Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. ×
Intel

Intel Chips For The Near- And Semi-Near Future 105

Brian writes "This article reports that Intel will release new chips at the Comdex trade show, its first low-power designs for super-thin servers. The new Pentium III model is a gussied-up chip taken from the company's product line for portable computers, which share many of the same constraints as ultradense servers. These systems can't consume as much power or give off as much heat as ordinary CPUs because overheating causes processing errors. The systems are the first swing of a one-two punch against Transmeta, whose low-power designs caught Intel flat-footed, first in the mobile market and then in the low-power server market. Intel now is fighting back just when most server companies using Transmeta chips are on the ropes." And albat0r writes: "Intel says that it will hit 3GHz on the mainstream Pentium 4 by the end of 2002. Intel will advance its Celeron line, currently based on Pentium III technology, with Pentium 4 technology by mid-2002." I look forward to good values on eBay when 2GHz is "obsolete."
This discussion has been archived. No new comments can be posted.

Intel Chips For The Near- And Semi-Near Future

Comments Filter:
  • Intel performance (Score:1, Redundant)

    by Spootnik ( 518145 )
    Their current roadmap still has them at 2.4Ghz next summer. At that rate, I seriously doubt they will have a 3Ghz out before the end of the year. I bet 3Ghz isnt released until some time near the end of the first half of 2003.

    Intel demonstrated a Northwood P4 running at 3Ghz with _supercooling_. (it actually got up to 3.5Ghz, but the demos were run at 3Ghz). Who the hell told you Intel had a P4 running at 5Ghz? Yeah, they could probably make it to 10Ghz within 3 years pretty easily if they use the same design concept behind the current P4. Maybe they will increase the pipeline to 40 stages and get there even faster!

    Oh boy.. Clock speed isn't everything. The P4 architecture is brilliant for a company trying to sell their CPUs to people like you. The chips are a hell of a lot easier to market when people just look at the clock speed. Rambus has little to do with being able to run between 3 and 10Ghz. Intel just doesnt want to admit that after spending all this time doing nothing but saying RDRAM is the only way.

    AMD already has .13 capability and the new SOI technology to go with it. They will be sampling Thoroughbred processors next quarter.
    • Somebody correct me if I'm getting my dates wrong here, but...

      It strikes me that the 3Ghz should be out sooner than the end of next year. It's been a couple months since the 2GHz was out, and so the total time in between there would be somewhere around a year, a year and three months.

      The leap from 1Ghz to 2Ghz took considerably less time than that. I know we're all sick of hearing about the widely-misunderstood Moore's Law, but shouldn't somebody out there be screaming bloody murder at this, that it should be out much sooner, that Intel is going to cave in, etc.?

      When AMD and Intel first hit the Ghz mark, they both announced that they were going to slow down their schedules, so they weren't left with a bunch of 600Mhz chips laying around while everybody wanted a shiny new 1.xGhz in their box. But there's shortages everywhere right now, so we know that Intel probably doesn't have a warehouse full of unsold P4s somewhere.

      Just pointing it out. Maybe we're all getting a bit too spoiled when it comes to speed. Anybody know what's up here?
      • The leap from 1Ghz to 2Ghz took considerably less time than that. I know we're all sick of hearing about the widely-misunderstood Moore's Law, but shouldn't somebody out there be screaming bloody murder at this, that it should be out much sooner, that Intel is going to cave in, etc.?

        Once again... Moores Law talks about the increase of transistors on a chip, not the speed of the chip itself. So you can not use it to reason that speeds will increase!!!!!
        • by Anonymous Coward
          Moore's Corollary of the Intel Marketing Department:

          Intel will double perceptual speed (read clock) every 18 months in order to stimulate upgrade purchases.
        • For the record...

          I know that Moore's Law talks about the increase of transistors on a chip, not necessarily performance and so on. If you look at my original post, I referred to it as "the widely-misunderstood Moore's Law" for a reason.

          I know this. You know this. Just about everybody reading on Slashdot, I'll bet, knows this. But it seems that the general populace that reads about Moore's Law out of Parade Magazine or USA Today probably doesn't know anything about transistors, let alone the way a processor is built, and so when some half-assed journalist writes about doubling transistors-- if he or she gets it correct even then-- John Q. Public at home is going to think "Double the transistors? Well, that must mean it's twice as fast."

          I didn't mean that the P4 should be twice as fast in 18 months. I meant that the general perception out there, thanks to the aforementioned uninformed-but-meaning-well journalists, is that it probably should be. So that's where we'd hear the complaining. (If there is any. Note, again, my original post, when I said "Correct me if I'm wrong...")

          C'mon, we all read Slashdot. I thought it was just a given that we think we're smarter than everybody else out there that doesn't.
      • Perhaps the only thing that Intel can reall do with their P4 CPU line is slap a "We were the first ones to reach X Ghz barrier" in order to sell their CPUs to the ignorant masses. There is nothing wrong with the P4 really, but in most cases, the slower (clock speed) AMD CPU is nearly always a better choice. How does AMD counter this marketing ploy from Intel? Why, they pull the old Cyrix PR idea. AMD has become a very resourceful company. I hope that they have a better idea in their bag of tricks.
    • 3Ghz...dang..if anyone remembers when AMD hit the 1Ghz mark first you might get this:

      So, Intel...Is that going to be a paper release or actual silicon?

      Oh, well, if nothing else we will certainly get another lively discussion on the Mhz/Ghz wars, speed vs HP and such.

      Yes I use a mac. Yes, I'm an NT admin. Yes, I realize the irony. And...yes, I'm keeping my mouth shut for the remainder of this discussion.

      Fair enuf?
  • In other, slightly unrelated chip news, ZDNet reports that motherboards with the new nVidia nForce chipset will hit the market next week [zdnet.com]. Boy howdy do I want me some of that!
    • The word on AnandTech [anandtech.com] is that the nForce motherboards won't be that great. Read the article NVIDIA's nForce 420/220: It's Finally Here [anandtech.com]

      If you do get one, get a TBird 1.4 to put in it and save your self some cash. The Palimino Chips (Athlon MP, Athlon XP) have improved pre-built in fetching logic - which negates the DASP on the nForce.

      The built-in graphics is the GeForce 2 MX - nothing spectacular.

      Where the nForce does have an edge, though is the built-in sound - it should rival the Creative Labs Soundblaster Audigy. I'd love to see a good review or comparison of which one sounds better with 5.1 sound.

    • Interestingly, I've just read a review of the first system (to my knowledge) that uses the nForce board as it's core. It's from Mesh [meshcomputers.com] and was reviewed in the December issue of Personal Computer World [pcw.com] in the UK. The review slateted it quite badly, saying that it's 3D performance was down even on Budget versions of the GeForce 2 card on which the gfx engine on the board is based. Mesh also seemed not to have bothered wiring up the cool onboard sound system the nForce carries. My advice : Wait a while folks - the first nForce systems are going to take a while to run really swish!

  • What?

    Are you currently searching EBay for that "steal" on a 486-DX4-100?
    • Are you currently searching EBay for that "steal" on a 486-DX4-100?

      At one time the difference between, say, a 386-33 and 486-66 were astounding, in terms of *feel*. But a few years ago I used NT on the job running on a 200MHz Pentium. Today I use an 866MHz Pentium III and it feels about the same. Compiles are faster, sure, games run better, yes, but it's not astounding. If I upgraded to a 1.4 GHz processor I doubt it would even matter to me.

      So I can profit from silly people who think that "1.4 GHz is slow" and constantly have to upgrade to whatever comes along next. The rest of us, the people doing actual work, have given up caring about CPU speed.
      • The rest of us, the people doing actual work, have given up caring about CPU speed.
        Define "actual work." Those of us who work with digital video (editing, compression, or whatever) will take all the speed we can get.

        That said, even if Intel does produce a 3-GHz processor, what will its real-world performance be like? We know their current 2-GHz P4 runs no faster than an Athlon running somewhere between 1.4 and 1.5 GHz, and I doubt that AMD will be standing still for the next year. It's also worth remembering which company actually had 1-GHz processors to sell and which company was only able to announce future availability.

  • Wall Clock time (Score:5, Insightful)

    by Anonymous Coward on Friday November 02, 2001 @04:26AM (#2511216)
    who cares what intels chips can get up to in MHz. I can design a chip that runs at 200GHz, does some useful processing and is slower then a 486.

    Transmeta wasn't originally meant as a low power processor. They tried to optimize transistors vs performance and did a good job. Unfortuneately they forgot that nobody really cares about it. They then decided to try the low power market, but since Intel made a chip specific to lower power of course Intel will beat them out.
    • I can design a chip that runs at 200GHz, does some useful processing ..

      I doubt this.

    • Unfortuneately they forgot that nobody really cares about it. They then decided to try the low power market, but since Intel made a chip specific to lower power of course Intel will beat them out.

      Silly me. I just pre-ordered a Fujitsu P Series [fujitsupc.com] specifically because it uses a Transmeta TM5800 and gets around 10hrs runtime with the second battery.

      As a commuter and longtime laptop user I care a great deal more about power use than I do raw speed. I don't think you should be so quick to declare "Intel will beat them out". Speed Step is pretty much worthless and I wouldn't count on the Intel M series to really be that much better. Intel is much more concerned with their core desktop & server business, and keeping AMD at bay.

      I'd rather take a minor speed hit with Transmeta and get 1.5-2x the battery life, than have a slightly faster laptop that gets 2.5-3 hours runtime and that can also double as a space heater.
  • What does the expression "on the ropes" mean?
    • What does the expression "on the ropes" mean?
      Literally, leaning against the ropes around a boxing ring. Used as a metaphor for being worn out and near defeat.
  • Speed? (Score:2, Insightful)

    by jkellmer ( 250509 )
    Who cares about speed?

    From the various fans inside (motherboard, graphics card, power supply (2x)), my machine sounds like a twister.

    Bring down the power consumption or I will stay with my "Low-End" 1GHz machine forever.
  • Oh forget it.
  • Speed Kills (Score:3, Interesting)

    by ll5 ( 522784 ) on Friday November 02, 2001 @04:45AM (#2511252) Homepage
    Low power is great, there is a definite need for using less power and producing less heat in some systems. As for MHZ increases, I truly wonder what is driving the need for speed anymore other than media types and gamers. Where are the next generation apps that will utilize this kind of firepower? Media producers, avid gamers, engineers, and server roles excluded, who else needs or even wants this kind of power? What will you do with it, besides *everything* you do today "faster"?
    • Re:Speed Kills (Score:4, Interesting)

      by AntiNorm ( 155641 ) on Friday November 02, 2001 @04:48AM (#2511257)
      Low power is great, there is a definite need for using less power and producing less heat in some systems. As for MHZ increases, I truly wonder what is driving the need for speed anymore other than media types and gamers. Where are the next generation apps that will utilize this kind of firepower? Media producers, avid gamers, engineers, and server roles excluded, who else needs or even wants this kind of power? What will you do with it, besides *everything* you do today "faster"?

      The same thing was said ten years ago.
      • The same thing was said 3 months ago.. :)
      • Re:Speed Kills (Score:5, Insightful)

        by jtdubs ( 61885 ) on Friday November 02, 2001 @06:53AM (#2511407)
        Yes, but that doesn't make it true. I honestly feel that in the near future. Maybe even the recent past, we hit the point where home users will NEVER need more Mhz power. I think the future is SMP, not Mhz. I don't mean dual-proc machines. I mean 32 processor machines and 256 processor machines. Multiprocessing is the future. Threaded applications running across several CPUs is the future. I think.

        More cycles will NOT make Word run faster. Word is I/O bound, not CPU bound. It won't make Internet Explorer run faster either. It's bandwidth bound, not CPU bound. It won't make games run faster. Game's have become bandwidth bound as well, only different bandwidth. Specifically, AGP and North Bridge bandwidth.

        There are things that will benefit infinitely from more Mhz though. Specifically AI and Physics simulations spring to mind. Haha, spring to mind, that's great, a figure of speach that combines both physics and AI. Whew. I kill me.

        Anyway.

        Faster memory, faster buses, more CPUs, that's what I think the future is like, not more Mhz.

        But, I've been wrong before. Almost too consistantly to be coincidence :-). So, we'll see.

        Justin Dubs
        • I honestly feel that in the near future. Maybe even the recent past, we hit the point where home users will NEVER need more Mhz power

          other famous quotes:

          "I think there is a world market for maybe five computers." -- Thomas Watson, chairman of IBM, 1943

          "But what is it good for?" -- Engineer at the Advanced Computing Systems Division of IBM, 1968, commenting on the microchip.

          "640K ought to be enough for anybody." -- Attributed to Bill Gates, 1981
        • I honestly feel that in the near future. Maybe even the recent past, we hit the point where home users will NEVER need more Mhz power.

          For home users we hit that point at 200MHz. Now we're at the point where developers don't even care. I have an 866MHz Pentium III and I do hardcore commercial software development (read "games"). No one where I work is running out and buying the latest 1.4+ GHz machines. It just doesn't matter to us; we don't see a noticible difference. Most of the games that run slow on so-called midrange machines are just because of sloppy coding, not because they really need 900MHz, and I think consumers are starting to pick up on this. Game X may require a 1GHz processor. Game Y may require a 500MHz processor and have noticibly more sophisticated visuals. Hmmm.

          More cycles will NOT make Word run faster.

          Again, this has been true since the original 200MHz Pentium.
          • *cough* *cough*

            Do you work on a 200 mhz pentium? Trust me...i have to deal with some crappy slow machines...and 200 mhz just izn't enough balls to do anything at any sort of "speed". Trust me, you notice the difference.

            My AthlonC 1.4 can boot win98 unbelievably fast. And, it's noticable: ever waited 2 minutes for your computer to boot, and watch your friend's comp boot up faster than you can say the ABC's?

            I always notice slowdown. Sure, i'm a power user...and sure, hardware can keep up with all the 'basics': Email, web browsing, media processing, etc. But we can always want it to be faster. Always want Winamp to stop skipping. Always want your program to load faster. And who doesn't play games?

            Sure. keep your box all nice and neat, few apps, no crap running in the background. Be a power user.

            But power users notice speed changes, they always want better.

            And average Joe User runs all the crap that slows a pc down...all the background apps...and he notices when IE takes 15 seconds to get up. When windows takes 3 minutes to boot on his new gateway with all the crap they package.

            Trust me, we always need more speed.
            • I wonder how much of that speed difference can be attributed to the 200Mhz CPU though.

              I bet it is also running on old 66Mhz or worse, 33Mhz (maybe even paired) RAM. Yuch.

              I bet the Hard Disk is running in damned PIO Mode 4 rather than UATA/anything.

              Sure, if you could stick a 200Mhz CPU in my current machine with PC2100 DDR and an ATA/100 EIDE RAID 0, it would slow my box down. Windows would boot more slowly. But not NEARLY as slowly as with the slow buses as well.

              Anyway, you are right. 200Mhz is too slow for many things. Just thought I should point out the numerous other factors involved. Have fun,

              Justin
        • I think the future is SMP, not Mhz....

          Word is I/O bound, not CPU bound. It won't make Internet Explorer run faster either. It's bandwidth bound, not CPU bound. It won't make games run faster. Game's have become bandwidth bound as well, only different bandwidth.

          And you know what? SMP will do absolutely nothing to fix bandwidth problems. Instead, it will make them worse.

          I agree that the greatest bottleneck in modern PC's is I/O bandwidth between the components, but how you jumped to SMP as the solution just escapes me.

        • Assuming the pattern of the home user's usage doesn't change.

          Lots of people, myself included, are now playing with hardcore audio manipulation- capturing albums/cassettes with their sound card and then digitally enhacning it to remove tape hiss and other noise...

          Any idea how long this takes? For the last 45 minute cassette I did (Beatles' Get Back LP Compilation #1, if anyoen cares), I tally up the total computation time at around 7-8 hours on my 400MHz Celeron. Of course, I probably could have shaved 4 hours off if I did everything in 44.1KHz (I used 48 for the editing), but hey..

          You could have told me when I got my 486 DX4/100 (coming from a 386 DX/33, which felt like a speed daemon compared to my 4.7MHz 8088...) that I'd never need more processing power, and I would have agreed. And that's true, too, if I don't use my computer to do anything that I wasn't doing in the early-mid 90s.

      • The same thing was said ten years ago.

        But in the era of the n86 machines, people would say 'why would I ever need more speed?' and buy the fastest or 2nd fastest machine on the market.

        On the other hand, right now, I'm shopping for a celeron 600-700 for my mom, buying a p120 for $20 to use as a firewall/router/mailserver at home. Celerons and Durons are selling like hotcakes, people are buying used computers like never before.

        A good friend of mine just upgraded to a blazing fast dual celeron 433 system, like the one I've been happily using for the past 2 years. I've upgraded it a lot, but extended it's functionality, not performancem, adding expansion cards for sound, firewire, iRDA, SCSI. Adding RAM (which I expect to start seeing in cereal boxes).

        It is fast enough.

        The only reason I actually use a GHZ+ machine is because I work on print at work and it's the machine my company gave me. I don't plan on upgrading until I can get a dual Athlon 1.4 for ~$250 here in europe. It'll be a vanity upgrade: I don't need the extra power, I just want to have it.

        • ...RAM (which I expect to start seeing in cereal boxes)....

          Just have to say that was funny. And it wasn't designed to karma whore either...I like it. Someone being clever on Slashdot for the sake of being clever. Whaddya know.
    • The reason why machines are getting faster and faster is simple: every operating system I know are adding more and more functionality as standard, and that requires more and more computing power.

      Sure, you can run Linux in command line mode quite well on older machines, but if you plan on running either the KDE or Gnome graphical environments better plan on something a bit more modern. This is especially true if you want to be involved with digital media content in any serious fashion.
    • What will you do with it, besides *everything* you do today "faster"?

      Maybe it will run Windows XP at a decent speed? I saw a demo running on a 1.3 GHz Athlon last weekend and it was decidedly sluggish...

      ....laura who tried very hard not to laugh

    • Maybe Intel is just being prudent and creating high speed processors before new apps come out that require such high speeds. This way they won't be caught producing slow processors that nobody wants and take a huge hit.
    • No, no, no! Anyone who thinks the world wouldn't benefit from faster computers has his head in the sand. I thought this was a college site? Ask any professor (errr... *real* professor: math/sci/cs) and I bet every one, unless they're a total dimwit will tell you how important it is.

      True, scientific computing as well as other things have been riding on the coattails of gaming, appreciating the miniturization and personalization of computers. Get a life, do some research, and then whine about the uselessness of chip power
    • Media producer and gamer are broad catagories. A couple years ago only a speed freak gamer would think of owning a 500 MHz machine, now my grandparents own a 500 MHz Gateway and a 700 Mhz HP and all they really do is type up greeting cards and use the internet. It is alot easier to do any sort of graphics work on a 700 MHz Celeron then it is trying to do it on a 66 MHz 486. On that point now all "media producers" are professionals. Lots of people are make their own audio CDs, touching up old family photo albums, and getting creative with home movies on their computers. More and faster are keys to these sorts of applications. I used to do video editing work on old PowerMac 8500AVs and it took forever to do even the simplest things with small video clips, you wouldn't think of doing a major project without access to a machine with a several thousand dollar hardware video card. With a 799$ iMac people can do in an afternoon what it would take me a week to do.
  • by Pathetic Coward ( 33033 ) on Friday November 02, 2001 @05:05AM (#2511278)
    The economy is in the toilet, no one is working, the machine you bought two years ago is just fine for your needs: revising your resume and dialing up Monster ...
  • I'm waiting for the PV (unless my PIII starts bugging me too much before that), the PIV line is just too plagued for me to touch. Which of course is not Intel's fault and all, but who cares? Maybe if I find a REAL STEAL on eBay.
    • ...the PIV line is just too plagued for me to touch. Which of course is not Intel's fault and all...

      I'm curious. Who's 'fault' do you think it is that the P4 is a dog?
      • Personally, I think that there was a breakdown in communication between rambus and intel, and that with everybody looking out for their own interests and the interest of meeting deadlines, things just got a little out of hand.
  • Still only 32-bit (Score:4, Insightful)

    by morbid ( 4258 ) on Friday November 02, 2001 @05:40AM (#2511321) Journal
    3GHz is all well an good, but we're coming up against the 4GB RAM addressing limit of 32-bit machines. Granted, the Pentium architecture can address 36-bits using segments, but we're back to the bad old days of LIM Expanded Memory and segments (who could ever need more than 640k?).
    I think AMD are on to a reall winner with their 64-bit Hammer architecture since that's completely backwards compatible and has a flat, 64-bit address space.
    • Re:Still only 32-bit (Score:1, Interesting)

      by Anonymous Coward
      Agreed. Intel seems to be missing the boat with Itanic. Constant delays, (we've been hearing about Merced since the mid nineties,) a complete lack of interoperability (I personally think that's a GOOD thing, but it makes it harder to sell,) and a ridiculous pricepoint (will eventually change, of course,) intel is missing out while AMD forges ahead with its consumer level product.

      Keep in mind, though, that you will not be able to use 64-bit addressing in 32-bit compatability mode with the AMD offering... you'll still be stuck with the 4 gig limit when running "legacy" code.

      If both products were available today at the same price, and if applications were available, I'd have to go with the Itanic. Intel has thrown away the last 20+ years of legacy crap that has been slowing development for so long. AMD, in an effort to ease migration and appease the masses of consumers, has retained those old roadblocks and built new roads around them.

      AMD also doesn't have the marketing to push their product and make it mainstream. Remember, in the modern economy it doesn't matter if your product is better: people have to BELIEVE that it's better. AMD isn't good at this, intel is.
      • So AMD won't sell many processors to the AOL crowd.

        Marketing hype only works if people don't know any better, which more and more people do these days.

        Right now, AMD may not be selling as many units as Intel but they have a far superior product. Sooner or later the marketing ploy is going to run out for Intel (Men painted blue doing stupid things - wtf are they thinking?!).
      • If both products were available today at the same price, and if applications were available, I'd have to go with the Itanic. Intel has thrown away the last 20+ years of legacy crap that has been slowing development for so long. AMD, in an effort to ease migration and appease the masses of consumers, has retained those old roadblocks and built new roads around them.

        Intel has traded a bunch of old road blocks and legacy crap for an entire new set of crap. Intel's developement of this new architecture has been bogged down for a while. You've said it your self that you've been hearing about it since the mid-nineties. Besides AMD's solution is really the same as Intel's. They both can execute IA-32 code on the chip, they both introduce a new instruction set which causes compilers to be re-written for. They both have their problems in design and manufacturing. If anything AMD has taken an approach where there is less to go wrong than in Intel's.
      • Re:Still only 32-bit (Score:1, Informative)

        by morbid ( 4258 )
        "Keep in mind, though, that you will not be able to use 64-bit addressing in 32-bit compatability mode with the AMD offering... you'll still be stuck with the 4 gig limit when running "legacy" code."

        True, but with 64-bit OS's on Hammer, you can run 32-bit apps along side the 64-bit ones. Have a look at the work SuSE is doing.
        http://www.x86-64.org

        It's pretty impressive, and it's all working (on the simulator). I've seen 64-bit Linux boot and run X on a laptop running the simulator :-)
    • First of all, Hammer does not have a flat 64 bit address space. Its address space is 48 bits (linear) and 40 bits (physical). Second, Intel's 36 bit addressing does not require segments, it requires paging. It's a flat space from a process's point of view (limited to 4GB per process, of course).
    • Question I've been meaning to ask (since my current, erm, desktop supports 32GB of ram)... does Linux support more than 4GB for the OS? If so, how gracefully? How do I enable it, if it's not enabled by default?
    • The 32bit Addressing limit isn't 'really' there though.

      True, you have to use CPU extensions in code, but all processors since the PPro have had > 32 bit address buses. IIRC, the P3's support something like 64GB of physical ram, the Xeon processors support more (1TB iirc - 40bit).

      The problem, of course, is the chipsets/motherboard manufacturers that fail to make use of the wider address bus.
  • I cannot help thinking that Transmeta's innovative technology has been overlooked. The cool thing about the transmeta architecture is not its low power consumption, rather it is the way it can do all kinds of translation on the fly. I hope this does not get buried just because people think transmeta is all about low power, and Intel come along and transmeta sinks without a trace.

    I wonder what Linus would think ?

  • by skroz ( 7870 )
    Anyone know when intel plans to release SMP enabled versions of the P4 and/or chipsets that support it? I've seen SMP P4 Xeons on their roadmaps, but nothing about the P4. They were supposed to announce something about it at microprocessor forum, but didn't. Any clues?
    • IIRC SMP will only be available for the P4 XEON. PIII will be the last Desktop/workstation/server CPU. Any SMP Offerings will have Xeon CPU's under the hood.
  • by sethamin ( 533611 )
    This looks like the beginning of the next phase in PC CPUs. It's possible that instead of seeing lots of innovation to give us better IPC because of speed limits, the major chipmakers will start to go the route of just denser and higher speeds, i.e. pump that MHz and make 2-way, 3-way, 4-way, etc. systems. Forget about re-architecture, just make a core and try to get as many of them in there at as high a speed as possible.

    I think the P4 was an especially telltale sign of the times ahead for the PC chip industry. While the AMD rivalry has helped to spark fierce pricing competition, I also think that it has prompted Intel to go the "MHz at any cost" route. Don't be suprised to see the P4 as the first in a long line of "let's increase that pipeline to pump up the clock speed!" While undoubtedly this can make a chip faster if IPC is not just cut equivalently, it also smacks of "MHz at all" marketing strategy.

    Intel has realized (more than AMD, who is still trying to "educate" those consumers) that the general mass of people don't pay attention to SIMD instructions, double clocked FPU units, superscalar speculative execution, full speed caches, or any of that other jive that gives you higher IPC. They look at MHz and just want to see higher numbers. And also more CPUs can't = bad either, can it? I mean, that's the next marketing blitz campaign once MHz stops working.

    • And also more CPUs can't = bad either, can it? I mean, that's the next marketing blitz campaign once MHz stops working.

      More is better, provided you can use it. Posted from a 150 MHz, 20 meg RAM, laptop with Mozilla running on a 650MHz Athlon thanks to SSH -X and Debian. Kind of nice to only have to install things on one machine. 66MHz 486 keeps the gate, 180MHz serves http, ftp and dns, 500MHz K6/2 and Red Hat drives a printer and makes the wife happy. There you have it, five freaking computers and it's only the beginning. Kerbos authentication would be nice, and I have a box that could do it if the mailbox gets slowed down. A box dedicated to voice recognition that had the authority to command would be excellent. A radio gateway to help eliminate the last mile rape is planned. A wireless hub for inside would be nice too. This hundred foot cable to the laptop can be a drag. None of this has cost that much. I've simply been collecting other people's trash and buying a hundred buck mobo from time to time over the last five years. My largest single expenditure was this $450 laptop.

      Yes, I share with my frinds.

  • P4 architecture. (Score:5, Informative)

    by Dwain_Snyders ( 412284 ) on Friday November 02, 2001 @08:51AM (#2511524) Homepage

    The P4 architecture is not brilliant, pushing up the clock speed won't help the fundamentally stunted technology. There are major problems with the architecture, the worst of which is probably their decoder implementation.
    The new architecture implements the U-V pairing and 4-1-1 in a nonsensical way. Multiple decoders have been eliminated and only one functioning decoder operates... the result of this is that just one instruction can be processed per clock cycle. Intel's theory was that the trace cache would eliminate the need to decode an instruction every clock cycle.


    However, this falls apart when a set of instructions is put forward that does not go into the trace cache.... the processor must call upon the L2 cache or put all that code into memory to pull in another 64 bytes of memory for each instruction - and then decode the 64 bytes of code each time! The end result is that the P4 takes a lot more cycles to decode these instructions. Compared to the AMD Palomino XP processor (the fastest Athlon chip at the moment, in fact, the fastest X86 chip at the moment!), the P4 performance is a bit underwhelming.


    The new Thoroughbred line of processors will introduce even better performance and completely blow Intel's offerings out of the water.

    • Dwain,

      What makes the AMD CPU's extremely good compared to the Intel CPU's is the fact that because AMD designed their CPU's with more much efficient FPU units (no Pentium Pro FPU legacy) and also more efficient access to L2 cache, the result is extremely high performance on a per MHz basis. Indeed, the Athlon XP 1800+ on a DDR-SDRAM motherboard will run rings around the Pentium 4 2,000 MHz on most apps except those that are optimized specifically for SSE2 multimedia extensions.
    • Intel's theory was that the trace cache would eliminate the need to decode an instruction every clock cycle.

      But they made the trace cache way too small, which made this even less likely to work.

      I agree with you: the P4 is broken in many ways.

      If they really can push its clock rate much faster than AMD can, the P4 might start to win just on sheer clock speed; but so far AMD has been keeping up. Since AMD's chips do more per clock, they don't need as high a clock rate.

      steveha
  • err (Score:3, Insightful)

    by RainbowSix ( 105550 ) on Friday November 02, 2001 @09:37AM (#2511605) Homepage
    I look forward to good values on eBay when 2GHz is "obsolete."

    Why??
    Let's see... you can wait a year until the 2ghz P4 drops to $100 and get one used on ebay, OR, you can take that $100 today and buy a new 1.4ghz AMD Thunderbird. You get similar performance... but on yea, you don't have to wait a friggen year!

    How much you pay = amount of money + transaction cost. A years worth of computing sure has value to me!
  • Low power CPUs certainly can help drive down the costs in less than obvious ways. With a CPU running at maybe 20W less power - a data centre of 500 dual processor servers will be saving an incredible amount of power - both in terms of electricity required to power and aircon to cool. My bad math says maybe 20kW, 24x7...
  • I look forward to good values on eBay when 2GHz is "obsolete."
    For almost all purposes, it is. AMD's top chips blow it away, for a lot less cash...
    • "blow it away". Seems we have a little bit of a reality/exagerration problem, here, don't we?

      In real life, AMD's top end is _comparable_ in some benchmarks to the 2.0Ghz. In other benchmarks (particularly highly optimized and memory-intensive) the 2.0Ghz beats them nicely.

      Please try to exist in the same world as the rest of us.
      • I recommend a visit to Tom's Hardware [tomshardware.com] to check out some benchmarking done on P4's and AthlonXP's. "comparable in some benchmarks" isn't accurate, but of course neither is the idea that the pentium 4 is blown away. Exaggeration is retarded; we are (or, should be) intellectually debating, not flaming. Yes, I am an idealist.

        In reality, AMD's best chip can compete with and usually beat the P4. (shrug). that's the way it is. But, the point is that AMD's pricing is better, they are loved as the underdog (and not hated because of Intel's Monopoly Price Gouging ((tm), licenced from Microsoft).

        Anyone remember the prices of Intel chips (especially high-end) before AMD came on the scene? I do. It was insane...the CPU was about half the system price. Now, I spend about $130 for a good chip, instead of $800. I attribute that to AMD. That is one of the reasons I love the company, that is why i drove 2 hours at 6 this morning to Detroit to their XPerience Tour to get a chance at a free chip and motherboard. People love AMD. No one 'loves' Intel.

        And yeah, it was worth it, to hear the chant of, "INTEL SUCKS! INTEL SUCKS!"
  • Any mention of an Itanium counter to AMD's Hammer. [slashdot.org] With mid-2002 not so far away this would be the time for them to be drumming up interest in it. To be fair, I've checked the Comdex site for news or presskit from AMD and nothing so far, but I expect they must have something there, as this is where they'd run it up the flagpole and try to get potential customers to salute it.

To thine own self be true. (If not that, at least make some money.)

Working...