Slashdot Mirror


CEO of Centaur Discusses x86 Strategy and Linux

An anonymous reader writes "This fascinating interview with Glenn Henry, founder of VIA processor subsidiary Centaur Technology, discusses the founding of Centaur, its strategy and products, and why Linux is fundamental to his company's success. Additional topics covered include: how to produce an x86 clone with a few million dollars and a few dozen engineers; the embedded x86 market, and how it compares to the traditional ARM and MIPS based embedded market; why Centaur doesn't compete with AMD and Intel so much as enable x86 to reach new markets; how Linux is enabling greater hardware functionality; the urgent need for pervasive security -- and much more!"

24 of 135 comments (clear)

  1. Wondering by swordboy · · Score: 4, Interesting

    I've been wondering why some company like this doesn't create a "network appliance" specification for all of us to hack on. It would be nice if I could just go buy a Netgear router and roll my own Linux installation. I purchased a Toshiba Magnia SG10 some time ago when they were a couple hundred bucks during the end-of-life period. For a 566mhz Celeron with an honest-to-goodness hard drive and switch on the back, it was hard to go wrong. I immediately wiped the stock Linux OS and rolled FreeBSD on there.

    Wouldn't it be MORE profitable for companies like the aforementioned Netgear to do this? What am I not seeing? Centaur: help us out!

    --

    Life is the leading cause of death in America.
    1. Re:Wondering by Anonymous Coward · · Score: 3, Informative

      If you want that, get a Soekris box. I run a few of them and they're very nice.

  2. Is x86 the best chip to use by millahtime · · Score: 4, Interesting

    Is an x86 or clone really the best chip to take to markey this way? Linux will run on other processors and the x86 isnt' the best archicecture. There are processors that are more efficient, use less power and can run linux.

    Although, I'll admit some of those embedded boards that I have seen are pretty cool and easy to use.



    P.S. I know I can't spell. That's why I'm not an english teacher.

    1. Re:Is x86 the best chip to use by stratjakt · · Score: 5, Informative

      Whether it's the best "textbook" architecture or not, it's entrenched itself in the computer world so deeply that x86 will be around forever.

      It works, engineers and coders are familiar with it.. The embedded world is still largely about ASM routines and tight little loops that dont leave the cache, etc, etc.. You can write and test your fancy-shamcy hand-assembled routine at your leisure on your desktop computer.... There are plenty of upsides, and the only downside is the "other stuff looks better on paper" argument.

      It's a newcomer to the embedded world, but will no doubt take a strong hold.

      --
      I don't need no instructions to know how to rock!!!!
    2. Re:Is x86 the best chip to use by squiggleslash · · Score: 3, Interesting
      It's a nice idea, but there are still a hell of a lot of binary-only packages for Linux (and GNU/Linux) varying from display device drivers to video decoders (such as Real.) Unless your chip is likely to capture a significant portion of the market, it's not terribly likely that many of those binary packages will be ported.

      On top of which, as long as it runs ix86, platforms with the chip will not be limited to Linux.

      --
      You are not alone. This is not normal. None of this is normal.
    3. Re:Is x86 the best chip to use by iguana · · Score: 5, Interesting

      As an embedded Linux developer moving from x86 to ARM, there are good reasons to stick with embedded x86.

      - My Linux development PC runs x86 so stuff compiles/runs on my development machine will tftp over to my embedded board and run immediately.

      - The GNU x86 toolchain is WAY easier to use than the cross GCC tools. All GNU (Linux kernel and gcc especially) projects target x86 first and everyone plays catchup. Bugs are found/fixed in x86 first. Features work on x86 first.

      Reasons to move from x86 to ARM.

      - ARM is designed as an embedded CPU and x86 is designed as a desktop CPU. So the ARM has a very simple memory map, you don't have to worry about working around 15 year old cruft (why the foo doesn't a modern x86 boot straight to protected mode?).

      - It's hard to find an x86, even one ostensibly designed for the embedded market, without the "kitchen sink" problem. Our product with >500 parts with an SiS x86 dropped to 300 with an Atmel AT91 ARM. Most of the passive parts on the SiS board were to turn OFF crap we didn't need (IDE controller, SD controller, video controller and so forth).

      - ARM is cheaper. ARM is smaller. ARM uses less power. ARM is cooler (temperature). ARM is cooler (it's a very nifty chip design).

      Just my opinions, lightly flavored with experience.

    4. Re:Is x86 the best chip to use by chrysrobyn · · Score: 3, Insightful
      Is an x86 or clone really the best chip to take to markey this way? Linux will run on other processors and the x86 isnt' the best archicecture. There are processors that are more efficient, use less power and can run linux

      Give PPC a try. I love my LinuxPPC machines, they're wonderful. I honestly fell in love with the PPC in college, first in real world x86 emulation (VirtualPC) and then in Computer Architecture and Computer Hardware Design. It's a lovely architecture, and you can see its low power uses in the iBook. I've read that the 750 in the iBook used less power than the southbridge, but I don't have references to back that up. Suffice it to say, the PPCs have been traditionally pretty low power, and have pretty good desktop acceptance as far as non-x86 processors go on desktop machines.

      After you get through the base packages of a useful distro, like Debian or a multitude of others, you'll start to find x86 only packages-- retail packages usually don't even acknowledge that Linux runs on non-x86 hardware! Is x86 compatible really the architecture to target for Linux acceptance? If I didn't love PPC and have a source of cheap old Macs, I'd rather be running a VIA / Cyrix derivative because of how cool they run. Since the Pentium class, x86 hasn't been about the native instruction set, it's been about the state machine that re-orders, re-names and actually issues native instructions on a pretty nice, flexible RISC inside anyway. So, Centaur's problem is basically updating their RISC core and keeping their x86 runtime emulator tuned. That way you don't have to bend anybody's mindset to make packages for some off the wall architecture, or beat another architecture at their own game.

    5. Re:Is x86 the best chip to use by iguana · · Score: 4, Interesting
      The embedded world is still largely about ASM routines and tight little loops that dont leave the cache, etc, etc.. You can write and test your fancy-shamcy hand-assembled routine at your leisure on your desktop computer....

      Not really. There isn't much assembly necessary on a modern embedded platform. Even the little 4- and 8-bit processors have specialized C compilers available. Assembly is necessary for startup but that's about it; then we just jump to main(). I think Windows game programmers write more assembly than most embedded systems folks do.

      Embedded systems development is just like any other development. How can I get the best product out as quickly as I can? ASM doesn't get me there quickly enough.

      That's why using embedded x86 can be a good fit. It may not be the best technical solution, but we'll get to market quickly because the tools and available leveragable source is so good.

    6. Re:Is x86 the best chip to use by AnwerB · · Score: 3, Informative

      Actually, the architecture I would recommend (and the one I use) for really small footprints is the ETRAX 100MCM from Axis.

      The new version (it's not on the web yet, but they are shipping in limited quantities) includes on chip: 4 Mbyte Flash, 16 Mbyte SDRAM, an Ethernet transceiver, Reset circuitry, and dozens of passives (resistors and capacitors), and all the usual: 4 Serial UARTS, a parallel port, SPI bus, etc.

      So if you wanted to build a credit-card sized Linux machine with Ethernet, BOA webserver, SSH, FTP, vi, sqlite, LUA scripting language, etc., you could (and you would still have at least 1.5 MB left over).

      This is what I use for my embedded designs, and it's astonishingly easy using the cross-compiling toolchain (basically, just gcc and a change in the makefile for target) that they provide. They also have a mailing list for questions if you don't want to call.

      Oh, and their architecture is now supported in the main fork of the 2.6 kernel.

    7. Re:Is x86 the best chip to use by Kazymyr · · Score: 3, Insightful

      Thumbs up for ARM (pun not intended). Plus you can take a synthetic ARM core and dump it to silicon together with the peripherals of your choice, and get a system-on-chip with very little effort. Try to do that with a x86 - not!

      --
      I hadn't known there were so many idiots in the world until I started using the Internet -Stanislaw Lem
    8. Re:Is x86 the best chip to use by Anonymous Coward · · Score: 5, Interesting

      I honestly fell in love with the PPC in college .. It's a lovely architecture

      PPC is nicer than X86 but it's hardly a "lovely architecture." In fact, it's one of the uglier RISC designs.

      - Backwards bit numbering scheme. The MSB is referred to as bit 0. Normally, it's just a matter of convention and makes no difference but there is some inconsistency in the ISA about this. For example, shift counts are stored in the 5 LSBs but indirect segment register indices appear in the upper 5 bits (probably because of the MSB convention.)

      - Some really complex instructions which increase processor complexity. Rotate-with-masked-insertion? That's not RISC. IBM's POWER chips actually need to break down some of the instructions into smaller ops, similar to X86.

      - Flags. It's nice that most flag computations are optional and are only done if explicitly requested, but flags don't belong in a RISC architecture. The way PPC implements them is even more complicated than on most CISC machines: A flag register with 8 4-bit CC fields and additional flags in a separate XER register! (??)

      - A bizarre MMU. The paging scheme isn't as efficient as other architectures, even including X86. Apparently this is IBM legacy cruft from a previous architecture (possibly S360, can't remember.)

      As far as RISC architectures go, Alpha was probably the best by far, and the actual implementation was great starting with the EV6.

    9. Re:Is x86 the best chip to use by chrysrobyn · · Score: 3, Insightful

      This discussion is perhaps more suited to ArsTechnica, as the issue at hand is whether or not x86 is the best instruction set for a low power budget chip maker to pursue. I hope you don't hold that Alpha would be a more suitable instruction set for that -- I don't even want to think of what packages aren't available for Alpha, let alone ask a vendor for a program ported to my "Alpha-Compatible". Academically speaking, however, I will reply to your PowerPC attacks. Regretfully, all my college textbooks are 300 miles away, and I have not looked through my senior textbooks in 5 years, so my rebuttal cannot cite specific Alpha shortfalls beyond anecdotal recollection.

      PPC is nicer than X86 but it's hardly a "lovely architecture." In fact, it's one of the uglier RISC designs.

      4 precise detractors against the PPC ommitted, I concede that these are 4 of very many

      As far as RISC architectures go, Alpha was probably the best by far, and the actual implementation was great starting with the EV6.

      The Alpha processor team spent years learning that many of the architecturally correct ideals they had held needed to be thrown out when it came to the real world. According to Torvalds, "And all the RISC stuff that tried to avoid it was just a BIG WASTE OF TIME. Because the _only_ thing the RISC approach ended up showing was that eventually you have to do the hard stuff anyway, so you might as well design for doing it in the first place."
      Reference

      Make no mistake, I make no claims about the Alpha deserving the fate that we all foresaw when DEC made the deal with the Devil. Alpha had its share of problems (remember the quick race to 500MHz, then the hard stop for a long time while they straightened out their unbuffered cross chip clock signal?). The RISC/CISC debate ended long ago when Intel effectively merged them. Since then, the two different schools have borrowed from each other so much that we expect chips to have some cruft to reflect their original intents. For this reason, we'll see all Itanium derivatives reflecting that they were made for pre-defined behavior where a compiler can make predictions of the instruction ordering, and likely this will continue to be in intensely predictable math operations. At this time, anyone who insists that modern PPC, x86, Alpha or Sparc chips are exclusively RISC or CISC hasn't taken the time to realize that there is a very blurry line between the RISC and CISC lately. Many have argued that this line is effectively erased and that the definitions have only had meaning in college courses for the past decade.

      The Alpha may be your ultimately pure holy processor, but I will continue to find PPC to be a great balance between theory and real world needs and heritage. If nothing else, x86 has taught us that we can have incredibly high performance architectures based on 1970s vintage processors like the 4004.

  3. I wonder by foidulus · · Score: 4, Interesting

    if they could take this technology and turn it into a "portable pc gaming console" of sorts(of similiar size and shape to the gba) In the article it states that they can run at 533 MHz fanless with a worst case power consumption of 2.5W. Wouldn't it be neat to create a gaming console(kind of like the phantom pc gaming console, only portable) for this with flash cards being the "cartridge"
    Though licensing for the abandon-ware would be a pain. As would trying to standardize the input across a large number of games. But still, it would be friggin' cool to play leisure suit larry while you are bored in class!

    1. Re:I wonder by foidulus · · Score: 3, Interesting

      Yeah, but with the small screen you are limited to on a portable, do you really want anything fancy? I'm not sure I could play unreal tournament on a small screen. There are a ton of old games that you cannot find anymore(Leisure Suit Larry, a whole load of freeware games that came on this cd I bought 10 years ago) that were a lot of fun, but not very portable. Plus, a lot of times you had to mess around with "low memory" and whatnot just to get them to run. It would be nice to have a console that just "worked" with all those games.
      It's too bad that the companies who made those games either:
      a) don't exist anymore or
      b) don't want to open source the games even though there is almost no chance of the games bringing any more revenue.

  4. Going after a different market by Da_Slayer · · Score: 4, Insightful

    I think it is a sound idea to go into the niche markets so to speak instead of just jumping into the fray with AMD/Intel. Everyone has had enough of the SSE2 vs 3dnow! extended vs the new kitchen sink it comes with. Those types of things have no real bearing on markets where companies are looking for solutions that are cheap, easy to deploy and know the company is designing the hardware for their problem. Not the company having to make their systems go on the vendors limited products.

    The more competition the better we the consumers are at getting the best products.

    --
    Push harder towards Open Media/Content
  5. Open source designs? by Trigun · · Score: 3, Interesting

    The biggest problem with most hardware is that it is exotic. I'm not talking about PC's, and routers have become cheaper with micto-atx mobos with dual ethernet onboard. Laptops and PDA's are where there seem to be a lack of standards. There are no displays which can be plugged into your PC without special driver circuits. Why not create a LCD that has a standard interface that is compatible with PC hardware now?

    If you standardize the equipment, then you will drive prices down, and you won't be stuck throwing out a PDA that has a broken screen, or has eol'd. Just upgrade it.

    1. Re:Open source designs? by Allen+Zadr · · Score: 3, Informative
      If your willing to search, you never have to throw anything away, but that's also a lot of work. If I think my time is worth, even as little as $15 an hour, I couldn't repair / upgrade a Palm/WinCE device for less time than the money required to buy a new one (well, maybe a high-end WinCE device).

      It's throw-away lock-in. Same thing that happened to Televisions and VCRs 20 years ago. Standardizing the parts and interfaces won't help - the labor required is simply too time conssuming.

      Laptops are amazingly upgradable, and even those are less and less apt to be worth the time required. You mention driver circuits, but really - there are somewhere around 80 different driver circuits in current use for laptop displays. Yet, there are some 500 laptop models out there. And without the the driver circuit, the Liquid Crystal display is far cheaper than an integrated unit would be. So, I think that part of your argument is counter-productive. LCD + accompanying driver costs a lot more money, yes. However, the video card hardware can be configured to talk to most of the driver circuits (they are close to standarized at the interface level). Again - it's just a lot of work.

      When you don't have the 90% air that most PC chasis hold, you can't have big bulky large finger capable standard connectors between every part. Sometimes, you have to route your signals through flat cables or custom bundles.

      Really that pain-in-the-ass to price point is even hitting PCs. Once a computer is more than 3 years out of date, it actually becomes cheaper to simply buy a package deal. If you really liked your case, swap it, the case was free with the bundle - along with yet another floppy and CD-ROM. As prices drop, the three years will turn to 6 months. Or about the same period between major CPU/architecture performance boosts.

      --
      Kinetic stupidity has a new brand leader: Allen Zadr.
    2. Re:Open source designs? by Trigun · · Score: 3, Interesting

      Flat cables aren't that difficult to work with. Proprietary flat cables are a pain to work with.

      As an example, I have an IBM thinkpad 380Z with a broken lcd cable. Simple enough, right? Replace the cable. They're mostly standard connectors. Except the cable has a variety of surface mounts on it, and can only be replaced with a $125 replacement cable. If it didn't have the surface mounts, I could use an old IDE cable to replace it. If it had a driver circuit that wasn't built onto the mainboard, I'd have a vga display to play around with.

      All I've got now is a broken laptop.

  6. centar by millahtime · · Score: 3, Interesting

    Centaur is positioned to tap the largest market in the industry- the sub-$400 PC market. Right now Centaur is designing a processor that will run over 2 GHZ.

    A 2 GHz machine that is x86 compatible but will it have all those nifty other features that windows can use to speed it up? If not, it would be much slower than AMD and Intel.

    Although, the thought of being able to use a chip like that for some embedded app would be pretty neat and pretty powerful because you don't need all the extra stuff that windows uses.

  7. Interesting trend by IWantMoreSpamPlease · · Score: 3, Insightful

    Every other day, it seems, someone is shouting about how their company is finding linux is crucial to their success/business plan/what have you.

    I wonder if it's a case of corporate "me too!" or if all the small firms were simply waiting for some large firm (IBM for example) to thumb their corporate nose at Microsoft, before they decided it was safe to do so.

    Don't get me wrong, I'm all for firms being able to decide who to attach their sail to, I was just wondering why it was taking until now...

    --
    So rise up, all ye lost ones, as one, we'll claw the clouds.
  8. Out of skew by b0lt · · Score: 3, Interesting
    Additional topics covered include: how to produce an x86 clone with a few million dollars and a few dozen engineers;
    Didn't Intel CREATE the x86 with less than a million dollars and a few dozen engineers?
    --
    got sig?
  9. Re:Nice strategy but... by sam_van · · Score: 3, Insightful

    While system costs are higher for the consumer assembling a system, I don't think that a product for the masses will encounter the same problems. Manufacturers enjoy economies of scale that we don't; otherwise the low price providers (Dell/HP/etc.) would still be selling generic beige boxes instead of snappy looking cases.

    --
    Thinking of starting a business in Minnesota? Me too! mnsmall.biz
  10. cheap laptops by zogger · · Score: 5, Insightful

    we have the cheap desktops now, like the walmart 200 buck boxes, but does anyone make a *new* laptop that uses this guys chips and a via mobo, and is it under 500 clams brand new? what he says is true, and I'm in that 90% range that what passes for a mid range speed is MORE than enough for my purposes. I don't do gaming or weather modeling, etc. That's the breakthrough and the sweetspot general pricing range I am waiting for, the linux laptop,comes complete and works outta-the-box, including wireless, under 500$, and *upgradeable*. Is this possible now? Say it is, I mean, 3-4 years from now I could replace the whole mobo with whatever is cool then, along those lines, yet alone just swapping in a new cpu, etc. And a REAL battery (or batteries even better, in some sort of standardized arrangement, with 12 volt DC input being standard) in it, I'll tote a couple extra lbs, I don't need a laptop to weigh sub-3 lbs, 6-7 is still quite acceptable, it's the same as the ones I have now. His chip at 7 watts sounds great, and 1 ghz is perfectly acceptable. Heck, even if it had a switch to toggle it back and forth between 3 watts and 7 watts would be nice, as in clocked/not clocked.

    Desktops are a different story, you can always swap around parts and do a little drilling and cutting, etc to make anything fit, but laptops are teh sucks for upgrading and working on, more or less, and they are too expensive as they are sold now to change out very often (for me I mean, but bet a lot of other folks feel the same way). I'd get one and use it for my main desktop most of the time then with an external keyboard and my regular mouse and monitor, scrap energy hog desktops, but retain the option of true portability.

  11. Re:Not Midrange performance by zogger · · Score: 3, Insightful

    Well, even so, it's still would be better than what I am using now, a PP200. I know all sorts of people still using older systems because they paid a lot of money for them and they still work well enough. So, you are right on what is top end now or midrange, but the centaur guys observations that his chips are good enough for 90% of what most people use a computer for are most likely still valid. Here on slashdot you have a much higher ratio of people who have brand new or pretty new systems than you do in the "general" population, and uber geeks tend to hang out more with other uber geeks, so perhaps they don't see it as much. I know most corporate bosses make so much money and live such a different lifestyle it's hard for them to relate to joe working stiff, that's why you see so many weird business changes-marketing and the bosses can't relate to most people, they lose track of anything outside their little niche worlds of much better off financially people than any sort of median-norm. I know there's no company out there that I have seen is offering any sort of new laptop that even possibly could be in my price range, so I hang on to the older ones, and I bought them used. Same with my desktops.

    There's different niches out there, and there's a huge niche where "cheap and good enough" is still king, and it was refreshing to see an article on slashdot where that is covered, usually it's some new thing that is the fastest/best and most expensive, and most energy hog as well, and seeing as how I am into alternative energy, I always look at that angle. I don't want an electrical sub station needed to power me and my house and stuff, I think that is beyoind ridiculous and getting into the generational greedy range. that's ME, others can think different, but it is how I think...

    This guy at centaur gets it on where his market is, and I'm exactly the sort of person he is thinking of, that's why I wondered if any company had an upgradeable cheap laptop based on his cpu/mobo stuff yet. Like my desktop, when I get ready I'll put a new board and CPU in it, but for now, this is "good enough" for me, just never seen a laptop I can do the same thing with, and I want low power, good battery life, and upgradeable by switching a few cables and slapping in a new board for cheap a few years hence.

    Just a-wondering is all, no biggee. If I really wanted or needed a liquid cooled monster like we see all the time here, I guess I would get one, but it would be overkill for my needs, and definetly suck way too much juice and throw too much heat. don't need it really, like I said, don't do games or modeling or whatever. Folks who do got tons of choices, folks who don't got almost squat for choices it appears, and I despise the forced upgrade business solutions we keep seeing.. There's plenty of choices out there in that expensive/faster/ mo powah direction, but very few choices the other way.