Forgot your password?
typodupeerror
Intel Hardware

Intel To Redesign PC With "Grantsdale" Chip 309

Posted by CowboyNeal
from the a-whole-new-bus-is-coming dept.
MarkRH writes "Over at ExtremeTech, we tracked down some Intel roadmaps that discuss "Grantsdale", Intel's most important chipset in nearly a decade. Grantsdale brings PCI Express to the PC, so get ready to toss out your motherboard, AGP graphics card, and maybe a host of other components, too. Also check out our articles on the "Tejas" microprocessor, Intel's first CPU to forego pins (check out the waffle iron socket!), as well as the real reason Banias saves so much power."
This discussion has been archived. No new comments can be posted.

Intel To Redesign PC With "Grantsdale" Chip

Comments Filter:
  • Hey (Score:2, Funny)

    by papasui (567265)
    Now you can finally play Doom 3.
    • Now you can finally play Doom 3

      The funny thing is that this joke will still be funny for years to come ..

  • by writertype (541679) on Friday February 28, 2003 @12:40AM (#5403844)
    It's going to be really interesting, I think, to see what this does for the holiday selling season. Since it's out there now that Grantsdale is going to have such a dramatic effect on PC architecture, what is this going to do for sales of graphics cards? Of sound cards?

    It looks like PCI will be supported in some way, but it's almost up to a motherboard manufacturer to come forward and say, "OK, we're only going to support one PCI slot, so figure out what you want to keep, now."

    My guess is that Nvidia's NV35 will be released later this year (fall?) on AGP8X, but that it will REALLY run well on PCI Express. So--wait, or buy? An old question, but with far more significance.
    • by Max Romantschuk (132276) <max@romantschuk.fi> on Friday February 28, 2003 @12:49AM (#5403892) Homepage
      It's going to be really interesting, I think, to see what this does for the holiday selling season. Since it's out there now that Grantsdale is going to have such a dramatic effect on PC architecture, what is this going to do for sales of graphics cards? Of sound cards?

      I doubt this holiday season will be any big break for PCI Express. Remember when PCI was introduced? Roughly around the time of the first Pentiums. You can still buy motherboards with ISA slots...

      It looks like PCI will be supported in some way, but it's almost up to a motherboard manufacturer to come forward and say, "OK, we're only going to support one PCI slot, so figure out what you want to keep, now."

      The same applies here, the transition won't happen over night. There is lots of stuff which runs just fine on the bandwidth that PCI has to offer. You will have to decide what to keep, but I'd say that years from now.
      • by buffer-overflowed (588867) on Friday February 28, 2003 @01:18AM (#5404012) Journal
        And sometimes you need an ISA slot. It's rare, but recently I've had occassion to really, really need one (in fact several...).

        Sure they're slow, ancient, legacy (appologize for the redundancy there) but sometimes you just really need an older piece of hardware or a board you can solder and design yourself without an EE degree.

        The same will be true of PCI. There are more PCI cards out there than ISA, so PCI-Express should really be backwards compatible, capable of both modes. Or at least have a few slots on it that are mutual, then faze it out over a few years.

        Why don't major vendors get the fact that some of us like our legacy stuff and don't want to move just because we "have" to?
        • Why don't major vendors get the fact that some of us like our legacy stuff and don't want to move just because we "have" to?

          Oh I'm sure they do, it's just they make more money this way.
        • > Why don't major vendors get the fact that some of us like our legacy stuff and don't want to move just because we "have" to?

          Why don't consumers get the fact that their hardware would be faster, cleaner, easier to use, and downright sexier if legacy stuff didn't have to be supported?

          Take Intel CPUs. They're a kludge. A terrible, messy, evil kludge. And they're a kludge because they have to support legacy applications that ran on the 8086.

          Intel, of course, is making exactly the same mistake by attempting to emulate x86 modes on the Itaniums.

          If you really, really want to use legacy stuff, then go and get a PCI to ISA Bridge or something. But don't try and force ISA compatibility into PCI-Express, because that's just going to make things slower (and messier) for everyone else.

          -Shane
          • Responding in arbitrary order:

            Intel, of course, is making exactly the same mistake by attempting to emulate x86 modes on the Itaniums.

            Because of its abysmal performance, Intel has abandoned this approach and now uses a Celeron coprocessor to handle x86 software execution.

            Take Intel CPUs. They're a kludge. A terrible, messy, evil kludge. And they're a kludge because they have to support legacy applications that ran on the 8086.

            Even though this is quite a popular sentiment on /., the reality is that consumers and businesses don't want their computers to be some sort of Platonic ideal of a perfect machine. They just want the damn things to work reliably, chock full of "evil kludges" or not.

            So the attitude you're propagating, that older is bad and anything supporting something older is therefore also bad, is pretty much not a viable approach to the economic realities at work here. You seem to think that sweeping revisions and corrections that throw out the work of previous generations are justified, but really all they would do is anger users and create chaos. It might be faster, cleaner, easier to use, and sexier, but it also be much less useful simply because there wouldn't be any software that ran on it.

            And you can argue that it's the principle of the matter, and that we shouldn't compromise our ideals just because they are unworkable in the market. But keep in mind that computers are a tool, not a religion, however tempting it may be to categorize them as such, and it's a bit silly to attach an ideology and a morality ("evil kludges" indeed!) to tools.

            • You seem to think that sweeping revisions and corrections that throw out the work of previous generations are justified, but really all they would do is anger users and create chaos.

              Yep, I know so many Mac users that just hate the PowerPC and want to go back to the old 68k CPUs. And OS/X is a big failure for the same reason.

              Hint: do it right, and 90% of the consumers won't even know about the sweeping revisions.
        • Actually, if you read the article, PCI and ISA can still be supported. ISA is supported on current PCI boards through a host-to-host-bridge. Likewise, the new architecture can support PCI, and there's no reason to think ISA couldn't be supported.

          The issue is that such a bridge costs money and space. The majority of people purchasing new computers (with no intent on upgrading HW) will want the cheapest, most-compact solution. This is the problem that faces video card retailers. They'll ahve to carry two lines of video cards (3 if they seriously still sell PCI versions), and they'll have to conflict amongst themselves as they market different HW. Especially in the months leading up to the roll out.

          No word on whether AMD boards have any chance of supporting the new BUS. If not, then AGP cards will persist for some time.
      • by writertype (541679) on Friday February 28, 2003 @01:36AM (#5404078)
        That's the question. If you read the article, there's going to be four PCI Express x1 slots.

        OK...so does that mean those are going to take the place of the PCI slots that will normally be found within a motherboard? PCI will be supported--but how many slots will we have to work with?

      • I have several ISA devices that I still use...

        One's a modem (a REAL modem that I can configure with jumpers, that works under linux... don't even get me started on winmodems)

        I also have a WINRADIO [winradio.com] (or LINRADIO [linradio.com] under linux); Those come as ISA cards if you get the internal version...

        Yes, I know it's a dinosaur hardware interface, but I still find it useful... and I'll bet I'm not the only one. Hardware may find itself deprecated and unsupported (Apple's Newton, anyone?), but still useful.

    • Nvidia's been releasing the NV30 cards (Geforce Ti4200 and MX series) as AGP8x modules.

      The problem is that the bandwidth that is offered by the AGP bus tends to be a PCI-AGP bridge, rather than a true AGP graphics card, so what you essentially have is a PCI card running at a slightly faster dedicated bus speed.

      If PCI Express can truly deliver, I'll be impressed... but Intel's known for making decisions that are not necessarily widely implemented in the long run (remember Rambus?). I'm taking a wait and see approach with this one.
    • Who cares? (Score:5, Interesting)

      by Anonymous Coward on Friday February 28, 2003 @02:06AM (#5404175)
      First of all PCI-express will come in second half of the NEXT year.

      Second, PCI-Express x 16 just double AGP8X bandwidth. We can expect same "dramatic" (1-2%) performance increase as we saw with AGP8X and AGP4X. It will take many years until this kind of performance is really needed. Since high-end video cards will have 512MB of very fast (~40GB/s) local memory in H2-2004, 4GB/s bandwidth offered by PCI-Express won't make much difference compared to 2GB/s AGP solution.

      PCI-Express add-on cards won't be popular anytime soon. Since:
      1) PCI replacement (PCI-Express x 1) offers just 250MB/s of bandwidth, thats isn't a lot more than current 133MB/s offered by PCI.

      2) >90% of users won't need any external cards in H2-2004. Currently we have following stuff integrated on the chipset/motherboard:
      -two 100Mbps NICs
      -Sound with better quality than original Audigy
      -Firewire/USB2 etc

      In 2004 we will also have:
      - NICs will be updated to 1Gbps
      - Wireless LAN
      - DSL modem

      3) In the server market PCI-Express won't be popular since it isn't compatible with PCI. Currently servers use PCI-X (1GB/s) and it will be replaced with PCI-X 2.0 (2GB/s). This is enough bandwidth for many SCSI-raids and Gigabit NICs.
      • Re:Who cares? (Score:4, Interesting)

        by poot_rootbeer (188613) on Friday February 28, 2003 @12:40PM (#5407262)
        Yes, PCI Express will only be an incremental improvement over the latest AGP spec. But there are other devices on the peripheral bus that need to move a lot of data around.

        Your processor runs at an internal clockspeed of what, 1.5GHz? And your PCI bus? IIRC, it maxes out out a paltry 66MHz. The peripheral bus is already a bottleneck, today.

        I don't care how much they can integrate onto the mainboard, it's still going over the same bus -- the only difference is that the connections are etched onto the board instead of having card slots.

        Furthermore, bundling peripherals onto a mainboard is exactly as bad as bundling web browsers and such into an operating system: it's harder to choose solutions from other vendors even if they're better suited to your needs, you're paying for features you may never use or need, there's no incentive for the hardware company NOT to cut corners and put the cheapest shite on there that they can find.

        The beauty of the x86 PC architecture, if any, is the extreme modularity. I hope that this feature of the design doesn't get eroded away by increasing levels of device integration, and a stronger, faster PCI spec can help a lot towards retaining openness and modularity.
    • I guess I don't understand what PCI Express does. If it's just more bandwidth, how could it make any difference? There's no game that saturates even AGP4x; why would you need something even faster than 8x? 8x will be sufficient for games a generation newer than Doom 3, I'd bet; why should we get excited about a standard that won't affect anyone for five years?
      • Re:What's the point? (Score:2, Interesting)

        by vofka (572268)
        Sure, one GFX card may not saturate the Bus, but what about more than one?

        Imagine having a board with several PCI Express slots - put a Good Graphics Card in, say, each of 3 slots, and multihead your games :)

        Also, if I understand it correctly, PCI Express is an upgrade to / replacement for PCI. Sure, it allows high-bandwidth comm to a Graphics Adapter, but also to SCSI/IDE Controllers, NIC's, Video Capture Hardware etc, etc.
    • I think we can be confident that both ATI and nVidia know about the PCI Express spec and both companies will introduce graphics card design based on ATI's R350/R400 GPU's and nVidia NV3X GPU's by the fall of 2004.
  • Big deal (Score:5, Insightful)

    by Jason1729 (561790) on Friday February 28, 2003 @12:41AM (#5403848)
    so get ready to toss out your motherboard

    Since when can you upgrade to a new generation CPU and not have to replace the motherboard?

    Jason
    ProfQuotes [profquotes.com]
    • Re:Big deal (Score:3, Insightful)

      by kfg (145172)
      I have to toss out the motherboard pretty much every time I buy a cpu, which makes the "quick change" feature of this unit rather moot.

      If they're going to hold the basic architecture steady for at least a few years this is going to be quite handy, but if each iteration is going to require a general upgrade to properly utilize the new speed and features anyway. . .big deal.

      The motherboard manufacturers like to see a steady upgrade cycle too you know and it almost always comes down to "gut the case" and hope a few cards are still usable.

      KFG
      • Since my MB and case are AT's, I have to replace the case as well. The one constant in all my upgrading is the floppy drive; it originally came with my 486.
        • I replaced my floppy drives with LS-120 Superdrives several years ago. The latest PC I built doesn't even have that; just a CD-RW and a NIC. I'd rather the MB makers eliminate the floppy interface than the PCI.. Does anything still come with a floppy? I haven't seen a new one in ages.

          Vat ees thees "fla-pee" you speek off?
    • I take it you have never owned an upgraded Macintosh. Almost every PowerMac can be upgraded [sonnettech.com] to at least a G3 (but most can go up to a G4) without changing the motherboard.

      Heh, and people say that Macs can't be upgraded. Sheesh.
  • Why NewCard? (Score:4, Interesting)

    by KiahZero (610862) on Friday February 28, 2003 @12:43AM (#5403861)
    I don't understand why revamped PC-cards are being pushed for desktop computing. I can understand increasing the bus speed on PCI cards (faster real-time TV encoding... yay!), but why does this need to happen in cards the size of two quarters?

    Is the goal to make it so that users with two PCs can carry peripherals from one computer to the other? I would also hope that there will be legacy ports. I'm not planning on buying a new chip for a while, but I really don't feel like having to buy brand new hardware when I do. I'll have to buy a new video card (no AGP port), but they could at least put a few standard PCI ports on the mobo so I could slap in my more expensive expansion cards.
    • but they could at least put a few standard PCI ports on the mobo so I could slap in my more expensive expansion cards.

      If you have expensive expansion cards, you'll be able to find PCI motherboards for a long, long, long time.

      Unless, of course, you think that a $150 sound card is "expensive." Then, you're fucked.
    • Re:Why NewCard? (Score:4, Informative)

      by Apple Acolyte (517892) on Friday February 28, 2003 @01:05AM (#5403952)
      KiahZero opined: I don't understand why revamped PC-cards are being pushed for desktop computing.

      The parent makes it seem as if PCI Express only defines the standard for new generation PC Cards. It doesn't simply do that; otherwise it would be nearly worthless as the next generation successor to PCI. Take a look at relevant quotes from the PCI-SIG [pcisig.com]:

      The "Mini PCI Express Electromechanical specification, an alternate for the existing Mini PCI form factor specification, is being completed for membership review and is expected to be finalized for publication in the first quarter of 2003."

      "IBM is excited about the PCI Express architecture because of its compatibility with the past and its high-bandwidth options for the future," said Peter Hortensius, Vice President of Development, IBM Personal Computing Division. "IBM embraces open industry standards and provides innovation on top of them, and PCI Express presents outstanding opportunities for solving real customer problems."

      Mini-PCI Express, then, is a spec in its infancy that is designed to replace the previous generation PC Card. It should make future laptops far more expandable, which is a great thing. And PCI Express is one of multiple candidates for desktop expansion. Yet, it seems that PCI Express is going to be backward compatible with important specs, and that it seemingly has industry support. I just wanted to make sure everyone understands the PCI card isn't going to be replaced by the PC Card.

  • by Anonymous Coward
    Arghahghaghaaa
  • by craenor (623901) on Friday February 28, 2003 @12:47AM (#5403883) Homepage
    If you are thinking about making a Portable PC Purchase and are looking for either a performance or "road warrior" system...just wait a bit.

    The ones I've been playing with at work just absolutely rock. You can clearly see the difference that 1mb L2 cache makes...and combined with systems that already have decent battery life you won't have to worry about whether or not you'll be able to finish the Braveheart DVD on battery power.

    Craenor
  • by tommy (12973) on Friday February 28, 2003 @12:48AM (#5403888) Homepage
    I beg to differ. My 10MHz Intel 286 had no pins. It looked like this [tcocd.de].
  • by Low2000 (606536)
    I'm all for improving hardware but... Why would this be done other then to foce people to buy new hardware? Is the current PCI spec so bad?

    I just see this happening.

    Hey. So you want a new sound card? Great! What? You only have regular PCI? I'm sorry we only have it in PCI Express. No worries. We offer this brand new Intel board and chip and ram that will solve your problem. Only $1,200!

    What am I missing? I hope I'm missing somthing =/
    • by Anonymous Coward
      Basically, there is no need for more speed from systems, I agree with your point there. I suspect this is a case of moving forwards for the sake of moving forwards. There are very few REAL applications that require anywhere near 3Ghz. Currently, even top line servers are rarely over the 1.xGhz mark, going in favour of more CPUs, something that can be done more easily and less wastefully with current chipsets. Compatibility is also kept.

      Honestly, a PC with eight $20 CPUs would end up far more responsive and just as useful for every task than one with one single several-hundred-dollar chip
      • One Fast CPU is always going to have an advantage over multiple slower CPUs. It takes a lot of bookkeeping in the background to assign different tasks to different CPUs. Not to mention programs need to be written multi-threaded to take advantage of another processor.
        • It takes a lot of bookkeeping in the background to assign different tasks to different CPUs. Not to mention programs need to be written multi-threaded to take advantage of another processor.

          There are lots of ways of using multiple processors for computations; threads is only one of them, and probably among the most cumbersome.

          One Fast CPU is always going to have an advantage over multiple slower CPUs.

          If that were true, the fastest computers would be like the supercomputers of the past. But multi-processor designs and compute clusters have pretty much killed the supercomputers of the past.

        • There a certainly tradeoffs, but it's not a true statement.

          It's worth paying for a more complex design, etc., to get a faster unitary CPU instead of two slower ones, for the bookkeeping reasons that you point out, but not worth paying an unlimited amount more. OTOH, simple parallelization schemes have their limits also.

          My view of where we are headed is:
          1) CPUs will advance to some optimum level of power.
          2) Clusters of CPUs will increase the power of the individuals.
          3) Clusters will be linked along a fast bus, with one CPU out of each cluster attached to the bus. (This cpu is effectively a member of two separate clusters, but one of the clusters does practically nothing but manage communications.
          4) A node at one end of the bus will be linked into an orthogonal bus, which will contain similar nodes...

          Now this is just a design for a maximally compute intensive processor. At each step you must pay additional overhead, so you would be better off it the problem could be addressed by a system one level simpler. But if you can't...

          At this point we come up against the issue of "but how do you USE it!?" This system will probably not be effective until compilers, or possibly interpreters or VMs can automatically partition problems and assign the pieces to various chunks. This will probably require a message-passing operating system. (Not too unreasonable. I think that Linux could be adapted into one.) But, e.g., when a job wants to open a file, it wouldn't need to know where the disk was, it would just send out a message asking for the file. This is like the separation between files and devices, so the basic layers are already in the design. File permissions would need to include lock status, though, or else file access would need to be managed by a dedicated cpu (which could do it in about the current manner). Etc.

          There's lots of details that will need to be thrashed out, and the design isn't going to happen this year, or the next. (Probably.)

          At some point, all higher levels would get turned over to a TCP/IP like connection set, slightly redesigned to optimize it for use within one connected computer system. (Probably just a matter of adding some additional protocols for internal use that are more efficient over the known network configuration, and are less concerned with security, but which will only talk internally. [And which have their own limits, so that a virus can't spread unchecked.])

          I see all external communication occuring over standard TCP/IP, and possibly even using only a subset of the standard protocols. (N.B.: I'm talking about limiting protocols, not ports. Think of it as a sanitation measure. You want the system to be able to evaluate incoming communications, and react sensibly. If someone tells it "Drop dead!", it should decide not to obey the literal, or even the figurative, interpretation, but rather to consider that this is an expression of frustration. And this should be true even if the remote user is "root". Some commands should require local access.)

          N.B.: I talked as if this were a strictly hierarchical system, and, to a large extent it would be. But it should take advantage of sideways links also. This would be largely for error recovery, as the high speed communications would be hierarchical. But it should enable reconfiguration and recovery (and diagnosis) in case of hardware errors. (Think of the cell system for underground agents.)
      • Have you ever tried compiling something in Linux. Even on a 2GHz P4, KDE takes 8 hours to do a full build. Renders still take a couple of hours. Even some simple video encoding can take several minutes. Before making overly general comments like "there is no need for more speed from systems" try actually *doing* something with your system.

        Also, there is no such thing as a $20 CPU. The only reason you can buy them on PriceWatch for $20 is because they're subsidized by sales of the $400 CPUs. And a PC with eight $20 CPUs would be dead slow. Most code is not that parallizable, and any interconnect that can handle 8 CPUs is going to be a whole lot more than a few hundred dollars.
    • I can see this with video cards, but sound cards? When was the last time anyone bothered to upgrade their sound card? You have a point, but this is not going to be an issue very often. 99%+ of the users out there, even the ones who tinker with their systems, never even think about upgrading their sound card. Most don't even upgrade their speakers.

  • transemta crusoe? (Score:4, Informative)

    by TerraFrost (611855) on Friday February 28, 2003 @12:53AM (#5403909)
    it'll be interesting to see how the "Tejas" processor compares to the Transmeta Crusoe - currently, the least power intensive x86 compatable cpu out there...

    also... if you're currious about PCI Express, this link seems to be pretty... informative:
    http://www.intel.com/technology/pciexpress/

    and is anyone else disappointed that the new "Grantsdale" chipset isn't supporting rambus ram!? i know i am :(

  • And Floppies too!
  • Oh Yeah! (Score:3, Funny)

    by Waffle Iron (339739) on Friday February 28, 2003 @12:59AM (#5403932)
    check out the waffle iron socket!

    I'm stoked. I'm going to pull in some serious coin on this deal.

    Every socket designer dreams about being chosen to do a major Intel processor. It doesn't get any bigger than this, baby!

  • Bloody tricks! (Score:4, Insightful)

    by HoneyBunchesOfGoats (619017) on Friday February 28, 2003 @01:00AM (#5403935)
    From the second page of the article on Banias:
    EBL encompasses four features: a non-synchronous vertical refresh control for displays, which will save between 200 and 800 milliwatts; an optimized LCD inverter design, expected to cut an additional 550 mW to 880mW; device performance state monitoring, which will reduce overall power by an additional 900 mW; and a device power profile utility tool, designed to monitor the power of all devices in the system.
    So basically, their "über-cool" power-saving processor isn't what's really saving power, but a bunch of other little tricks in the rest of the system. All they did with the processor was take a PIII-M, ramp up the frequency, and slap on a bigger cache.

    Not that I hold this against them or anything; if in the end it increases battery life, that's a Good Thing. I just wish they wouldn't hype up their new processor as being so great, when really there isn't much more improvement over the PIII.
  • Nice (Score:2, Insightful)

    by pheared (446683)
    <rant>

    I can't wait until I have no choice but to buy some hardware that's not compatible with anything I might possibly already own. What's even cooler is that I get to do my part and add my obsoleted hardware to our local dump.

    P.S.: It would be nice to get the computing companies to do a bit more in the way of reuse. I don't think it's a good idea to use until there's no more, and then just move on to a new resource.

    </rant>
    • Re:Nice (Score:3, Insightful)

      by cheezedawg (413482)
      So are you suggesting that the industry should ignore the increasing system bandwidth needs (gigabit ethernet, USB2, 1394, Infiniband, SATA, etc) just so you can use your old hardware for longer?

      And your hardware will be compatible for years to come. Legacy interfaces linger for a LOOOONG time.
  • PCI Express FAQ (Score:5, Informative)

    by Anonymous Coward on Friday February 28, 2003 @01:08AM (#5403968)
    PCI Express FAQ here. [pcisig.com]

    Quick summary: Formerly known as 3GIO, Software compatibility. Point-to-point instead of bus. 1 to 32 bits wide @ 2Gbps per bit = 16 GB/sec max (vs. 1-4 GB/sec for regular PCI; this is about AGP16X)
  • Joy of joys (Score:4, Interesting)

    by buffer-overflowed (588867) on Friday February 28, 2003 @01:11AM (#5403982) Journal
    Even more stuff that as someone who uses computers primarily for work, I don't need.

    Sure it looks good, yea, I'm all exited about a "new era of computing," but it breaks backwards compatibility with all of my old stuff and I bet it still can't outperform the mainframe I program on now in terms of raw MIPS.

    Why did we ever move to PC's from thin clients in the first place? We have consoles for gaming, windows for PC gaming, and *nix for serious work (try doing something else under say Solaris, and posting to slashdot doesn't count.) now. Why all of the redundancy? Aren't we in an economic downturn? The bus speeds and improvements are nice, don't get me wrong... but in a PC? It removes the PCI bottleneck problem, but I don't see where it removes the HDD bottleneck in terms of raw speed.

    All in all i'd say it's a nifty gadget.

    When we get holographic/full immersion, give me a call. I'd love to see what my brain can output in raw source without needing to actually type.

    --I'm just continuing my tradition of posting drunk, pay me no head. Don't post to slashdot under wine.
    • Re:Joy of joys (Score:4, Interesting)

      by be-fan (61476) on Friday February 28, 2003 @02:18AM (#5404232)
      Grognard. It's not 1992 anymore and PC's aren't lumbering beasts.

      Thin clients: How are people going to use this at home? Over their 28.8 dialup connection? With the work I do, I can peg pretty much anything you throw at me. You think they want user's like me on shared systems? You think I want other users slowing me down?

      PC architecture: A modern PC has more resources than most RISC workstations that are 5x the price. Ever since the P4 came out, PC memory bandwidth (one of it's traditional weak points) has skyrocketed. By the end of the year, it will be up at 6.4 GB/sec, which is an impressive number even for an SGI or Sun machine.

      Bottlenecks: What do you do where the HDD is the bottleneck? After an hour of use or so, my Linux system pretty much runs out of RAM. On workstation tasks, the HDD is often not the benchmark. It's not the benchmark for the 3D rendering I do, the scientific sims, the gaming, the programming, pretty much everything. In fact, I thought it was going to really suck moving to a P4 laptop, because of the slow 4200 RPM hard drive. Ever since I put 640MB of RAM in there, I don't noticed any slowdown at all.
      • Re:Joy of joys (Score:3, Interesting)

        by MtViewGuy (197597)
        I think PC architecture is going to undergo some drastic speed improvements over the next 24 months--and that's not including the CPU.

        Between faster chipsets, big increases in memory bandwidth (PC3200 DDR-SDRAM is only the beginning), and Serial ATA, you'll see overall faster computers anyway.
    • Why did we ever move to PC's from thin clients in the first place?

      Trust me...you don't want to be working on the same mainframe as me when I'm editing/rendering huge 3D animations :)
    • did you mean "it removes the PCI barrelneck but doesn't remove the HDD bottleneck"? 8-)
    • Re:Joy of joys (Score:4, Insightful)

      by jejones (115979) on Friday February 28, 2003 @08:27AM (#5405133) Journal
      [PCI Express] removes the PCI bottleneck problem, but I don't see where it removes the HDD bottleneck in terms of raw speed.

      I thought that was what serial ATA [explosivelabs.com] is supposed to do.

      • SATA, higher areal densities (Maxtor has 320GB+ drives planned for release very soon) or 10,000 RPM SATA drives will pretty much eliminate the bottleneck at the hard disk.

        Modern ATA drives are already pushing 25-50 MB/Sec, doubling the capacity also doubles this sustained rate. Within a year we will likely see single drives that are able to max out the initial 150MB/Sec SATA spec.

        I think we should be more concerned that most people are still on 100Mbit LANs. Duplicating a single 200GB hard disk over 100Mbit LAN takes ages. Gigabit is still pretty expensive, compared to how extremely cheap 100Mbit is. An example, we checked into adding a 8 port copper Gbit card to our Cisco at work, and I think the figure was something like $5000. We have a few fiber Gbit slots, but just the GBIC for those costs around $400 a port.

        I just have the feeling that by the time Gbit becomes cheap enough for the masses (i.e. someone is heard saying "we might as well go Gbit, it's not much more expensive"), it will already be obselete.

        Not to mention that there are a whole lot of crappy Gbit NICs that can do no where near Gbit speeds.
    • breaks backwards compatibility with all of my old stuff

      Whine whine whine. Have you looked at a modern system? Unless you have specialized needs you probably have one card - the video card. Everything else - sound, USB, firewire, network, HD, other I/O - is on the motherboard. The most common add-ons are a TV encoder/decoder (although increasingly becoming part of the video card) or a SCSI controller - which will certainly be available for PCI-X (and you'd want to move anyway to get the additional bandwidth).

      Oh, you have a specialized card available only on <insert ancient standard here>? Guess what - specialized means costs more. Get over it. And I bet you'll still pay less than anything else on the market.

      I bet it still can't outperform the mainframe I program on now in terms of raw MIPS

      I bet you're wrong. A top line PC often outperforms everything else on MIPS per CPU basis. The issue is that the bus sucks and it can't do anywhere near as many transactions or I/Os as a high-end workstation (Sun, MIPS, etc), much less a mainframe.

      Of course, PCI-X helps address this bus issue - it doesn't solve it, but it'll go a long way toward improving the problem. And will narrow the gap between PC and mini/mainframe performance even further.

      Why did we ever move to PC's from thin clients in the first place?

      Besides, not every problem needs the throughput of a mainframe or even a Sun class box. Why spend several hundred thousand or a couple million on a box when you can get the job done with a $1000 PC? What was that you were babbling on about regarding an economic downturn?
  • by bluesoul88 (609555) <bluesoul@ t h elegendofmax.com> on Friday February 28, 2003 @01:12AM (#5403990) Homepage
    I just hope it ain't the short bus. I hate getting on that friggin' thing, and I don't do well with ramps on a friggin' bus.
  • by bushboy (112290)
    It would be great to have an idea of the cost implications of the new chipset - the fact that you'll need a new motherboard, CPU and graphics card means this setup isn't going to be for the masses right away.
    The price to upgrade could easily reach $1200 US for early adopters.

    I don't see much of a problem with the PCI slots as the majority of current modern systems have a lot of components onboard already, such as LAN, Sound, Video etc.
    I guess the safe bet is they'll include 2xPCI slots which should be enough for most peoples purposes.
  • Grantsdale? (Score:5, Funny)

    by Anonymous Coward on Friday February 28, 2003 @01:51AM (#5404124)
    Feh. Anybody who's seen intel's roadmap (as I have) knows the Grantsdale chip is just a stepping stone.

    Personally, I'm waiting for the Higgenbotham chips in early 2005. After that, the Ranmatheau chips. In earlier 2007, expect amazing performance from the Cleodranvier chipset.

    2008 brings us the amazing 10-GHz Hefnestranthellhaller chipset, and 2009 unveils Intel's most impressive chip: the Quackenbush.

    But the true surprise comes in 2010, when the world experiences the amazing speed of the Gentrecktagazunt.

    Truly wonderous times ahead.
  • by mesach (191869) on Friday February 28, 2003 @02:00AM (#5404154)
    there were pins on the slot1's?
    • there were pins on the slot1's?

      Yes.

      What, you didn't see the processor on the center of that board? With the pins soldered to the board?

      Even so, the statment is incorrect, since there have been previous CPUs that were pinless - such as some revisions of the 80286.

      A CPU this complex without pins is a pretty nifty thing though.
  • by rainwalker (174354) on Friday February 28, 2003 @02:30AM (#5404284)
    From the article, ..."Granite Peak" initiative, which limits the number of driver revisions to one every six months, making the launch of each new chipset even more significant.

    So, what exactly does this mean? If I have a problem with Intel's drivers that, say, prevents my machine from booting (not that THAT has ever happened) I have to wait 6 months for the next revision? I don't understand what driver revision schedules have to do with product release cycles.

    Also from the article: "...[people buying] the latest GeForce card near the end of this year, when six months later it won't work [fit] inside a new PC?"

    This is a non-issue for most people, I think. Those people who buy new video cards every six months (you know who you are) aren't really going to balk at replacing motherboard, CPU, and video card all at the same time, if it yields a 25% performance improvement (or more). At the other end of the scale are people who upgrade video cards by buying a new Dell (or whatever), for whom this is also not an issue. Those of us in the middle just won't buy a new motherboard/CPU until we can afford to replace the whole shebang anyway. Once we do, we will most likely build a whole new machine.

    Anyway, it's not like nVidia and ATI are going to stop making AGP cards; I'm sure that both connections will be supported. If you look around, you can still get PCI versions of most cards on the market (shudder).
    • Don't forget that since you can't have multiple AGP cards in a machine, there is a darn good reason to have PCI cards around. Multi-head display adapters alleviate the problem some, but it's still nice to be able to have, say, four displays on one PC.
  • by Thagg (9904) <thadbeier@gmail.com> on Friday February 28, 2003 @03:23AM (#5404436) Journal
    The article mentions that Intel may do away with the USB ports in the Grantsdale systems, that PCI express may get rid of USB entirely -- but if it does have USB it will have at least 8 ports.

    OK, that's pretty weird. But why would they get rid of a popular, reasonably high-performance, and cheap interface like USB? Is Firewire 800 going to take it's place? SATA? Is everything going to be wireless?

    thad
  • by master_p (608214) on Friday February 28, 2003 @06:16AM (#5404802)
    Sure, I need faster speed (Doom III comes to mind; and a whole load of cinema-quality games-imagine Half Life II for example, or Duke Nukem For Ever), but there are some other things that bother me:

    1) the PC BIOS!!! for how long should we tolerate the shitty 16-bit PC BIOSes ? I mean, in the days of PCI-X and 800MHz memory buses, the PC's BIOS is still 16-bit and operating systems need to perform wild tricks to boot.

    2) the partioning scheme. Only 4 partitions!!!! this is an artifact from the days of the original PC.

    Ok, not so important but irritating nevertheless.

    • > 2) the partioning scheme. Only 4 partitions!!!! this is an artifact from the days of the original PC.

      You can definitely have more than 4 partitions. I think
      2 primary plus 8 logical is the maximum (total of 10).
      It would be nice to have an arbitrary number of partitions
      but this is not very important in my opinion.

      Regarding the BIOS, I have to say that I like it the way it
      is. Complex things fail and BIOS should be 99.999% error
      free. I don't see the need for a themeable GUI-Bios with
      transparent windows. If it works, don't break it.

      P.
  • What's Obsolete? (Score:4, Insightful)

    by maizena (640458) on Friday February 28, 2003 @06:44AM (#5404873)
    No matter how long you wait, the day after you buy/upgrade your PC it will be already obsolete.

    We shall not forget that, as any other enterprise, Intel's business is to make MONEY. Cutting edge technology is just a plus...

    It's in their best interest to push forward the their latest family of products. This is how Intel works and obsolescence is carefully planed by them.

    It's up to us, as consumers, to set the pace and not get swept by the low-tech fears. An upgrade is really only necessary when your PC performance gets in the way of you doing your usual tasks.

    Therefore, we must keep in our minds that obsolescence is dictated by our needs, not by theirs.
  • your old stuff (Score:2, Informative)

    by Ratchet (79516)
    There's an PCI-X to AGP bridge that vendors can take advantage of to offer AGP ports on PCI-Express motherboards, so you'll likely be able to hang on to that new $500 vidcard you just bought (not that you'll be seeing PCIX anytime soon mind you). Your "old" PCI devices should still work as well.
  • by adzoox (615327) on Friday February 28, 2003 @07:36AM (#5404985) Journal
    Who exactly is in the PR/Graphic/Ad Design Dept at Intel? I think the only markettable ideas Intel has ever come out with were the names Pentium and little tone at the end of commercials. The names they come out with are very rarely tongue rolling or memorable. Same goes for their commercials. I mean, one campaign has Homer Simpson getting his brain replaced by a Pentium. The next campaign has aliens using it. The first tells me that stupid people use the Pentium, the second tells me that only aliens think a Pentium is cool,

    Not being Apple bias, but you have to hand it to Apple Computer's PR/Design/Ad/Graphic Design Departments. They even get press for what they name variations of the operating system. It's not goofy either. Jaguar and all the promotional material has spawned the entire design industry into using animal prints, especially Jaguar.

  • However? (Score:2, Insightful)

    by AyeRoxor! (471669)
    "Tejas uses a 775-contact pinless Land Grid Array (LGA) that far exceeds the 478 pins used on the Pentium 4, and Prescott. However, the additional pins were required for the additional I/O and power requirements of Tejas, the documents say."

    Ohh, for I/O and power ! I thought they would be used for..... umm... Well, that's about all they can be used for. Why does that sentence begin with "However"?
  • by dpilot (134227) on Friday February 28, 2003 @08:06AM (#5405049) Homepage Journal
    As others have said, so what if a new motherboard is needed - they're obsolete about as fast as a CPU chip, anyway. Another post indicates that PCI eXpress is a reasonably open standard.

    But the IP/lock-in aspects still bother me. Intel behaved like a spanked puppy for a few years after their Rambus fiasco, but lately they seem to be back at those games, again.

    They've taken steps to ensure that Banias/Centrino only sells with their chipset. It's only a logo program, but it probably carries a heavy enough advertising kickback behind it to have the force of law.
    The Itanium is *the most proprietary* CPU on the planet, or at least a contender for the crown. No second sources, no cross-licensing on any of the IP.

    So in this light, anyone want to bet that Tejas is not tied to Grantsdale?

    Assuming it is, the net effects are questionable. It appears that Intel is driving compatibility away from the CPU pins, and out to the motherboard plug interface. I seriously doubt they have the capability to push it any further than that. In the long run, this probably opens the market niche for AMD and Via C3, because it's closing the market for low-cost chipset providers to service Intel CPUs.
  • Nope. I distinctly recall the Intel 80186 in an old wang I had once that had no pins, just a bunch of contacts.
  • Throw away your video card? Throw away your other slotted peripherals? Replace them with something that uses an Intel-proprietary bus?

    Didn't IBM try this a decade and a half ago? Intel needs to read up on something called Micro Channel and learn why it didn't work for IBM, and it won't work for them either.

    Intel needs to tread carefully. They may be Chipzilla today, but something like this could be the turning point, like Micro Channel was for IBM, where they turned the screws a little too tight and the customers fled to something more open.
  • This is dumb (Score:2, Interesting)

    by Anonymous Coward
    The ultimate in CPU packaging will be RF, that is, one high bandwidth interconnect using a very dense modulation scheme like 1024 QAM. The bus of the motherboard will be Infiniband.
    The CPU will look like old style ceramic power triodes, with a built-in bonded heat sink. There will be two low inductance connections for power, and a hard line SMA connector for everything else.
  • My Pentium III-M 600-MHz notebook in performance-optimized mode doesn't have the kick to play a DVD

    I've played DVDs on Pentium II 300 MHz laptops without difficutly....what's this guys problem?

    Worse still, Brookwood said, was when he takes his notebook to the local java joint. "I just want to sit there for a couple of hours"

    A couple of hours? Don't you have anything better to do? Go get a job.

    whopping" 316 minutes of battery life using a Centrino notebook"

    I'll belive it when I see it. The two biggest draws of power in a notebook are the LCD backlight and the hard-drive. "Centrino" doesn't change either of those.

    Don't believe the hype.

    -ted
  • no pins (Score:3, Informative)

    by chunkwhite86 (593696) on Friday February 28, 2003 @11:24AM (#5406612)
    This "waffle iron" design which lacks pins is old news to the DEC/Compaq/HP Alpha processor team. They have been using this packaging for almost a year now in production systems.
  • by default luser (529332) on Friday February 28, 2003 @11:33AM (#5406689) Journal
    Did I read that right? 250MB/s?

    PCI spec is 133MB/s, which is hardly a marked improvement. 16-bit ISA, by contrast, was barely 16MB/s.

    If I am to believe the theoretical numbers for AGP, then PCI Express as a graphics bus makes even less sense:

    AGP 1x = 264MB/s ( 66.6 mHz, 64-bit )
    AGP 2x = 528MB/s ( 66.6 mHz rise and fall, 64-bit )
    AGP 4x ~ 1GB/s ( 66.6 mHz - 4 strobe, 64-bit )
    AGP 8x ... you get the picture.

    What in the hell do we need a PCI replacement for that has zero potential for handling enormous video bandwidth as well as or better than AGP?

    What in hell do we need a PCI replacement for that doesn't even utilize the PCI-X or 64-bit, 66MHz PCI already being pushed for servers? Not to mention that fact that any device that can push the bandwidth of PCI is already available in one of the above formats, who wants to build yet another model for PCI Express?

    Honestly, if you need to find emerging technologies, just look to the server path. Intel has always been about trickle-down, this move doesn't make any sense.
    • I think you misinterpreted the numbers. PCI-X is a point-to-point, rather than bus technology. A connection is 1 to 32 bits wide @ 2Gbps per bit.

      So if you're using a measly 2 bit connection sure you might only get 250 MB a second, but if you read the article you'll see that Intel is planning on using a 16-bit wide connection for the graphics card. This would give you 32Gbps, or roughly the same as AGP 8X. There is potential to go much faster by using a 32-bit PCI-X connection. In comparison, the other PCI-X slots on the motherboard for peripherals will be much slower, probably only using 2 bits.

      I fully expect server boards to have multiple 32-bit PCI-X slots for maximum I/O throughput. In addition, we will probably see new technologies for clustering that utilize a PCI-X expansion card as a high speed server-to-server bus.
  • by heroine (1220) on Friday February 28, 2003 @01:21PM (#5407641) Homepage
    It's amazing that CPU's haven't increased performance on a clock for clock basis since 1997. Imagine if modern CPU's really ran 3000 times faster than a 6502.

The key elements in human thinking are not numbers but labels of fuzzy sets. -- L. Zadeh

Working...