, and isn't (fast) memory something of a killer for low powered devices?

    If Transmeta can produce something that emulates other architectures, and uses a comparable amount of power to the low power versions of those architectures, it has to be one of the most impressive breakthoughs ever.

    I do worry though - you know what they say -

    A chip can be fast, cheap or effecient - pick any two.

    Okay, I made up the quote, but I think it is slightly accurate at least, esp. in the early generations of a design.

    What else.... Oh yeah.

    If they are really going to announce this in January (or at Comdex), I don't think we will see it in use anywhere for a couple of year. If Tranmeta had contracts with fab plants somewhere, someone would have said something by now.

    I doubt very much if you can go down to your local chip maker, and say "We want you to switch your plant to making our funcky new designs - forget about this multi-billion dollar contract you have", so they can't just get manufacturing facilities like that. It takes a long time to build a fab plant, too, and it's not like you can just convert a derelic factory to a state of the art chip fabrication plant.

    --Donate food by clicking: www.thehungersite.com [thehungersite.com]

  • Ah Mahir [xoom.com], the cause of and solution to all of life's problems... or just the real name of Transmeta's new chip... you decide! [salon.com]
  • TransMeta has been around long enough that silicon may be close if in fact test silicon isn't already there. Given TransMeta's reluctance to say anything at all regarding their product I'd say that if they do announce a processor on November 15th they'll also announce a relatively short time till actual delivery.

    You don't need a fab to build a microprocessor. Many CPU's are built using third party fabrication resources such as the MIPS microprocessor.
  • I realize that, but while Transmeta people seem to keep their mouth shut because they really enjoy their work, do you really think some mid-level manager is going to care?

    Here Dave, make sure we have enought people to staff the canteen to server X people from 15 January.

    Oh, Okay Jim, what's happenening?

    We've just signed a new contract with some manufacture, but you can't tell anyone.

    - Infact you should probably have some factor in there for the motivation of them, too.

    --Donate food by clicking: www.thehungersite.com [thehungersite.com]

  • Nono :) Transmeta's cpu is a VLIW cpu. That means that MANY MANY transistors that in other architectures are devoted to out of order execution and/or superscalar esecution are not 'wasted' on transmeta's cpu. So, this mean less transistors and less power consumption. All the translation stuff is done by software (with hw aid), so complexity is moved out of the hw (cpu die) and moved into software. (emulators and compilers)
  • Yeah, but you know what they say about having kids: they never grow up the way you want. Linus's kid will probably be totally anti-drug and work at microsoft (probably writing Windows 2030).
  • Keep in mind that not all CPUs run user os's.. Perhaps it's some embedded system that isn't even positioned to compete with Microsoft Windows. (Of course, it would be competing against MS's lame attempts at an embedded os.)
  • Didn't I hear a rumor a while back that transmetta was talking to IBM about having them build whatever it is that they are developing? I very well might be mis-remembering, so take this with a grain of salt.
  • Soft instruction sets ? That's gonna make it really easy for compiler writers - NOT.

    If you want to see truly excellent processor technology take a look at the ARM processors.
    www.arm.com

    Very low power, very fast and modular.

    For innovation look at the clockless designs they've prototyped - they're really funky - fast and use next to zero power.

    They're also Intel's biggest headache - having inherited the rights to the StrongArm they don't know what to do with it. It craps on all their processors and is CHEAP!


  • Maybe the chipset needed to run it will be called Friday. ;)
    Or they may be implementing "advanced Friday cache architecture". =)

  • Does anybody else find it odd that this processor is being touted as "primarily for notebooks". If it is really the revolutionary beast that it is rumored to be, it would make much more sense to have it first on desktops than a notebook. Years after, when it will have replaced the Intel processors, we'll see it on notebooks, just as we have Intel compatible processors on notebooks now.
  • to quote theregister

    That said, Transmeta's filed patents hint at technologies to make applications run
    considerably faster than they do now, at a given clock speed. So if you're willing to run them
    at standard speed, you could, we imagine, run the chip rather more slowly than your
    average PIII and thus make a big saving on power.
  • Are you the guy who makes all these really strange fortunes? :-)
    --
  • or it could just be named crusoe to give us a witty name if it flops...

    will crusoe=gilligan?

    :)
  • Targeting at the low-electrical-power market does usually mean that it isn't (yet) competitive in the high-computing-power market. It makes sense to sell your new product into the niche that suits it best, just to get a foothold and some income before you try for a more difficult area. If it can take on the high-power end adequately, there's not a lot of point in restricting it to a particular market segment.

    The Arm is reasonably powerful now, but wasn't always : before StrongArm, it was computationally powerful for it's price and wattage, but not really comparable with 486/Pentium.

  • Ha, I guess there ARE people out there that know what the hell those patents were talking about! :)
  • Yeah, because we all know notebooks are cheaper than desktops... :P
    Stan "Myconid" Brinkerhoff
  • Hmmm, I guess I might just have to wait before buying my new laptop.

    Why not wait until the next generation of chips, or the next next? I heard thoes cpus 3 generations are mad sweet...
    Stan "Myconid" Brinkerhoff
  • This is the biggest news story of the week... it should be on the front page :^)

    Steve

  • There are two replies to this article that are rather interesting. Follow the links containing:

    No boards (nevertheless!), and faster despite cheaper (x, 11,11,1999)

    (Enlish butcherization)
  • Actually,no. Haven't you noticeda shift over the past few years to putting better technology into notebooks?

    Many OEMs (correctly, I think) see notebooks and even desktops derived from notebook technology like Gateway's Profile as the wave of the future. It's likely that conventional "desktop" technology will die off over the next few years.

    This will be a good thing - computers will get smaller, quieter, more power-efficient, more flat-panel-ready, and finally, the huge gap between notebook and desktop computing costs should close considerably.

    Notebook technology is the future...
  • Here's a question for people who know more than me:

    Suppose Transmeta actually does have a really cool new chip. Obviously they are going to need some help developing software (compilers, etc.) for it. I assume that's where Linus fits in. But seeing as the trend these days is to get Linux to run on everything from the Palm Pilot to old 286's, don't you think that Linus would be at least slightly interested in developing a Linux kernel for this thing...

    So my question is: does he have a Linux kernel? Can he develop code for it and keep it secret from the rest of the community? Is 2.4 going to support this chip?

    LL
  • According to netcraft, transmeta runs apache and linux (Who would have guessed ;-) ). But it runs apache 1.1.1! Isn't that risking it a bit, even if there is nothing to hide on that box?
  • The Gilligan's Island theme keeps going thru my head: "Like Robinson Crusoe / as primitive as can be.. / as primitive as can be.. "

    Not that I'm dissing Transmeta here (hey, maybe they're using a very simple "primitive" instruction set).

    Hopefully, their sales force will provide a nice "three-ee hour tour/ a three-ee hour tour"..
    -----

  • Good point. It's pretty much all we've got to go on for the mo' though, so until more emerges...
  • OK then... Windows doesn't really have self-modifying code in it.

    Windows generates code and executes it, which is subtly different to having self modifying code. This is done for things like Blt routines... the code for a particular ROP is generated on the fly. I believe the SGI software OpenGL implementation for Win32 generates scanline routines in the same way, taking into account the relevant renderstates.

    And, of course, the context of this discussion is the TransMeta processor, which is said to generate native code as part of its emulation strategy.

    Do you have a problem with self-modifying code?
  • Well, not Hype persay; or hype for hype's sake... but the fact that TransMeta has so MUCH Hype for not marketing any itself, is rare... that's what gets me excited about it. They still have to prove themselves of course with an actual product (or do they? Economics students should respond here).

    And the side effects of the silent publicity are cool too. Any time Linux can get in the media in a non-negative (if not positive) way, I'm probably going to support it. TransMeta has been doing that, so yay!
  • But you've left one of the most interesting possible instruction sets that the transmeta chip could support.... JAVA BYTECODES!

    Imagine Java running natively... ease of development and native speed.
  • Actually their page was updated with that weeks if not over a month ago...and yes the, err, comments are still there in the HTML.
  • and only runs on Apple hardware
  • Ziggy! I've found you at last!
  • Remember that Microsoft's Paul Allen is one of the major investors in Transmeta. And I definitely remember seeing somewhere that the Transmeta CPU is supposed to run Windows NT as its primary target.

    My guess would be that they are writing a new HAL and recompiling the performance-critical parts to native code. You can afford to run MSWord in emulation. Even your soundcard driver won't mind too much.

    Now all that remains is to get a few CPU-hogging killer apps like Lightwave or Adobe Premiere to recompile to Transmeta native code. A really fast JVM would make a Transmeta box an attractive middleware application server, too.

    But I am definitely looking forward to a linux kernel that can execute both i386 and transmeta executables...
    ----
  • When National Semiconductor [national.com] announced that they were selling the Cyrix line to VIA, they also said that they were looking to sell a good-sized piece of their interest in their South Portland facility [national.com]. I haven't heard anything more on that part of it, so maybe it's still in play.

    BTW, as far as Transmeta's PR budget goes: it's Linus' salary. They get enough coverage out of the fact that they employ him (and give kernel.org the server space), that they can come out with an enormous IPO even if all Linus does is answer his Email all day. (NB: I'm not trying to imply that that's what he actually does... I'm sure he would not be comfortable in an environment where he was wanted only for his celebrity PR value.)
  • Don't forget how many companies are absolutely biting the big one on low-end desktops.
    Hell Packard Bell is out, IBM is out of direct-sales, etc.
    Laptops still offer a nice profit margin.

    Hell, if you're PC Magazine, you can then compare the $1599 iBook against a $2499 IBM ThinkPad and beleive it's fair! :P


    Pope
  • A lot of /.'ers hear "low power" chip and think "laptops." Don't be so limited.

    Obviously, low power chips are good for any battery-powered applications: PDAs, cell phones, devices we haven't thought of yet.

    Low power chips are also important when you have a lot of them. Say, for the sake of argument, the Transmeta chip is very well suited to parallel processing, maybe massively parallel processing. You'd have a lot of such chips in one box. You'd want low power chips, both to reduce power consumption and to ease the cooling requirements. I presume low power chips also generate less EMF.

    For example, along these lines, low power chips are useful in the telecommunications market. I've been associated (loosely) with some hardware that needed to be redesigned to have more fans. One customer was the electric company's third biggest customer in that city (and you've heard of the city and the two bigger customers).
  • That's the catch. According to the patents they have filed, it'd most likely be able to run ANY OS, and hence, any software. There chips would morph x86, etc, calls, to it's own internal architecure, hence, 'emulating' nearly any chip on the market currently..
  • I'm pretty sure it's just rumor. If they were actually close enough to start constructing the chip itself, they'd have a patent for that. Once all the patents are in order I'll start considering the truth of these rumors.
  • That's the catch. According to the patents they have filed, it'd most likely be able to run ANY OS, and hence, any software. There chips would morph x86, etc, calls, to it's own internal architecure, hence, 'emulating' nearly any chip on the market currently..

    Bingo on the emulation. If we wre to explect a native Linux on this platform, we could also expect a large chunk of new, untested code in the kernel; something Linus has been heavily against in the past.

  • HeHe.. I submitted it as a story nearly 3 weeks ago, got rejected soon after... ;-P
  • Indeed. Your point is well-taken.

    On a purely historical note, however, the ARM series was, at its inception, more than comparable with the x86 processors of the day; that'd would've been about 1987, I guess.

    ARM Ltd was spun off to develop the processor and aimed at the embedded market in particular, resulting in the less spectacular mid-range chips such as the ARM6 core; as you note, it took Digital's involvement in the StrongARM project to make another high-end processor.

    Of course, no ARM ever ran the x86 instruction set, which is where (we think) this may differ...


    --
    This comment was brought to you by And Clover.
  • Linus has been saying for quite a while he sees Linux being big in embedded computing systems and appliances. If these rumors about Transmeta releasing a low-power chip are true, could this chip be the killer chip for embedded designs running Linux?
  • Unfortunately, you can't run Java 100% natively because there are no bytecodes for accessing hardware or anything like that. And I don't think having the garbage collection algorithm executed as bytecodes would work very well, either...

  • > Yeah, because we all know notebooks are cheaper > than desktops... :P

    They are at least cheaper than servers or high end workstations for which something like a Xeon or a high end Sparc processor might be reasonable.
  • Forget Java bytecode: I see the return of the LISP machine!
  • ... that they are running a Shoutcast server.

    Go fig!


    Joe

    ChozSun [e-mail] [mailto]
  • Gateway has dropped all AMD systems, true. But this, when taken with other recent Gateway actions, is more likely to scare Gatway shareholders than non-Intel fans, as it's a fairly clear indication of troubles within Gateway.

    The most visible of these moves is that they have dessimated their customer/tech support group. In restricting their products to Intel-only, it's likely they're shipping pretty much the same thing in every box, CPU/motherboard wise, with all motherboards made by Intel. This will definitely cut support costs, simply because support staff needs only know one or two PCB variations.

    It's also a retreat back to what Gateway's confortable with. They were one of the last major companies to start using non-Intel products, and they're one of those with the least internal development around (which they have also cut back recently).

    This isn't to say that fear of Intel, or even Intel themselves, hasn't hurt the Athlon introduction some. But that's not the only explanation; the Super7 introductions were plauged with chip and software issues for some time. Some vendors will want to believe that's not an Athlon problem before they jump on board.

    It's also a standard cost-benefit thing; is Athlon compelling enough to jump now versus later, or as a flagship, or whatever. Maybe? If the Transmeta fulfills its design goal of delivering a 4x cost/power advantage over some comparable state-of-the-art chip, it'll catch on fast. Especially on laptops, which, being fairly well self-contained and made by larger companies, are less succeptable to fear of Intel.

  • A good thing???? Yes, just what I want, my cheap, very modular, and easily upgraded system to because an overpriced, very proprietary, stuck with it until I buy a new one because of no real upgradeability system. Forgive me for not agreeing. Yes, desktop computing costs and notebook costs would be more equal. but closer to the notebook side. That is expensive. Let's look at what you've said.
    smaller: Ok, I don't see the big deal here really, my tower isn't to big, and I *like* room inside to put whatever I want... Be that, 5 pci cards and 4 harddrives, or load it with fans if I desire....
    quieter: Once again, the sound is already negligable on most desktop/towers. The small gain is not, in my opinion worth the cost of swapping to laptop-type technology.
    more power-efficient: For the average home user who runs one system, the power the computer is drawing is not an issue. A business might like this, but I still don't think it is worth it.
    flat-panel ready: Huh? Ok, I think you may mean one of two things. You could mean the whole system will be in one small, flat panel. In this case this is the exact some poit as "smaller" and I've already addressed it. Or, you may mean flat panel monitors will be supported. Well, this is an absurd notion to think we should change the whole computer architechture for a problem that should be addressed in the video card.

    Anyone who wants a very high performance desktop picks and chooses *each* component to get the absolute best in all parts-- soundcard, video card, motherboard, processor, and even the case is scrutinized. If what you suggest were to come to fruition, all of this power is lost. We would be stuck with what dell, gateway, compaq, or whomever threw together in an attempt to get our business. I have never once been impressed with a large company predesigned computer, and I don't want to be forced to have one.
  • The author who wrote Robinson Crusoe also wrote Gulliver's travels, which was a play on the word Gullible, since most people believed that his stories were true. Robinson Crusoe was one of these stories since travel journal stories were popular at the time. Keeping this in mind I have to wonder if this is just another rumor being passed off as truth.
  • It's not that difficult to develop a very confidential manufacturing relationship with a Taiwan based fab (i.e. TSMC, UMC, Mosel-Vitelic, ASE, Amkor, etc) to manufacture this product, particularly within the past twelve months.

    Until recently, when memory prices started gaining ground, these Taiwanese fabs were practically paying companies for their business. I have a client that developed relationships with two fabs in Taiwan to produce their first analog product and the market never had a clue during the entire product development process until the day of the press release.

    As you probably know these fabs went through huge capital spending programs to get leading-edge process technology only to be left hanging with much more capacity than they could utilize. As a result people were able to approach them early on and get long-term manufacturing commitments at reasonable cost.

    Experience tells me that in order to go this route you have to have someone in your organization that has very strong relationships with the Taiwanese, be they transplants from Taiwan or they have a history of working in that environment. Does anyone know if such people exist at Transmeta?

  • Hasn't anyone else wondered if there is a connection between Linus' change of attitude towards minor kernel numbers and his work at Transmeta?
    Compared to the development time of the 2.1 kernel, the following pace is downright breakneck; 2.1 was released in 1996, but 2.2 and 2.3 were released in 1999, and 2.4 might appear by year's end. Would a port to Crusoe (the Transmeta platform) be justification for kernel release 3.0, despite being one port of several? At this pace, 2.9 is surprisingly close.
    Forgive me if this seems naive or just dumb. I am not a kernel developer, and I do not work for Transmeta - ha! I wish I did.
  • by hazydave (96747) on Thursday November 11, 1999 @08:03AM (#1543685)
    From a read of the patents, from other more open work at other companies, it's fairly clear that if Transmeta is really making a CPU (they have tried to deny it without really denying it, I think they are), what they have is a VLIW engine of some kind with lots of hardware support designed to allow it to run not just a popular instruction set (say, x86) as fast as you'd expect a modern CPU to run this, but also emulate the hardware subsystem. So they could release a thing that looks and behaves like a PC, but actually doesn't have either x86 or a real PC architecture living down below. You all know this is possible; x86 emulators have lived on 68K, PPC, SPARC, etc. machines for years. The Transmeta stuff would make it fast, say, 50% or so of the native CPU speed.

    The Linux connection (strongly hinted at by Jim Collas, former President of Gateway's Amiga division, when I spoke to him about the various AmigaNG rumors) is, of course, Linus himself. But the coolness factor is this: here we have a CPU, running emulated instructions as fast as some version of the Pentium II/III, and doing it cheaper and with much less power. Only the first one is new work -- it's easy to find CPUs that run faster than PII/IIIs using less power (PowerPC, for example). The thing is, the x86 code runs at 1/2 or so of the native Transmeta CPU core speed, perhaps based on the limits of their dynamic recompiler, the loss of abstraction in binary code, etc. What about native code?

    The modern trick in all this, same thing Sun's doing with MAJC, is to make the idea of a VLIW processor legit by never tying a system to native binaries. You run x86 or Java Byte Code or whatever through a translator, and when the machine architecture, and thus instruction set, change, you build a new translator, everything's hunky-dorey code wise, and you get to forget entirely about hardware legacy. But Linux and other open source stuff doesn't have a binary legacy problem, period. So there's no problem in coding Linux native, and if you did this, Linux would run twice as fast, relatively, as any closed source OS on this platform. The need for Linus on this would be building a mixed code manager for Linux, so that the kernel could cooperate actively with the dynamic translator and run your choice of x86, native, or other code modules (full-speed Java, etc).

    This is the last layer of HW abstration. Some day, all CPUs may be built this way.

  • Two comments.

    smaller is better

    The other reason why smaller isn't really better is that if you put your components in a huge enclosure (like the Supermicro SC801-A ) you've got oodles of room for fans, elaborate vapor-phase cooling devices, PCI cards, drives and alien technology.

    more power efficient is better

    I agree with the original poster. I didn't realize how much it was costing me to leave The Beast powered up at night. Answer: Enough to make me turn it off before I go to bed.

    The other reason why it's good to be power efficient is that it's easier to overclock a processor with low-voltage requirements than a processor with high-voltage requirements. (At least I think that's right...)
  • Absolute nonsense. These are two diferent authors - Jonathan Swift and Daniel Defoe.

    OK, just notice you're from Canada. You're forgiven, my son. Being from Canada is a good excuse ...
  • Bzzt! Wrong, thank you for playing, but Daniel Defoe wrote Robinson Crusoe, while Jonathan Swift wrote Gulliver's Travels.
  • OK, let me have a shot at this....

    Smaller: My case is ~2.5 feet tall, and is mostly empty (but useful) space.

    Quieter: 11 fans and 2 hds do not a quite computer make.

    blah blah blah....

    Anyway, you're missing the point. In fact, you're so far from the point I had to hike out to send this.

    PC's were in the same position a few years ago that you claim notebooks are in now. They were expensive, proprietary, difficult or impossible to upgrade, etc. Guess what? Things change. Now their (fairly) cheap, (mostly) open and (relatively) easy to upgrade. The same will come to pass in notebook tech as it migrates to the desktop. People will demand that their waffle sized computer have slots for their shiny new NV11, and some company will comply. Then they'll want to be able to upgrade processors, and another company will comply. And so on. If you're still looking for over-priced lock-ins, try the workstation market.

    Anyway, you can't stop it; the lower-end of the market demands it. The iMac is a laptop (Powerbook mobo's, etc) in a big colorful case, and people love it. Some are starting to complain about the limited upgrade path (i.e., none), and hopefully Apple will do something about that. The reason they're doing so well is that most people don't care about proprietariness; they want a computer that works, without having to know anything about it. That is what this tech will give them (though it's not really at any advantage over the current systems in this). However, this does not mean that we will *all* be using iMac's (see above paragraph).

    So, in five years, you'll see basically what you do now. Compaq, Gateway, Apple, etc. selling all-in-one miniPC's for the average consumer, while those of use with more refined tastes will order our Abit mobo's and Transmeta processors and slap together monstrously overpowered notebooks. I fail to see the downside. (Esp. if I can get dual processors, firewire and an NV1x in something I can take to LANparties in a backpack. ;-)
  • ...through babelfish produces some humerous results: "Linux is an as far as possible stink-normal Unix."

    feed this url http://www.ix.de/newsticker/forum/go.shtml?read=1& g=19991110as000&msg=95
    through babelfish [altavista.com]
  • Here is some info on Trnasmeta's fab arrangements, though this is a bit dated:

    http://slashdot.org/articles/98 /09/26/1956233.shtml [slashdot.org]

    http://www.redherring.com/mag/issu e60/intel.html [redherring.com]
  • Heres a random thought...

    Computer languages are really just a useful convention providing an abstraction layer that a compiler translates to machine instructions. In some applications, I wonder if its possible to eliminate an abstraction layer with the transmeta architecture and "compile" to a customized instruction set instead of a given machine instruction set (virtual or otherwise)

    e.g. Suppose I wanted to create a router. Instead of writing source code that is translated to machine code which is translated by the transmeta chip. Could I write an instruction set that defines "router" instructions?

    Practically speaking, current compiler technology / language specification isn't setup to handle this paradigm, at least with my limited understanding. But why not have a X-window instruction set and a html instruction set? In a way, the protocol becomes the insruction set.

    Of course, if you do this, I think you'd might be tying yourself to machine architecture again -- hmm, maybe not...

    Well, what do people think -- flame away.



  • Imagine yourself shipwrecked on an uninhabited semi-paradise island on the Pacific knowing there were similarly shipwrecked wickedly wonderful geekgrrls on a neighbouring island but there were no physical means of getting there... Since you only managed to rescue the only items of any importance to you - a notebook and a solar panel - you'd want to waste no time setting up a chat connection to that real-paradise island.

    Before setting out on that perilous Pacific cruise you had read on Slashdot about someone called Gilligan Bates (also known as Billigan) - a former billionaire who had found himself stranded on that very same island a few months earlier. Like you, he had also managed to hang on to his beloved laptop. But his was a new Pentium model running Windows Y2K. The bloated OS ran - when it wasn't sporting a blue screen - barely half an hour per recharge; and it always took more than half a day to get the battery full anyway. Worse, since everything from MS-Office down had been integrated into one humongous bundle most of that productive half hour got spent on operating the HD, and running MS-Diagnostics after bluescreens. Not that Billigan was even interested in the grrls across the straits - he was too small and limb to have found the courage to even wave at them - but he was desperate for a Word from the outside world on the situation of his declining worth. Well, the value of his declining stock fortune anyway. But none of the proprietary protocols newly-integrated into his software had been adopted by the near-by islanders so he couldn't even communicate with anyone. Before banging his head against a trunk of a palm tree til the final bluescreen got to _him_, he had carved one final message on lid of his laptop. It said: "These bloody savages, so obviously happy and care-free, must have been using free software and open protocols that weren't included in my all proprietary preload setup. God, oh god, I miss the preloading civilization! And cause I'm getting ever smaller and limber by the hour, I can't even reach the bananas any longer. The end is nigh..."

    Thinking about the poor Billigan makes your grin ever wider. You plug in the solar panel, boot a few days old revision of MandHat and in half hour you've created a wireless LAN and started hacking your way through your favourite project while chatting with the increasingly interested geekgrrls from the other island. You look at the power gauge and, after an hour it tells you're at 90% and holding steady. As you start recompiling the kernel and one of the geekgrrls has started to become intimate online you laugh aloud. Not because the grrl told you that there aren't any bananas on their island, but because you know it really paid off to get that cheapo laptop with a curiously named CPU - Crusoe.
  • This is only for the true fiend among us, but it just might satisfy your hunger. You neglected to include your email address, but I promise I'll send you some personally next time, okay?

    -b

APL is a write-only language. I can write programs in APL, but I can't read any of them. -- Roy Keir

Working...