Forgot your password?
typodupeerror
Intel Hardware Technology

Is the x86 Ready for Consumer Appliances? 105

Posted by Cliff
from the is-intel-the-way-to-go dept.
rckymntrider asks: "By now, it's pretty obvious that the movers and shakers of the PC industry are shifting their attention to consumer electronics. Consumers today demand capabilities from their set-top boxes that PCs already deliver (examples: HDTV and gaming). They just don't want a bulky, hot and noisy PC next to their beautiful new plasma TV. Intel, for instance, announced several initiatives for bringing their technology to the media/home automation front, including the establishment of a $200M fund for companies in that arena (small change if you ask me). As a small manufacturer of media-centric devices (I will not name the company and product -- this is not a plug), I have become very frustrated at the availability of hardware for 'consumer' type of applications. ATX? Micro ATX? Too big. Eighty watt CPUs? You're kidding me! Mini ITX? Better but not powerful enough and *way too expensive*. Besides, every new piece of hardware that comes out is practically designed for Windows, and we all know that this is not the operating system that will drive consumer appliances, right? So to sum it up, do you think that the traditional x86 architecture, even with the advent of PCIX and the likes, is suitable for consumer anything? What other platforms do you see on the horizon that could still offer things like High Definition video capability and not double as mini-heaters? Have you ever heard (or envisioned) of a platform designed for powerful but still cost-effective consumer appliances? VIA tried with their EPIA platform but - in my opinion - they failed. Do you think Intel will do it? If not, then who?"
This discussion has been archived. No new comments can be posted.

Is the x86 Ready for Consumer Appliances?

Comments Filter:
  • by ObviousGuy (578567) <ObviousGuy@hotmail.com> on Wednesday February 18, 2004 @11:34PM (#8322911) Homepage Journal
    They sold off their x86 Geode platform to AMD a year or so back.

    The Geode is in plenty of consumer devices, if you care to tear them open to take a look.
    • Yeah, right. Geode has clockspeeds in the 200-300MHz range. Even a VIA C3 at any clockspeed can murder a Geode.
      • by Anonymous Coward
        What do you need all that speed for, though?

        The real speed that's necessary is in the graphics chip and keeping the board low cost like the Geode allows you to splurge in the design on a more powerful video chip.
        • Mainly for transcoding video streams on the fly. a 1 ghz celeron should buy you the ability to transcode one stream to disk while watching the other. The graphics chipset will not do this. I suspect that video transcoders will come out to accelerate this at some point but I know of none at the moment.
          • Yes, but this is an embedded device. YOu don't transcode videoon the fly with one of those. You do it in hardware with specialized DSPs- the cost is HUGLY cheaper with hardware decoding. If you need a GHZ processor for an embedded device, you're not taking advantage of having a hardware platform.
            • Tivo did quite well by doing a first iteration
              on an easily accessible platform, and a second
              iteration with more sophisticated (expensive)
              hardware design.
              • Tivo was in a race to be first to market. Being on the market was more important than the money hit of the moree expensive components. This is a rare situation. Noone addressing this kind of product now is. You'll be competing against entrenched solutions. You need to compete on price, which means an x86 solution is out.
      • Anyone that argues that the EPIA platform failed due to its hardware hasn't looked into it much. The EPIA failed due to the worst freakin drivers you've ever had the misfortune of using. Rarely have they had all the features working at one time... actually I think it was NEVER that they've had it all working. Most of the features don't even work at the BIOS level. If they had gotten that right, they'd have done much better.
  • Just a Question... (Score:4, Insightful)

    by Ieshan (409693) <ieshan @ g m ail.com> on Wednesday February 18, 2004 @11:36PM (#8322932) Homepage Journal
    Besides, every new piece of hardware that comes out is practically designed for Windows, and we all know that this is not the operating system that will drive consumer appliances, right?

    Well, that's a good question. Windows is sorta big and bulky, but it runs on an awful lot of things. I mean, think about how versatile the code really is, even if it does crash. Take that requirement out of the picture - that the OS has to run on Nteen thousand different hardware configurations, just one, your superblender - and it might not be the worst choice one could make.

    But then, I might be completely uninformed. It's just conjecture.
    • by Feztaa (633745) on Wednesday February 18, 2004 @11:41PM (#8322973) Homepage
      I mean, think about how versatile the code really is, even if it does crash. Take that requirement out of the picture - that the OS has to run on Nteen thousand different hardware configurations, just one, your superblender - and it might not be the worst choice one could make.

      I dunno, that sounds a lot like the old argument "windows is crashy because it supports so much hardware, MacOS is stable because the hardware is tightly controlled" -- then linux came along and provided much stability, and greater hardware support than windows (more processors than just x86, anyway).
      • I wonder what kind of fantasy world you live in where you can say that Linux has greater hardware support than Windows. Every single box in my house has a piece of hardware that has poor or non-existent support in Linux. Yes, I can run Linux on sparc, powerpc, x86, and IA-64, but it doesn't support my wireless card on any of them! And why would supporting more hardware make Windows "crashy"? That's not even remotely logical, unless you argue that Windows has to support hardware that is somehow inherently un
        • by Feztaa (633745)
          Did I say that windows is inherently crashy because of it's hardware support? No, I said that it was a common argument, put forth by other people.

          Remember the early 90's? People who would argue about the relative merits of Windows/Macintosh would say that MacOS is more stable because of Apple's tight control of the hardware, while windows lost stability because it needed to support such a wide range of configurations. It was like common knowledge, or something.
        • I certainly wouldn't agree that Linux has "greater" hardware support than Windows. But, with the smaller set of devices it supports, it has infinitely better hardware support (for certain pieces of hardware, that is).

          This is what I love about Linux hardware support:

          • USB hotplugging -- PnP in Windows 98, and even Windows 2000 is a joke and never worked properly for me. I think Linux is a breeze compared to those. As long as I've compiled the proper kernel module, I plug it in-- zip, there it goes!
          • All the
          • by Grab (126025)
            Yeah, PnP was lousy in Win98. But how good was Linux at PnP 6 years ago? WinXP is just fine on USB, and that's what would be used. I suppose USB might be used on consumer devices to offload recorded TV programs or music files onto a USB stick, maybe.

            Re driver inclusion, this is no argument at all. This is a consumer electrical device, remember? Everything included in one box. Any drivers needed would be preloaded.

            The main argument for Linux would be efficiency. Consumer electronics needs to be chea
            • Yeah, PnP was lousy in Win98. But how good was Linux at PnP 6 years ago? WinXP is just fine on USB, and that's what would be used.

              sorry to disagree with you but usb pnp support is definately NOT fine! I know of at least 3 cases where installing a scanner is a major problem on win2k (not xp, wont touch that ;) ), if you do anything thats not described in the manual in the exact order (install software, plug in).

              once you plug in the scanner BEFORE you installed the software (what you usually do with usb dev
          • Yeah, how many people know what the correct kernel is and how the hell to compile it?
            • well, not many, I guess. I only learned about kernel compilation a year or two ago, using Mandrake Linux (do yourself a favour and don't get Mandrake if you need to recompile your kernel -- the kernel sources are nigh impossible to find). But from a very very subjective point of view (which was what my original post was intended as), Linux works just peachy for me in a way that Windows doesn't.
        • by Ieshan (409693) <ieshan @ g m ail.com> on Thursday February 19, 2004 @03:57AM (#8324367) Homepage Journal
          Of course it's remotely logical.

          Windows gets accused of bugs it's often not responsible for. Because some shoddy developer never realized that people might be running on system configuration X instead of system configuration A, when his application crashes, people often blame Windows hardware support.

          I'm not advocating for Windows, I'm just saying it's got decent hardware support and that when writing an application for one specific set of hardware it's made to appear infinately better because the consumer can't screw with the default application environment.

          Most of the problems aren't with unstable *hardware*, it's unstable and ugly Windows Malware.

          • Windows gets accused of bugs it's often not responsible for.

            OK, I'll give you that, but the problem is that it doesn't ever fail gracefully. If a 3rd party program sends a hardware call that causes a problem the whole friggin system stops cold and you have to turn it off! You can't blame Malware for that. It is no ones fault but Microsofts that their system can't handle errors. Thier .NET languages are moving in the right direction with the Try, Catch, control structure but it will be a while before that h

        • Hardware is never "inherently unstable".
          It is the combination of hardware and software
          which may be stable or unstable. As Marc
          Andreeson observed, Windows is just a badly
          debugged collection of device drivers. As
          such, it is unstable.
    • by cypherz (155664)
      >it does run on an awful lot of things.
      Winders doesn't really run on very many platforms does it? I mean there's the versions for the PocketPC, and there's x86 (Is there still an Alpha version?), but what else really? BSD and Linux of course have been ported to everything under the sun.
    • by torpor (458)
      Windows is sorta big and bulky, but it runs on an awful lot of things

      WTF. Last I checked, Windows only ran on x86 hardware. That's not really 'an awful lot of things', its just pure x86. Oh sure, it used to run on a few other 'exotic' processors, yeah. The last time I really cared about Windows was when I could run it on a dual-CPU MIPS box, and that was years ago. Ain't so no mo' ...

      Linux, on the other hand, you can *DEFINITELY* say that it runs on an awful ... and really ... that list grows daily .
      • That's not really 'an awful lot of things', its just pure x86

        I tried reading it in the same context as you and thought the same thing. He must have meant an awful lot of computers -- not different types of architecture.

    • I mean, think about how versatile the code really is, even if it does crash. Take that requirement out of the picture - that the OS has to run on Nteen thousand different hardware configurations, just one, your superblender - and it might not be the worst choice one could make.

      Take a look at Windows CE. It supports very little hardware, is stripped down to bare bones, and it's still an unstable little bugger...
  • Of course! (Score:3, Insightful)

    by 2nd Post! (213333) <(gundbear) (at) (pacbell.net)> on Wednesday February 18, 2004 @11:47PM (#8323018) Homepage
    They're already here! There's an AMD 486 inside my graphite Apple Airport Base Station : )

    Of course I'm joking, but what the poster really wants to know... is *Windows* ready for consumer appliances?

    I think not, myself ^^
  • by Marillion (33728) <ericbardes@gmREDHATail.com minus distro> on Wednesday February 18, 2004 @11:50PM (#8323048)
    Well okay, it has market share.

    I've written programs on VAX, Dec Alpha, RS6000, PowerPC, PA-RISC, 6502, Sparc, Ultra Sparc, 68000 and every version of x86 since the original PC. Really, don't get hung up about x86. In the grand scheme, it's just another CPU. Unit cost, energy cost in a million unit device will more than out weigh nearly anything that might make you choose x86.

    • by Paul d'Aoust (679461) on Thursday February 19, 2004 @02:37AM (#8324076)

      "if you build it, he will come." x86 the architecture itself may be ready for the set-top box and the digital streaming stereo thingy, and good software foundations are out there (for appliances, think QNX Neutrino, embedded Linux, PalmOS 6, and so forth), and there are low power chips like Geodes and C3s. (I've even heard that people are experimenting with Transmeta's processors for appliances.) So the architecture is ready and the software is ready. But there aren't a lot of people out there who seem really interested in making good hardware (mainly motherboards) to fit this niche, and I think that's mostly what the author is frustrated with. (Mind you, VIA is going in the right direction with their new nano-ITX board, if only they'd drop their price a whole heap.)

    • Wrong. The x86 series of processors has long been used in consumer products. Why? To paraphrase Intel Inside (A documentary written on the early days of Intel), the world is overflowing with engineers capable of programming and developing using the x86 platform. Not only that, but there is a proliferation of tools available for developing with x86.
  • What's the question? (Score:4, Interesting)

    by Kris_J (10111) * on Wednesday February 18, 2004 @11:53PM (#8323080) Journal
    If you're asking if I believe that we're likely to see an off-the-shelf PC motherboard in every new fridge, then the answer is no. If you're asking if there are entertainment options for small quiet x86-based motherboards, then I point you to Mini-ITX.com [mini-itx.com] (depsite the cost), particularly to "Lippert's Passively Cooled Thunderbird".

    If you're asking if modern consumer OSes based on the x86 range are bullet-proof and idiot-proof enough to power a device as easy to use as an answering machine or VCR, then I'd have to say no, these are still hobbyist devices.

    • I think the argument that the cost of mini-itx boards being too expensive is totally off base. The VIA mini-itx boards have a 1ghz CPU, sweet built in audio, network card, hardware mpeg2 decoder, and a graphics card with TVOUT and they cost around $120-$160 new.

      Now, total up the cost of a full size mainboard ($50) + cpu ($50) + graphics card with tvout ($50 and you get about the same price, with a little more performance. In fact, the only plausible argument here, IMHO, would be that the performance is l
  • In some ways... (Score:2, Insightful)

    by Bobdoer (727516)
    It's great. For larger things that you want to run Windows on (refrigerators or other such devices), the size is no issue. For media things, people expect VCR sized devices, and a VCR could easily have a Micro ATX board inside.
    As for smaller devices, I for one do not want Windows or *nix to run on my toaster; it does not need a general purpose x86 chip inside, as it just does one thing. I want these little ones to do what they are asked of promptly and easily.
    Not to mention, a Widows/*nix enabled popcorn
  • MiniITX (Score:3, Interesting)

    by Hungus (585181) on Wednesday February 18, 2004 @11:57PM (#8323110) Journal
    I don't know what you are trying to do but I have an epia m series 1Ghz windows box that i use for a media PC. Even running XP I have yet to find a codec it can't decode realtime. It is uber responsive when doing software decoding? No but it runs my PVR even while Watching video from disk. The only thing I had to work around was disk access and adding a second and third drive fixed that. 120GB storage drive, 6GB swap drive and 20GB OS/ Software drive and everything runs fine.
    • Even running XP I have yet to find a codec it can't decode realtime.

      Any processor can process any codec... The question is, how big can the content be.

      My 233MHz system can decode lots of MPEG2 videos, but it can't handle 1080i (HDTV) videos worth a damn. Size matters, a lot.

      • OK better phrased would be I have not yet found a file that I cannot decode using a software codec or the mobo intergrated systems in real time. HOw's that? and yes evilviper that includes HDTV signals from my Sat eq.
        • The one reason you can decode HDTV is because of the hardware MPEG2 decoder on the video card, otherwise that system wouldn't have a prayer.

          Want a few videos your computer won't be able to handle? Just go here: http://www.microsoft.com/windows/windowsmedia/cont ent_provider/film/ContentShowcase.aspx

          (The files are '.exe', but are just self-extracting zip files, you can use any unzip program)
      • Re:MiniITX (Score:3, Informative)

        by AKnightCowboy (608632)
        My 233MHz system can decode lots of MPEG2 videos, but it can't handle 1080i (HDTV) videos worth a damn. Size matters, a lot.

        Not really since the Via EPIA boards have hardware mpeg-2 decoding built in. My MythTV box easily handles 720x480 mpeg2 streams with hardware decoding using around 10%-15% of the 1GHz CPU for mythfrontend.. compare that against 90% CPU utilization for software mpeg2 decoding. It could probably handle HDTV if I cared, but I don't.

        I don't know what this fascination is with incredibl

        • I don't know what this fascination is with incredibly high resolution television broadcasts.

          I can't speak for everyone else, but I am mainly interested simply because it finally gets us away from analog, and to fully digital video... Same reason I like DVDs. That said, I don't expect to be able to recieve HDTV for a LONG time. No TV stations are within 50 miles of me (max range of HDTV) and my cable company is only offering HDTV broadcasts to people willing to pay a big premium (and sacrifice their priv

    • I think you mean _because_ you're running XP it can decode in realtime. There are Linux drivers which enable the hardware decoder, but they're closed-source and buggy. My Epia M Nehemiah is a slow and quiet workstation (though quite good at that) rather than a DVD watcher, because it can't play them without installing Windows. Official drivers: http://www.viaarena.com/?PageID=325 Unofficial drivers: http://www.ivor.it/cle266/

      • Umm no I didn't mean _because_ I mean that XP is a bloated Hog ( no offense intended) IF you are planning on working in the embedded market you can write a driver for your board's chipsets.
    • at full resolution?

      Last year I though about making my own media box (via-based) and so I googled a lot looking for performance issues.

      The only place I found (where someone did talk about via processors decoding divx) was somewhere in the newsgroups (cant find the link). And they said it didnt handle it at all.

      Please dont tell me it works fine... Not now that i have an Xbox.
  • SiS 550 (Score:3, Interesting)

    by mnmn (145599) on Wednesday February 18, 2004 @11:59PM (#8323132) Homepage
    The SiS550 is an x86 SoC like the geode, elan. Theyre getting faster and smaller and more ready for embedded markets. 256MB flashes are cheap, and can carry full distros of WinCE, QNX Linux, BSD or anything you want.

    Theyre still a far cry from ARM cores though, and I'd only use x86 where win32 binary compatibility is absolutely required. Things can and do get complex on x86 SoCs, and ARM cores will give you that 'simple and efficient' feel nothing else will.
    • Re:SiS 550 (Score:2, Interesting)

      by torpor (458)
      I also use ARM cores and have a lot of x86 experience, and I have to say here that the ARM wins hands-down for 'fun' factor, ease-of-use, and sheer bang for the buck.

      On the other hand, its much more fun to debug x86 code "out of the box" ... with the ARM, you have to do a bit more work setting up a remote debugging environment, or depend on your board vendor for all the tools, which can be a serious drag at times.

      My only ARM wish is that I had a beefy ARM-based system to use as my *main* machine in develo
  • I fail to see... (Score:5, Informative)

    by Stigmata669 (517894) on Thursday February 19, 2004 @12:00AM (#8323140)
    how MiniItx boxes are too underpowered and expensive. The 1ghz VIA C3 runs $170ish (not cheap i'll admit), but it boasts the onboard features consistent with most top of the line motherboards and a chip that draws something like 11 Watts max. (Firewire, USB2.0, digital sound out, 5.1 support, tv out etc)

    As far as not being able to HDTV you're dead wrong, I've got an HDTV decoder in it which runs flawlessly (want a 40gb HD version of the superbowl? mail me a harddrive). Gaming is a no go for modern FPS, but even without using one of the 2 pci slots (riser card) the onboard video will run Half Life rather well, and most RTS (save WC3) and of course anything MAME can throw at it... Who wants to play an FPS on a TV but doesn't want a console anyway?

    In short, if you think VIA has failed with their MiniItx form boards and the C3, justify that conclusion. All your complaints are either incorrect or baseless. Divx DVDs and HDTV all run beautifully on the VIA. As for gaming: the most powerful console on the market runs at less that 1 ghz and boasts a far from cutting edge graphics card so it's not lacking in power, just in development support. PC game companies aren't interested in supporting anything but bleeding edge tech, and in all likelihood people who want games on their TV will be looking to the real players in the market: Sony Nintendo and Microsoft.

    • It fails in the $170ish part.
    • by sysadmn (29788)
      If the 1 GB Via board + CPU is ~$170 (retail), you cannot put it into a product that costs less than ~$170, can you? There are scores of consumer electronic devices in the under $200/range. The next big thing is figuring out which ones can be mostly implemented in software, on commodity (or at least high-volume) hardware. What about a $99 DVD player that also plays internet radio? Or a cable set top box that can print photos from your digital camera?
    • I know the Nehemiah series chips are much better than the Ezra variant, but I had trouble with latency on my Ezra-based EPIA-800.

      I intended to use it as a 'multi console emulator' emulating game boy, NES, Genesis, SNES, and some VERY old DOS games. I kept noticing 'lag' when using the joystick and Mario 3. I would hit the jump button and the delay was much more than the console had accustomed me to. DOS games were bad too, but I attribute that to BOCHS more than the CPU.

      Are you running Windows on your EPI
  • by k4_pacific (736911) <k4_pacific@NOspam.yahoo.com> on Thursday February 19, 2004 @12:00AM (#8323143) Homepage Journal
    "Is the X86 ready for consumer appliances?"

    I'd say so, I have an old 486 chip supporting the short foot on my dryer. It has very low power requirements in this capacity and it does a fantastic job of keeping the dryer from wobbling across the basement floor.

    The downside is I have to use plenty of Bounce sheets to keep from ESDing the chip.
  • But WHY? (Score:5, Informative)

    by moosesocks (264553) on Thursday February 19, 2004 @12:01AM (#8323145) Homepage
    The only reason that x86 has endured on the desktop is that it was rapidly adopted by the masses in the early 80s, and being intelligent companies, intel and IBM built upon the platform while maintaining FULL BACKWARDS COMPATIBILITY

    This doesn't make it the best for most uses. It just makes it the most practical for a general purpose computer. But not necessarily an embedded device.

    In the 90s, new, better architectures were introduced, but x86 endured mostly because of the large installed user base. PowerPC, Alpha, and SPARC, if given enough funding during development, would have easily toasted any of intel's x86 offerings. DEC had 64-bit chips before intel had pentium.

    Many new platforms designed specifically for embedded devices such as MIPS and ARM (only ones which come to mind) have developed over the last few years. Backward compatibilty is not an issue here.

    Look at TiVo. They used a 66mhz PowerPC in their 1st generation boxes because they ran fast and efficently, and without active cooling, and it was open and cheap (PPC is a VERY open platform). There is no way that an x86 at this speed could have performed the complex tasks TiVo needed it to.
    • Re:But WHY? (Score:2, Insightful)

      by optikSmoke (264261)
      Thank the heavenly motha-fscking GOD someone said this! Clearly, there is ALMOST NO POINT in putting x86 in a consumer device if it isn't the cheapest solution, as it is certainly not the best. If you're putting linux on the thing, it isn't even a requisite, so I don't really see why you would consider it unless you can find a small and cheap x86 setup. Since the poster has indicated that they can't, I think they've pretty well self-answered that part of their own Ask Slashdot question. NEAT.
    • Also note that because the first generation Tivo was on Linux, the 2nd generation moved to MIPS with no problems.

      F.O.Dobbs
    • Indeed, the PPC platform, IMO is better suited to the embedded market than x86.

      IBM's 750 range (the G3) draws a mere 6 watts at 1Ghz at present - more grunt than most embedded devices will ever need really, and at a much lower power consumption than x86.

      The G3 in my laptop, while having a fan, never needs to turn it on, no matter how hard I work it.
  • by bergeron76 (176351) * on Thursday February 19, 2004 @12:14AM (#8323242)
    It's the primary thing that keeps x86 out of the "appliance" market. No one wants to wait for their device - be it refrigerator, TV, etc. - to boot before being able to use it fully.

    Before you argue that many machines don't need to boot, please keep in mind that MANY do, and can't stay on 24-7. Hell, even the ones that can should shutdown or hibernate in order to keep peoples' electric bills sane.

    Anyway, x86 needs to defeat these hurdles to compete in the embedded arena:

    a) boot a kernel that is bigger than 1M (like ARM can) - why? because if you want to boot a device FAST, you use an uncompressed kernel; and uncompressed kernels are BIG ... and ...

    b) get over the POST time - POST'ing on most motherboards (yes, even VIA EPIAs), takes 10 seconds or more. I know firsthand because our app was initially built on one[an EPIA-M]. Asking a user to wait for 10 seconds for the Hardware to POST, PLUS another 10-20 for the OS to boot is highly unacceptable. Mark another win for ARM here...

    Anyway, the answer to your question is ARM. It's Intel's existing answer. Have a look at an XScale CPU solution young grasshopper.

    • Most newer BIOS... (Score:5, Interesting)

      by Ayanami Rei (621112) * <rayanami&gmail,com> on Thursday February 19, 2004 @04:16AM (#8324427) Journal
      can post in under 2 seconds if not in hardware-change-check mode.

      Besides, if you were making an appliance, I'm sure you could write your own BIOS (take LinuxBIOS, for example). That'd make it boot instantly into any size kernel image you care.
      • You made my point.

        "Most newer BIOS" can POST quickly. I agree. However, does anyone want an AthlonXP powering their fridge? The heat that the processor generates would melt everything in it! There is a very defined niche that embedded systems fill.

        It's not always a good idea to stuff a Square into a Circular hole.

        • with a Celeron 700 in it.
          The system boots into networked linux in seconds from a stock BIOS and the system consumes less than 45W idle with a normal, 7200RPM hard disk spun up.
          Pretty good, huh?
          It's essentially my firewall/router/WAP/webserver.

          And you could easily do a lot better in the power department by using flash storage and lower the clock speed and chip voltage, or by going to a newer CPU (coppermine).

          I'm, I get your point, but it's a bad example.

          What needs to be emphasized is the difference betwee
    • Part of the Windows XP program requires the windows startup screen to appear in 6 seconds. Or at least they proposed that one, didn't Microsoft force that through?

      I agree POST can be a problem. My system takes over a minute to finish POST (yes I have timed it).

    • I have a x86 appliance that boots fast, it's called an XBox. IIRC the stripped down Windows 2K kernel fits in 256K of onboard firmware and also contains the opening animation (so it LOOKS like it is instantly booting). Full boot (from pushing the button to seeing the dashboard) takes probably 10 seconds. If Microsoft can make a x86 appliance, ANYONE can.

      (BTW, my XBox also runs Linux great, and is a terrific emualtion platform.)

      • If what you're saying is true, Linux on the XBox must boot in under 6 seconds!

        I highly doubt it. I'm not saying you are lying, just that you're obviously missing some facts somewhere.

        • If what you're saying is true, Linux on the XBox must boot in under 6 seconds! I highly doubt it. I'm not saying you are lying, just that you're obviously missing some facts somewhere.

          What are you talking about? I was clearly speaking of the boot-up times of the native XBox OS (a stripped down Win2K kernel). Hence "Full boot (from pushing the button to seeing the dashboard) takes probably 10 seconds." The dashboard is the UI of the XBox if you weren't aware. At the end of my comment I was merely me

  • by PurpleFloyd (149812) <<moc.ibtta> <ta> <02onez>> on Thursday February 19, 2004 @12:36AM (#8323403) Homepage
    It seems a lot of people here are assuming that the x86 as an embedded platform somehow still requires an OS like Windows or Linux. It doesn't. Instead, it would probably use an embedded OS like QNX [qnx.com] or VxWorks [windriver.com].

    The issue here is whether the x86 platform's issues, like excessive heat and power consumption and the requirement for a separate memory controller, outweigh its advantages, like the large variety of hardware already available to interface to everything under the sun and the fact that it's a well-understood architecture.

    Now that's out of the way, here's my two cents: the x86 architecture, or at least the implementations currently available, simply isn't cut out for most embedded applications. While x86's limitations have been addressed with lots of extensions (MMX, SSE, 3dNow, etc.), those end up adding complexity and drawing more power than a chip designed without those limitations. Also, the x86's pitiful lack of registers compared to architectures like the PowerPC (another choice for embedded applications that require a good deal of power) means that almost any complex operations mean lots of going in and out of cache, or, worse, main memory. While x86 is acceptable in an environment with a 300W+ power supply and user tolerance for a good deal of noise, it won't cut it in your VCR. x86 might see some use in applications which require rapid development and lots of power, but in most cases there is already a good solution available.

    • It seems a lot of people here are assuming that the x86 as an embedded platform somehow still requires an OS like Windows or Linux.

      Just so that its perfectly clear, Linux is both a desktop Operating System, AND an embedded operating system. Linux scales better than a whole lot of systems, QNX and VxWorks included ...
    • It seems a lot of people here are assuming that the x86 as an embedded platform somehow still requires an OS like Windows or Linux. It doesn't.

      I would have to say, the main reason companies are interested in using x86 is because it will run the common OSes, thereby allowing them to draw on that large pool of pre-existing resources. The price of x86 hardware really isn't that much less expensive than other hardware.

      the x86 architecture, or at least the implementations currently available, simply isn't cu

  • by T-Ranger (10520) <[jeffw] [at] [chebucto.ns.ca]> on Thursday February 19, 2004 @01:17AM (#8323641) Homepage
    You sholdnt be looking at what commercial available, mass produced consumer grade, desigined for PCs MoBo's are out there.

    Get on the phone and call up the manufactures. Get something custom desigined, or at least get pointed at the non-consumer grade web page. If your doing any kind of volume at all, it wont be that expensive. Its not quite as easy as building a computer from componets in your basement, but PC technology is standardized components. Hell, if they have an autorouting board designer they could likely so something from scratch in an afternoon.

  • by sr180 (700526) on Thursday February 19, 2004 @01:31AM (#8323715) Journal
    Dialogic Telephony E1/T1 cards have used 386 chips as embedded chips (mainly for encoding/decoding) for quite a while now. The architecture is obviously good enough for the job.
  • by cypherz (155664)
    It all depends on how much it has to do and how much it has to cost. I really don't know why the EPIA platform "failed" as you say, I don't think the EPIA platform is meant to be a straight-up appliance. if it was the m1000 boards wouldn't have VGA, they would have NTSC (or PAL) outputs only.
    It seems the way to implement x86 appliances would be to use low power procs like the VIA and outboard processors to take the load off the little CPU. A VIA M1000 board with the built in MPEG decoder comes close. With
  • Is This A Troll? (Score:3, Insightful)

    by swdunlop (103066) <swdunlop.gmail@com> on Thursday February 19, 2004 @02:05AM (#8323903) Homepage
    The x86's have been used in embedded apps for quite some time now. I distinctly remember discovering an 80186 in a microwave I disassembled, and many industrial applications use them, due to their well known characteristics.

    All these comments about POST and other silliness come from the PC architecture, not the CPU itself. Amazing how many people are willing to comment when they really have no clue.
  • Hell No! (Score:4, Informative)

    by sheapshearer (746106) on Thursday February 19, 2004 @03:52AM (#8324341)
    The x86 architecture, well, is just plain silly by today's standards...

    A RISC CPU and a few DSPs could perform a lot of set-top applications, with tremendmous savings in both power usage and perhaps area (size).

    High Performance doesn't not mean +100W consumption. If you don't need 4-way out-of-order execution (which is a really, really, really complicated thing to implement), complicated branch prediction, large multi-level caches, etc, then your power consumption will be ** A LOT ** less.

    The fact is that many signal processing applications, don't require large amounts of memory, and they are highly parallelizable. Their algorithmns tend to be much more predictable.

    Also, all processing & interrupt delays are known precisely in DSPs (this is a requirement in realtime stuff). This is also why caches, etc are not desirable, since their performance is not constant.

    Simple DSPs can outperform desktop PCs for a great many applications, using 1/100th the power, cost, etc....

  • Mainstream (Score:2, Insightful)

    by Tune (17738)
    First of all, IMHO it's not about the x86 architecture per se, but rather the concept of a (mini) PC. PCs appear to be attractive in any area simply because most of the hardware already exists, and there's a huge and relatively mature codebase available. In this light, embedding PCs is like making it do what you want and then rip-out anything you're not using.

    Now what happens when you do not start out with a standard PC, but one with special hardware? For example: an x86 processor WITHOUT PCI or a PCI main
  • by torpor (458) <jayv@@@synth...net> on Thursday February 19, 2004 @04:20AM (#8324454) Homepage Journal
    x86 is only one of about 60 different processors that can be used in consumer electronic devices.

    The only question that needs to be asked about whether or not x86 can/should/will be used in consumer electronics devices is the per-unit price.

    If you can't get an x86-based chip for integration into your embedded system for, say, $10 - $15, then its not going to happen. The competition in this sector is too fierce. Other, nice, lower power, fun-to-use (RISC, even...), easy-to-integrate processors are out there, which will definitely give the x86 a run for the money.

    The only thing x86 has going for it in this space is the development realm - yeah, its great to cross-compile for your target processor, but in the end, its also fun to just run the same binary you just built and run on your PC.

    x86 has to get cheaper. Show me an x86-based chip that has tons of SOC-style integrated peripherals, and I'll show you a chip that is just too expensive to compete with the other cpu's we're already using to control stuff, just fine, in consumer electronics-land ...
  • by SmallFurryCreature (593017) on Thursday February 19, 2004 @05:36AM (#8324661) Journal
    The hardware platform doesn't matter much but there has long been a fascination in the computer world with running everything from one CPU. This has left the cpu as the bottleneck and a very hot and expensive bottleneck too.

    However there are some minor signs this may be changing. Most of this is rumors but I think the X-box2 and the new gameboy are both going have more then 1 cpu. Plenty of phones already have more then 1 cpu to spread the load of the increasing demands of the software on them. All the chipmakers seem to be working on putting more then one CPU on a single core.

    So if phone companies, console makers and chip companies think it is the way of the future why not for consumer appliances?

    Think about it, exactly how much cpu power is needed to decode a video stream when a cpu can be dedicated to that task and nothing else? You don't need to go with a SMP like setup. You can simply have one simple processor wich does all the interface stuff. One wich decodes the video. Another perhaps wich decodes the sound. All geared and dedicated to their specific task. Costly? Well to a certain point this is already how PC's work. GPU for visuals. Soundcard for ehm sound.

    Of course such a board will be far more expensive to design then a simple board you pick up of the shelf. With consumer electronics like this still extremely unproven the cost may be too high. Until then simply accept the bigger size and other bad points of PC architecture. Have you ever seen the first generation tv's? Video records? Mobile Phones? Etc Etc? They all had one thing in common. THEY WERE HUGE. Hell the first pocket transistor radio's were so big that the sellers had special shirts with enlarged pockets. (got it from an interview with sony people years ago I am sure someone else can better tell this anecdote).

  • Laptop (Score:2, Interesting)

    If my laptop can play back video, than you can use x86 as a consumer platform, go to whoever supplies the boards to the major laptop manufacturers, and buy how ever many you need, stick a moble chip in, mold a plastic case and cut the headers off of any port you don't want. for that matter, if your just making a demo, close the laptop, buy a PS2 stand thingie, and stick it upright and there you go, i mean, if the Playstation can sit next to the tv, so could my laptop. the trick it standing it on its side,
  • From what I can tell there has been a big trend to move to lower powered devices, while at the same time look for more processing. The x86 is a poor player in this field, since compared to other chips on the market it consumes a lot of power. The two big chips seem to the PowerPC and the ARM chip. Both are RISC chips. My perspective is mainly based from looking at PDAs and the direction that game consoles are going.

    If you think to the contary, then please tell me.

  • I have and help set up a group for hacking an old set top box made by Acer. Inside was a nearly bog standard PC motherboard. It had all the regular parts of a motherboard. There was one ISA slot (used for a modem or NIC). There was a header for a COM port (you had to add a particular Maxim SMT part to get it to work - most people used it for a mouse). It came with its own wireless keyboard, and used a AMD 586/133 for the CPU. IIRC, it had 8 meg of RAM. It also had a smart-card reader. All packaged in a slic
  • Slashdot article... (Score:3, Interesting)

    by herrvinny (698679) on Thursday February 19, 2004 @01:07PM (#8327923)
    Wasn't there a slashdot article about someone using a tiny Intel CPU to build a whole computer on a chip only 5-6 by 1-2 inches big? Wouldn't that be useful? I visited the site when the slashdot article went up, but now I can't find it. Anyone have a link?
    • I actually own one of those. ;-) It is a complete pentium-class computer in a 2x3x3/4 package. It uses an AMD Elan processor, and IBM Microdrive, and a few other components (memory chips, cmos battery, etc.)
      I bought it a few years ago from Tiqit Computers [tiqit.com] (a company founded by some people from Stanford). It was their now discontinued Matchbox PC model. I believe Slashdot had an article about some guys at Stanford using one as a webserver back then.
      A friend of mine actually installed Windows 98 on it..
  • The reason why you'd want an x86 is to leverage all the existing code which runs on it--practically none of which is relevant to embedded devices.

    It's not as if you were going to program your embedded application in Visual Basic, or as Lotus 1-2-3 macros, or something.

    Designing a consumer appliance with an x86 processor in it makes about as much sense as designing a rocket ship that runs off 97-octane gasoline, just because gasoline is more familiar, more available, safer, and more competitive in pricing
  • by MountainLogic (92466) on Thursday February 19, 2004 @04:44PM (#8331413) Homepage
    As an engineer who has been on the inside of creating mass consumer electronics let me tell you the decision I made were very different from what you might expect. When you order anything in quantity one million everything is off the shelf. Or to put it another way the NRE (Non Recoverable Engineering) costs per unit for doing something from scratch is so little there is no reason not to do so. In fact, I've yet to see a major chip that was not worth doing a turn on to drop the unused parts and to reduce the cost (reduces area, reduced I/O pads, increased yield and reduced testing). If you talk to the project leads at the major chip houses they do not expect major electronic companies to use what's in their catalogs. They view those as resumes to show what they can do and why you should give them a call. Only under funded startups and niche plays use catalog products. The only reason to use off the shielf is time to market and in consumer electronics if it isn't as cheap as it can be its not ready for market.

    Open up a TV or a VCR. do you see a standard bus in there. No! And why? Because everything in there is designed from scratch to play nice with each other. Standard products carry extra capabilities that a fixed large volume product does not need. You'll know that STBs and DVRs have hit the big time when you see a single board with custon chips, the code in ROM and no I/O besides required cable/AV in and AV out.

    If it were me, I'd buy a small CPU like an ARM7 and do all my heavy lifting like transcoding in hardware.

  • Advansys (Score:3, Informative)

    by aminorex (141494) on Thursday February 19, 2004 @10:07PM (#8335316) Homepage Journal
    Advansys is selling Ezra-800MHz SOM (system on module) in ETX form factor (3.7 x 4.5 ") for twice EPIA prices, at this [advantech.com] link; while their EVA SOC (system on chip) only reports 80186 performance levels, and have embedded RTOS, probably TRON [tron.org]so it's not what you're looking for.
  • can u say OMAP (Score:2, Interesting)

    by sydres (656690)
    TI's Omap processor; low cost, high performance(some bench it to be as fast as 400 mhz xscale)ARM core+dsp+controls for various subsystems such as usb, serial, sd/mmc, etc.
    got a 126mhz model in a palm zire 21 runs for
    12 hours continuusly on a single charge.
    and it has a native linux port as well as various codecs that use the dsp for acceleration 2d/3d. also unlike the XSCALE this chip has builtin coprocessor and 192k L1/L2 cache as well as a 1.5mb cache all done on 90nm process
  • Why bother? (Score:3, Informative)

    by AuMatar (183847) on Friday February 20, 2004 @05:05AM (#8337690)
    Why bother with x86 for an embedded hardware system? its going to be much more expensive to build a unit than doing it theright way with an embedded processor and DSP's and ASICs for the heavy lifting. I could see it if you're in a race to prototype or hit a market first, because you can leverage a lot of code. But otherwise you're going to do alot of work trying to make x86 fit a niche it isn't made for, when you can do it quicker and cheaper with other solutions.
  • by AlecC (512609) <aleccawley@gmail.com> on Friday February 20, 2004 @09:28AM (#8338593)
    Do you really want to embed windows in a consumer device? Probably not - I certainly wouldn't. Since you are posting on/., you probably have some affection for Linux. But there are plenty of much lighter wight systems around, like QNX, vxWorks, which are also much more suitable for ebedded work. Windows is very, vety heavy. Yes, you can buy a 3GHZ x86 to run Windows - but a 1GHX Risc processor optimised for embedded work wil outperform it by two or three things and undercut its power consumption by 20 times.

    Once you have ditched Windows, all the other OSes run on multiple platforms - Arm, PPC, MIPS, Coldfire, Hitachi H series... Linux is certainly available on Arm, PPC. Most of the others are available on more architectures.

    Which means that if you chose one of these OSes and (usually) C++, you can move platforms with a recompile. (Not quite true, but near enough for overnment work).

    I have experience of the Arm family, and they go like lightning when programmed right - much faster, MHz for MHz, than you would expect compared to Windows. And the power consumption is small to minuscule. And there are some very interesting new CPUs coming along obviously targeted as set-top boxes (sorry, NDA doesn't permit details and Google doesn't know yet). So what you need is for the set-top manufacuters to agree on a common OS like the mobile phome manufacturers have done (how about the same one, as a suggestion) and use the best of the new generation embedded processors.
  • Xbox.

    I'll leave it to you to decide if this example is good for x86 or not.

    (IMO, Xbox is a great example of the x86 chip succeeding in the CE world.)

    As already mentioned, the older Airport Base Station has a 486 in it.

    Even with these successes, it really depends on the device. x86 is a general purpose architecture, designed to do everything good enough. There are other chips that are much better designed for specific applications (example being PPC, which while still being a general purpose chip, smok

No amount of genius can overcome a preoccupation with detail.

Working...