Stories
Slash Boxes
Comments
Search

News for nerds, stuff that matters

Slashdot Log In

Log In

Create Account  |  Retrieve Password

T-Engine Enables Ubiquitous Computing

Posted by timothy on Tue Jun 28, 2005 07:13 AM
from the does-not-stand-for-turbo dept.
Technology Hardware
An anonymous reader writes "A Japanese-government sponsored research consortium that include five chip makers and 17 other Japanese high-tech firms, has announced that the T-Engine, a ubiquitous computing platform is ready for prime time. The engine is featured in a IEEE Computer Society article (PDF) and discussed more on Windley's Technometria. The system is based on the iTron real-time OS and includes multiple boards for different applications."
+ -
story
This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.

T-Engine Enables Ubiquitous Computing 50 Comments More /

 Full
 Abbreviated
 Hidden
More
Loading... please wait.

  • Recycling Great Technology! (Score:3, Funny)

    by creimer (824291) on Tuesday June 28 2005, @07:20AM (#12930222) Homepage
    It's the return of the Model T engine [modelt.ca]!

  • um? (Score:4, Interesting)

    by tomstdenis (446163) <tomstdenisNO@SPAMgmail.com> on Tuesday June 28 2005, @07:21AM (#12930225) Homepage
    So it's an embedded computing platform?

    Ok, if I was a company [like say Motorola] and wanted to make some sort of portable device [say a cellphone perhaps?] I'd take a READILY AVAILABLE ARM core and drop the sucker into my design.

    What really are they offering there other than perhaps a "standard" [though amongst ARM cores there are standards and they use well documented interfaces, etc...]

    Is this just better because it's newer or?

    Tom

    • Re:um? (Score:5, Interesting)

      by torpor (458) <jayv@sy[ ].net ['nth' in gap]> on Tuesday June 28 2005, @07:31AM (#12930278) Homepage Journal
      Is this just better because it's newer or?

      its better because its older, not newer. i-tron, and its descendants, are the results of 30 years of computer-science research on ways to get collaborative computing systems into operation.

      the ARM core scenario is derived from the desire to have common platforms being used by multiple, different vendors. it was i-tron which prompted the industry to adopt ARM and similar initiatives, and it is the i-tron philosophy of common cores and platforms which have allowed ARM to flourish in the embedded world in the first place.

      JAVA was an 'Americanization' of the i-tron initiative, only it hasn't had as much success in the embedded world because of the lack of hardware adaptation that i-tron has prompted; at least, with the big Asian chip foundries, anyway, this is true, and we all know that the embedded space is dominated by the Asians ...

      this latest instatnce of the T-Engine is the realization of some very old, honored traditions in the embedded space. the dream of having your microwave oven use your cell phone for that little extra calculation power it needs to get your meringue fluffed right is just one step closer ..
      • I'd really be impressed when thay get a microwave oven to fluff meringue. All this time I've been using a whisk! ;)

        Still, I think it's interesting that things like cell phones are as standard as they are. That old joke about standards beign so great because there are so many to choose from... tends to ring true.

        Of course, the other joke is how standards codify obsolescence. What do you think this will do to hte ability to upgrade later? (Especially things people don't replace every year, like... microwave
      • If my microwave oven is using my cell phone for a little extra power then my cell phone is pumping out god damn too much radio waves.

        -
      • we all know that the embedded space is dominated by the Asians

        The first thing that springs to mind is that they're the only ones small enough to fit!

        Well, I was amused.

    • Re:um? (Score:3, Informative)

      by RegularFry (137639)
      So it's an embedded computing platform?
      Nope, it's a standard for implementing a family of embedded computing platforms.

      This is better because it gives you a highly configurable operating system, with all that that implies, on top of your READILY AVAILABLE ARM core. Or an i486. Or any of a small bucketload of other SOC configurations. Anywhere from 8-bit to 32-bit. You get scheduling. You get (soon, I think - can't remember if it's actually in yet) a TCP/IP stack. You get memory management. And mor
    • Except that Motorola has it's own cores based on the PPC. Like everything else it is a trade off. Everyone and their dog can use an ARM core. You get a trade off between time to market and how easy someone can match your features.

  • wtf mate? (Score:5, Insightful)

    by inkdesign (7389) on Tuesday June 28 2005, @07:22AM (#12930236)
    "With any luck your 2007 Toyota Camry and your Mitsubishi food processor will be exchanging recipes in the not too distant future."

    I'd prefer my car stick to driving, thank you.
  • is futile; all your devices will be assimilated!!

  • Embedded Development (Score:3, Insightful)

    by putko (753330) on Tuesday June 28 2005, @07:29AM (#12930270) Homepage Journal
    Standards Conformance usually gets cut when doing embedded development if conformance ruins the cost of components or power the device requires.

    I don't think they will be able to get everyone to hew to the party line; there will be too many economic reasons to deviate.

    Otherwise, sounds neat.
    • It sounds like they have some ideas on not just the hardwrae side but on the software and protocols for interoperability. If so, then maybe they are trying to say "Hey do it our way, and here is some hardware. But if you roll your own hardware, license our software." Could be a good strategy.

    • Re:Embedded Development (Score:4, Interesting)

      by torpor (458) <jayv@sy[ ].net ['nth' in gap]> on Tuesday June 28 2005, @07:36AM (#12930313) Homepage Journal
      I don't think they will be able to get everyone to hew to the party line; there will be too many economic reasons to deviate.

      umm .. take any single Asian-produced cell phone, and at its heart you will find pieces of the i-tron initiative. it has already been proven in this space that there are far, far greater reasons to comply with the party line than to deviate. insta-deviation for the sake of it is anathema to the i-tron initiative; it is this very powerful fact that has resulted in such growth in the Asian core and embedded mfr. market in the first place.

      american electronics/semiconductor giants ridiculed i-tron, and its resulting policies, in the 80's, and Asia has been eating the carcass of former US' manufacturing prowess for lunch. if it weren't for i-tron, philosophically, we wouldn't be buying EUro 5,- MP3 players, made in Taiwan instead of Kansas, at the Aldi checkout lines ..

      if you're a US comp-sci person, and you haven't boned up on i-tron, you've got some history lessons ahead of you. quick, before its too late.
    • Uh, mods? Insightful? Everyone *is* hewing the party line! Well, in Japan, anyway. The whole point of this standard is that it's scalable enough that the component cost can be controlled, and the interoperability gains you get are worth it - as is not having to retrain/retool for a new embedded platform when your design criteria change.

      And yes, it does sound neat.
    • This diagram [sakamura-lab.org] on this page [sakamura-lab.org] shows in general terms how they're addressing this.

  • iTron (Score:2, Insightful)

    by Yjam (893817)
    "TRON, ITRON, and ITRON do not refer to any specific product or products."

    This is from the official TRON website.
    So now... how are they going to sell something based on nothing.

    I too can build and sell the top-notch-most-powerfull stuff ever built.. and I won't be selling it cheap.. oh, of course this would only be theory-reselling.

    I've seen nowhere in TFA that this techno is actually going to be used. Bu anyway, if it's gonna be.. maybe we should all beware of the Attach of the Killer Tomatoes [imdb.com] wh

    • Re:iTron (Score:2, Insightful)

      by torpor (458)
      So now... how are they going to sell something based on nothing.

      what i think you need to do is recognize the difference between the word 'initiative' and 'implementation'.

      lets take this to another context: Free/Open Source Software.

      F/OSS is an initiative. Linux is an implementation.

      Get the point? iTron is an initiative which has borne much, much fruit. Look around you, find a "Made in Asia" component which contains a computing system. Therein, you will spy aspects of the iTron initiative, undernea

        • Re:iTron (Score:2, Informative)

          by torpor (458)
          check the table headed "Table 4: ITRON-specification kernel implementations".. each one of these companies has an iTron kernel implemented somewhere.

          i myself have personally worked with/reverse engineered the Morson Japan kernels, as these are commonly used in high-end/professional digital audio devices, such as the Yamaha A3000/A4000/A500 samplers, digital mixers, etc.

          iTron is out there, but you really have to pry open the box ..

  • hmmmmm... (Score:5, Funny)

    by new death barbie (240326) on Tuesday June 28 2005, @07:36AM (#12930307)

    if there aren't security and privacy implications, you're probably not doing anything very fun.


    I think I used to date this guy...

  • ob. link (Score:3, Informative)

    by pario (675744) on Tuesday June 28 2005, @07:36AM (#12930317)
    As always, the TRON Web [super-nova.co.jp] is the most valuable source of infomation on the TRON project in English.
    You can find some good articles on the T-Engine platform here [super-nova.co.jp].
  • I've been waiting for a system of this calibre to come along for quite a while. To me, the T-Engine specification, along with iTron, is a tool that can revolutionize how we look at our daily life. For some, it means that they may not have to worry about having to leave their PC to make coffee during a Gentoo install. This specification can and will change the way we look at how we view computers in our daily lives. This is the age of the computer, plain and simple. This system will make computers an even more important, and more critical part of our daily lives.

  • Commerical and criminal abuse of this (Score:5, Interesting)

    by G4from128k (686170) on Tuesday June 28 2005, @07:44AM (#12930350)
    When everything is networked, the potential for commercial and criminal abuses becomes that much higher.
    1. A phisher sends a small worm to your stereo, the stereo asks the user to re-input the password on the DRM system, the phisher collects that password and uses it to re-sell all your music (possibly making you lose the right to listen to it). Even at $0.02/song and a 5,000 song collection, the phisher gets $100 per cracked stereo.
    2. Phishers attack one, low-level device hoping that the phished password from that device is also used on other, more important devices. How many people might use the same password on their stereo's DRM system, their refrigerator's automatic reordering system, their car's ignition system, or their bank's online account access?
    3. Spams starts arriving on ALL audio devices -- audio pop-ups ("audiups"?) start intruding on iPODs, VOIP phones, stereos. Worse, the infection could attack anything with a sound chip. Imagine suggestive Viagra ads coming from your pop-up toaster oven.
    4. A sleazy marketer buys access to or plants spyware in your vehicle's navigation system. You start getting pop-ups for oil-lube places or the database of locations of competitors becomes corrupted misleading the driver on their location.
    5. Digital cameras become spam-sending zombies. Anyone who walks within bluetooth range of you suddenly finds an image file on their device that contains an ad for whatever is the latest spam du jour.
    6. ...... I'm sure there are a million other scenarios, but its early and my coffee hasn't sunk in.
    The point: Cool technology, but I wonder if the core OS has needed security layers to prevent exploits like these. I wonder if the systems designers have embedded a strong sense of permissions on processes and interfaces.
  • I've been hearing about the TRON project for something like 20 years now, starting with the TRON microprocessor... the ultimate CISC. We're talking about a processor that has "insert a record into a doubly linked list" as a fundamental instruction. And they're still pushing this super-CISC as an improvement over RISC.

    Has anyone outside MITI actually done an objective comparison of TRON with any contemporary RISC? The examples I've seen are ludicrous... comparisons "proving" that TRON is faster than RISC by comparing individual highly specialised TRON instructions with a straightforward unoptimized translation of the same code to an unspecified RISC processor. They don't even do any common subexpression elimination... who would write code like this?
    MOV @(RDQ_TBL+4,R2*8),Rn
    MOV R1,@(RDQ_TBL+4,R2*8)
    MOVA @(RDQ_TBL,R2*8),@(R1,FOR)
    MOV Rn,@(R1,BACK)
    MOV R1,@(Rn,FOR)
    http://tronweb.super-nova.co.jp/tronvlsicpu.html [super-nova.co.jp]
    • You and the anonymous coward ubergeek who replied to this need to settle this once and for all.

      Benchmark Quake on these for a DEFINATIVE answer as to the superior machine.

      Stupid geek techno babble doesn't impress the ladies as much as a good frag!
    • It's an academic boondoggle that has enough support to keep it alive, but not enough to actually keep it up to date. Implementations of it, like T-engine, feel like circa 1990 era SDKs, particularly if you want to write actual user applications (you know, if you have a job) rather than low level drivers. For the most glaring example, there is no way (in the SDK that I have, at least) for process-based user applications to write to the screen. None. Yes, I'm serious. The one example in the SDK that atte
    • Well, the TRONCHIP subproject is considered to be dead for quite a while, or at least so I read. The last TRONCHIP to be produced was the Gmicro/400 [google.com], which came out in 1994.
      • From what I've heard, its easier to make a risc compiler than a cisc one. You don't have to look for all those weird, special cases and try to optimize them. You don't have to work really hard to understand that the code writer is inserting an element into a doubly-linked list and output the correct instruction for that; you can just output the logical, obvious, unconnected instructions.
        • From what I've heard, its easier to make a risc compiler than a cisc one.

          Actually, it's easier to make a simple CISC compiler that produces OK code on a regular CISC like a 68000 or a PDP-11. For these "4th generation" CISCs and things like Altivec or SSE you don't bother, you leave them for library writers and assembler boffins. The ones that are really hard to code for are the 3.5th generation CISC like the VAX and iAPX432 that have instructions designed to help compiler writers like "set up stack frame
      • The result is that this high-level instruction allows for processing is carried out in roughly one half the time, while at the same time realizing a two-thirds reduction in code size.

        Yeh, that's the case I was talking about. First, that's an awfully CISCy RISC they're moving stuff to, with "Register plus scaled register plus offset indirect" as an addressing mode. Converting it to a sane RISC would use two more registers, add two arithmetic instructions at the beginning but eliminate all the complex addre
      • Oh, yeh, man, the RISC code in the second example isn't using the (n&-n) trick from HACKMEM, which is four times faster on a typical processor and would thus reduce the RISC time to ... 61 cycles, about the same as the CISC (again, this is unsurprising, since the CISC code is interpreted by microcode that's similar to the RISC code).
  • Although it will be tough to get the T-Engine out of the Hive...

  • Nobody cares about TRON outside of Japan (Score:4, Informative)

    by heli_flyer (614850) on Tuesday June 28 2005, @08:10AM (#12930488)
    I used to work at Hitachi America so I know about TRON/iTRON/microITRON/etc. It's this weird, baroque API that's non-POSIX and not standard C library compatible. Kind of like an extreme example of Not-Invented-Here syndrome. For some reason Hitachi Japan thought everyone wanted it without realizing that nobody outside Japan cared about it. We kinda tried to humor them..."sure, we will distribute microITRON if customers ask for it".
    • It's this weird, baroque API that's non-POSIX and not standard C library compatible.

      That's not necessarily bad. Particularly in embedded systems, where most of the resources POSIX and stdio manage (files, virtual address spaces, etc) don't exist. Real-time programming deals with an environment more like the internals of the UNIX kernel than userspace, which is why things like microkernels are so attractive even if they don't directly make implementing a POSIX environment any easier... what they do is crea

  • TRON is in decline (Score:3, Interesting)

    by kahei (466208) on Tuesday June 28 2005, @09:16AM (#12931065) Homepage

    I can remember when the cool parts of TRON were still going -- the bTRON desktop, which had its own hyper-ergonomic keyboard with about eight shift keys, and the TRON charset which included Unicode and Mojikyou, so you could actually have a fair shot at representing old Asian texts on a computer without using image files for the characters.

    Now, only the embedded iTRON part of the project is left. And it's been very successful -- I think at one point it was the most-used OS in the world, although to someone from a Linux/standard C background it seems kind of weird. But there's seriously no news here -- T-Engine is the attempt of the TRON project to remain relevant now that hardware can run embedded Linux or Windows or Symbian and what have you, and it's too little too late.

    TRON rocked once, and for industrial robot arm controllers and what have you maybe it still does, but it's never going to break into the IT world now.

      • Re:Unicode? (Score:4, Interesting)

        by kahei (466208) on Tuesday June 28 2005, @10:22AM (#12931767) Homepage

        Ken Sakamura's a bit of a nutcase -- he always had this idea (common in Japan at the time, but wrong) that Unicode was some kind of conspiracy to take away Japanese identity and make everyone use a sinister Sino-American character set. So most Japanese computing initiatives have tended to avoid Unicode, and TRON insisted on seeing Unicode as just one charset among many, all mapped into a 'meta character set' space.

        Thus when you say 'A' in TRON, you have to specify whether it's a Unicode 'A', a Mojikyou 'A', or some other 'A'. I am simplifying a bit.

        In practise, ironically, everyone uses Shift-JIS, which really IS a sinister American conspiracy :)

        Sakamura used to have a web page containing the most extraordinary rant about Unicode, with A LOT of factual errors, which was quite interesting for those wanting to see how certain very reactionary parts of the Japanese business community think. It wasn't exactly a good advert for TRON, though!


          • I used to say just what you are saying. In fact, I used to go round going "But don't you see, character X and character Y are seen as TWO SEPERATE CHARACTERS! You can't force people to use one!". And of course it's true that Han unification is a difficult job that can never be done perfectly.

            But, it's a difficult job that Chinese, Taiwanese and Korean representatives were all able to see the need and true meaning of, and co-operate to do it well. The Japanese input into the discussion was clouded by th


              • I don't think an academic has a right to be wrong. They have a responsibility to be right, especially on basic facts.

                Come to think of it, in Sakamura's case I think that might be the problem -- he might be fine as an academic. But the move into politics / ideology was not a wise one and it prevented him from getting things academically right as well.

                I think the big mistake made by Sakamura and others is to see it as a case of globalism versus Japan. It's not; it's a case of getting JIS right. But man
  • You may be more interested about a related project, ubiquitious id also by Ken Sakamura, at the uID Center [uidcenter.org].
  • I've seen the various TRON project [wikipedia.org] technologies / OS' mentioned a few times around here ... Apparently there's even a commercial closed-source desktop OS [personal-media.co.jp] based on it ...

    Since we've got a topic for Be (which seems to be sorely lacking in Haiku / Zeta updates) , I think that this OS might merit one...

    (Or perhaps an "alternative OS" category?)
    • Hmmm. BeOS seems to have an interesting file system and GUI, but the OS underneath is a bit language-specific and hoggy for my taste. How about implementing BeFS and the BeOS GUI on top of TRON? You could call it "Beatron" and finally have an OS that's got enough pop-culture references to achieve a kind of pop-Zen perfection and eliminate Microsoft.
    • The article states that MontaVista has ported its real-time version of Linux to run on the T-Engine. Additionally, Windows CE and Java have also been ported. However, running these systems on the T-Engine is known as running 'guest operating systems', which is a fancy word for 'translated operating sytems'.

All trademarks and copyrights on this page are owned by their respective owners. Comments are owned by the Poster. The Rest © 1997-2009 Geeknet, Inc.