Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Hardware

Coppermine vs. Athlon 205

Posted by CmdrTaco from the battle-of-the-titons dept.
SaDan writes "I checked out a comparison of the new Intel Coppermine processors and AMD's Athlon chips at Tom's Hardware Guide last night. It's kind of interesting, and I thought others would be curious about how Athlon stacks up against Intel's latest offering. "
This discussion has been archived. No new comments can be posted.

Coppermine vs. Athlon More

Coppermine vs. Athlon

Comments Filter:

  • .. (Score:1)

    by Anonymous Coward
    Athlon still wins in Sysmark98 tests. I like AMD vs Intel stuff. It's similar to the linux vs M$ issues. Go AMD!
  • The referenced review basically shows that Intel once dropped to .18 and clocked 5% faster can pretty much keep pace with AMD's shipping product for $150 more.

    Add to that Coppermines OLD core, Athlons scalability and Fab30 coming online soon and it is clear that INTC is going to have to do more than in the past to stay relevant.

    If it isn't great news for AMD investors, this is at least great news for CPU buyers as INTC will have to WORK for it's money for a change!
  • There's also a comparison of several different processors ranging from a K6-III 400Mhz to the Athlon 700 with a couple Intel processors thrown into the mix...

    http://www.aceshardware.com/Spades/read.php?arti cle_id=84

  • It's always good to see the underdog getting ahead. Athlon seems to be the superior game performer, except for quake 3. But still, it really looks like AMD is becoming the best choice for gamers, not only because it's faster, but because it's cheaper.

    Maybe they should both seek seperate markets, AMD should go for gaming/low cost 3D workstations, and Intel should stick to servers and the like. That may even out the market, and get rid of some of this "do everything we can to keep the competitors down" attitude. I mean really, when you have that much of the market, can they really be that much of a threat?

    Start working towards real technological advances, rather than mediocre enhancements to beat the competition.
  • AMD i'm hoping you are still pushing to put out those mobile k6-3s. Along with a SMP athlon system, i want nothing more than a very efficient fast cheap notebook (only other option is the cool as hell, yet expensive, g3 powerbooks). Anyone else think a k6-3 mobile would sky rocket AMD sales in the portable market (though k6-2s seem to be doing great already)
  • it still pushes AMD to improve their design, perhaps to the point that the K7 can also be used in portable devices. i think that the major appeal of intel's newest processors will be in the laptop arena. it would be awesome to have the same power in a laptop as on the desktop with an x86 processor.
    i really hate to say it, but it's getting to the point where it can be cheaper to buy a new system than to upgrade an older one (especially with DRAM prices so high.) of course, with the introduction of new processors and technology, older (obsolete?) models become much cheaper, and hey, a better processor is a better processor. competition is good, as long as the competition doesn't force either company out of business. that would be bad.

  • With the recent attitude of "innovate, retaliate, counterretaliate" in the chip market, both companies are really giving the consumers a lot to look forward to. The only thing I wonder is, how long will it be before Intel and AMD are forced to slow their shrinking price to performance ratio in order to protect their bottom line?

    On an unrelated gripe, "Tom's Hardware Guide" isn't helping its credibility much by having a major error in the very first sentance. "pushed from its thrown"? Ugh.

  • Bad News (Good News) (Score:5)

    by Christopher B. Brown ( 1267 ) <cbbrowne@gmail.com> on Monday October 25, 1999 @10:36AM (#1588241) Homepage
    The bad news for AMD in this is several-fold:
    • Intel steps things up on the number-of-megahertz side of things, which is good for marketing.

      (Of course, anyone that should be considering these high-end processors should be competent enough to know that performance is only positively correlated with the number of MHz. They don't run lock-step...)

    • Coppermine comes better prepared for laptop configurations

      (Of course, there aren't many really-high-powered laptops; there was only ever one Alpha-based laptop, few SPARC-based, and such... I agree with others that availability of faster AMD chips in the K6 series is more important at this time...)

    • It's probably more important news that engineering-quantities of Itanium chips are starting to be released.

      (I half-expect to see a report from VA-Linux Systems some time soon...)

  • I'm the kind of user who doesn't give a crap about who makes the chip, or how much it costs, just as long as it's the fastest mainstream consumer chip available. And as the basis of that, I'm still going with the Athlon for two reasons. It beat the new PIII in just about every test. Second, I haven't seen any super cooled PIII chips that are like what Kryotech has done. Can you say 1 GHz Athlons by December? Mmmm [kryotech.com]
  • ... and I've been told that it supports PenIII's with a simple bios upgrade. Does this mean that I can use Coppermine processors in it? Anyone know?
  • Remember, the goal of Intel and AMD is not to put out the studliest CPU, but to maximize profit. The $150 premium you pay for the Intel P-III over the Athlon is probably a fair estimation of the business goodwill Intel has invested in the Pentium trademark. And most people (although perhaps not most slashdot readers) are probably willing to pay that premium to get the "Intel Inside" sticker.
  • The bad news for AMD in this is several-fold:

    Also, the new Intel chips run much cooler than an Athlon, which eases cooling issues. It might make more of a difference for those wanting to run multiprocessor machines.
  • What I like is how intel sidestepped the clock of the beast issue by making a 667Mhz processor. They really missed out on a great marketing opportunity! Frag your friends like never before, with a computer made by Satan's own hand: 666Mhz!

  • Slot 1 vs Slot A? (Score:4)

    by G27 Radio ( 78394 ) on Monday October 25, 1999 @11:16AM (#1588249)
    I'm hoping someone can clarify something for me. If I understand correctly, the slot that Pentium II/III's go in is called Slot 1. Intel patented Slot 1 so that other companies could no longer make replacement processors.
    So AMD had to stick with the Socket370(?) form (which was the non-proprietary Pentium Pro socket.) Until they came out with the Athlon. AMD (or somebody) created the Slot A socket which is suprisingly(not!) similar to Slot 1 but not compatible. Main question: Is Slot A proprietary also? Or can other companies make processors for it other than AMD?

    NOTE: The above is based on many guesses, assumptions, things I've read, and things I may have imagined reading. Please correct me or clarify (I'm sure there are errors.)

    numb

    ?syntax error
  • Soon after their new Pentium III chips were found to be outperformed by AMD, Intel released a very disappointing profit report, which sent stock prices down as much as 10 points within a couple of days. Immediately following this, Intel makes several announcements about Willamette and Coppermine, and their stock goes back up. It seems that Intel may be pushing these chips a little too quickly in an attempt to make their investors feel better.
  • So AMD had to stick with the Socket370(?) form (which was the non-proprietary Pentium Pro socket.)


    up until athlon all of amds chips past their 486 chips went into (super) socket 7 with the exception of mobile proccessors which i have no idea about.

    pentium pro used socket 8
  • The day you see Intel bowing out of the low-end of the market, is the day you'll see the SPARC-ification of the x86 platform. um, or the MIPS-ification, or the PPC-ification, or the Alpha-ification (say that ten times fast).

    The low-end is Intel's bread and butter, their marketshare, and they can't live without that. It's what the hegemony is based on. Once they lose their marketshare, they'll have to scale back on their infrastructure, and without the Fabs, they lose their major advantage - the fact that when someone orders 10000 CPU's, Intel's got them out the door. AMD or Motorola just cries.

    I wish I had a nickel for every time someone said "Information wants to be free".
  • AMD i'm hoping you are still pushing to put out those mobile k6-3s.

    There's a reference to K6-III-Ps at speeds of 350, 366, and 380 MHz in AMD's retail employee website [c-3.com] (registration required), and there's this less-detailed information in their public website [amd.com]. I don't know how any of this translates to notebooks that you can actually buy that have this processor...K6-2-Ps are available at speeds up to 475 MHz, and people seem to be fixating on megahertz alone. (Not that the K6-2's bad...I have one myself and it runs like a champ, but the K6-III, from all I've seen, is substantially faster, especially at high clock speeds where the L2 cache speed difference gets totally out-of-control.)

  • R-E-S-P-E-C-T (Score:5)

    by tekman ( 95776 ) on Monday October 25, 1999 @11:28AM (#1588255) Homepage
    The main thing that AMD has gained out of this whole Athlon thing is respect. Real respect. Not just, "Gee, since I'm a very poor geek, I'm going to put an AMD chip in my box."

    To illustrate: my college's career fair was just a few days ago. AMD was there. They have been there in the past. In the past, only die-hard computer engineering hardware geeks talked to them. This year, however, as they had Athlon processor periphenalia and even a couple actual processors (none of which they were giving away) there was a line. I'm talking about a long line. Everybody wanted to talk to teh AMD guys. I waited in it for over twenty minutes then decided that since hardware doesn't make me feel an extreme amount of joy inside I would go talk to someone else.

    Bright Young Minds (at least, I think that's what we are) are taking notice of AMD and are intensely interested in being hired by them. This seems like a Good Crowd to have on your side when speculating on the future.
  • Slot 1 is Intel's P2 (and 3?) slot.
    Slot 2 is Intel's Xeon (and P3?) slot.
    Socket 370 is Intel's Celeron socket (lower cost)
    Socket 8 is Intel's Pentium Pro socket.
    Socket 7/Super7 is for Pentium, Pentium MMX, AMDs, Cyrixes, Rises, and WinChips.
    Slot A is AMD's Athlon slot.

    As for your question, I have no idea. Probably not though.
  • Ok, yes, Intel put wierd liscencing terms on the slot 1 so no one could use it. But it did that for Socket370 too. Socket370 is a socket they came up with for celerons and future socket chips. The pentium pro went in the socket 8 (which was also under strange liscencing terms). AMD's former chips (the k6-2s and 3s) went in what they called the super socket 7 which was the 66mhz socket 7 used in pentiums raised to 100mhz (thank you via). The slot A is nothing at all like the slot 1 because it doesn't use intels GTL+ bus. They worked with digita^H^H^H^H^H^HCompaq to move the EV6 (alpha 21264) bus to an x86 compatible chipset. Which is where a good deal of their speed comes from (200mhz bus, better general architecture of the bus (especially for smp)). As for whether other companies can make slot A chips? I don't know...it depends on how compaq liscenced it to amd.

    Joe

  • .18 not impressive (Score:1)

    by Anonymous Coward
    shipping G4's are .15 and 500 and 550mhz will be .13 microns
  • Okay, here we go:

    Socket 7/Super 7: Pentium, Pentium MMX, AMD K6, K6-2, K6-III (Super 7 is a Socket 7 w/ 100MHz FSB)

    Socket 8: Pentium Pro

    Slot 1: Pentium II/III, some Celerons

    Slot 2: Pentium Xeon?

    Slot A: AMD K7/Athlon - Slot A is mechnically but not electrically compatible with Slot 1.

    Socket 370 - Designed by IBM - used by Intel for some Celerons & Pentium IIIs. The future for Intel - Slot 1 goes bye-bye over the next few years.

    I don't know about the proprietary nature (or lack thereof) of Slot A, sorry.

  • It's there for you to click on Pooky. It ain't great but it amuses me.

    You the *man* doin' "fulltime" x 2 OK.

    CC

  • German article in c't 22/99 (Score:1)

    by Anonymous Coward
    for all those germans :-) look at
    http://www.heise.de/ct/99/22/132/

  • It's not just brand recognition. Intel has marketing arrangements with major computer companies that virtually guarantees them a market for these CPUs on corporate desktops and low end servers. And corporations will pay well to get the "fastest" chip (read highest Mhz number).

    AMD, unfortuately, can only really sell into the "home market" channel from the major manufactures. That and the "clone" and hobbiest market needs to keep them afloat until they can get into the business market.

  • Re:Please not Tom (Score:1)

    by Anonymous Coward
    All I have to say to that is... **Please** put down your crack pipe.
  • The entire concept of media objectivity is a pointless exercise in wishful thinking. Any person who wishes to share a researched article with others should endeavor to make their bias KNOWN (as Tom has said on numerous occasions, he likes to "root for the underdog" [not a quote from Tom, just paraphrasing]) and with that in mind, present both sides of the issue fairly. Whether the author brings the research to a conclusion or leaves the matter up to the reader is an exercise in knowing what sort of article deserves which ending.

    There are enough myths out in the world already. Please don't contribute to another.

    -l
  • Is it just me or does everything that comes out of T.H. seem really reliable? There was a time that I trusted him exclusively for all my HW advice (I'm really a software guy at heart). I've been a bit skeptical about his site since he went commercial, but what does the rest of /. think?
  • remember that PPC chips are RISC w/ some CISC-ish features and thus have smaller transistor count requirements by default. Since that count is smaller, it is easier to "squeeze" the chips onto a smaller die.

    this is my understanding anyway. i am not an engineer. any of those folk care to speak up?

    -l
  • You forgot...

    Slot B: Athlon Ultra / DEC Alpha
    socket 410: Intel NG connector

    There were some rumors that S370 would be used by AMD, Cyrix, and others eventually as the next socket format. The reason is that socket chips are cheaper, and slots were only used because of difficulties with large caches. Eventually you go back to socket (like Intel has done), and reduce cost across the board.

    Oh, and Slot-2 is for Xeon, P2 and P3 generations. Slot 1 is used for Pentium II/III and old Celerons, and S370 is used for Celerons, P3s, and Coppermine (just another P3).

    I wish Intel would get their next IA-32 architecture out. The P6 architecture is anchient, and was supposed to last 3-4 years. They just seemed to put to many developers on the IA-64, and lag IA-32... though I remember reading somewhere that the NG IA-32 is supposed to have mny simularities to the IA-64 design...

  • Re:.18 not impressive (Score:2)

    by Anonymous Coward
    shipping G4's are .15 and 500 and 550mhz will be .13 microns

    The argument can backfire: it looks like they had to go .15 and .13 microns to reach 500 and 550 Mhz. If so, how do they expect to reach 1 Ghz ? FYI, the Alpha 21264 at 600 Mhz is 0.35, at 800 Mhz ~0.25 and at >= 1 Ghz is expected to be 0.18 micron.

  • On an unrelated gripe, "Tom's Hardware Guide" isn't helping its credibility much by having a major error in the very first sentance. "pushed from its thrown"? Ugh.

    I don't know. I thought it fit right in in Slashdot's battle-of-the-titons dept.

  • Perhaps if you're going to complain about someone else's spelling troubles, you should be extra careful to make your own post error free. It's 'sentence.'

  • Re:Tom should be more objective (Score:1)

    by Anonymous Coward
    To say Tom has a bias towards AMD or nVidea would be the understatement of the decade. You can see him grudgingly accepting how Intel put out a good P6 based chip, but a half a sentence later he's already preaching about how AMD is putting out a new design Real Soon Now.

    He also seems a little too fond of the 3D Studio Max benchmark, it's his little toy that shows AMD doing 2x better, and he sort of glosses over the gaming performance (what was once his shouting point for AMD's and 3DNow).

    In the end, it was an ok review if you just read the benchmarks. His recommendation at the end should have taken price into account, not the one 3DSMax benchmark. He just goes back onto the AMD soapbox, even including data on K6-3's and crap in the conclusion of an Intel product review.

    (Note, Intel is not my favorite company. AMD has made a great chip, but Intel is responding to competition. Tom doesn't want to concede that fact. This is a Tom bash, not an AMD one).
  • Grammer? Grane?

    sigh...

  • If you cool it well enough, you can overclock the 550 to 733, by changing the FSB setting. However, the Athlon can be overclocked also, by FSB and Multiplier, though the multiplier setting requires a soldering iron and a lot of guts. But at least it's possible. I'm personally going to choose the Athlon.
  • Actually, that;s not the case. The BUS is really 33 1/3 MHz and 66 2/3, leading to exactly 100, 233, etc... not 231 or whatnot. the 66 MHz machine is closer to 67, and 233 is 233.33333333 etc.
  • I mean, the Athlon is a really cache dependent CPU. As everyone saw in the 3DSMAX benchmark, the FPU is fine, but when even a little bit of cache is necessary (see: Q3Test, Descent3), the Athlon starts lagging compared to the CuMine, especially at higher clock speeds where cache becomes a bigger factor.
    I can't wait to see an Athlon Ultra. ^^
  • First off, check your core voltage. Going from 0.25 um to 0.18 um requires you to lower your voltage, possibly to a point your MB doesn't offer. Best of luck to you!
  • Where do you get the info that they are shipping 0.15 um? I am working in the long term R&D (grad school) and have not heard of anyone shipping below 0.18 um product. Please offer a source.
  • I'd really like to see Athlon's scalability in the real world.

    What?

    You can't?

    There's numerous advantages you can tout with the Athlon: Speed (in MHz - up until today), Price, sheer Performance, but let's please hold off on scalability until AMD can point us to a supplier of 2-4 slot motherboards.

    I also doubt that anyone cares how old a CPU's core is, except for the people that actually care. Kids are going to buy a CPU based on how fast they percieve it to be, regardless of architectural constraits. In that reasoning, the first company to tout 1 GHz will have something to rub in the other's face for at least a little while.

    It'll be good news for AMD's stockholders when they can sell a chip that performs identically to the comparable Intel product and it costs more or at least comparable. Right now, you have to wonder how much longer their bankers will let them float with all their debt. (I'm only assuming they've acquired debt due to 6 or 7 quarters of losses.)
  • All this article talks about is how cool the Coppermine is, and how it's better than the Athlon. It looks more like Intel propaganda than a true comparison.

    I can't even get a Coppermine box yet, so why should I care that some computer is better in theory?
    --------
    "I already have all the latest software."
  • They didn't HAVE to go smaller to get more speed. They (Mot) did so they could reduce costs. The more chips you squeeze onto a wafer, the more money you generate from said wafer. That's all there is to that argument.

    Intel drags it's heels in order to maximize it's profit from each successive generation of tech, only switching when they absolutely cannot do anymore with what they've got (witness the 7th generation x86).

    Don't think that all the other semi-conductor co.'s follow their footsteps in that regard.
  • OMG, don't talk man.

    Intel HATES him. He released a Pentium II preview with a pre-production P2 which *sucked* it up bigtime. He's probably the most overtly anti-Intel man out. Why do you think he made a reference to a "fishy article" (pun at Sharky - who is biased towards Intel.)
  • You fail to consider the fact that he glosses over the gaming benchmark for a good reason. The G-Force probably had optimized SSE and Coppermine drivers (nVidia is really quick on new techs) while the 3D Max benchmark was a more even race. Second, he DOES say that PIII might be better if you are into gaming, but you should definitly stick to Athlon if you are running a workstation.
  • You fail to consider the fact that he glosses over the gaming benchmark for a good reason. The G-Force probably had optimized SSE and Coppermine drivers (nVidia is really quick on new techs) while the 3D Max benchmark was a more even race. Second, he DOES say that PIII might be better if you are into gaming, but you should definitly stick to Athlon if you are running a workstation. And don't doubt Athlon's gaming performane. Under the TNT 2 Ultra, it beats the PIII 120 fps to 90 fps.

  • I will let this exchange make my point.

    CC

  • I'd really like to see Athlon's scalability in the real world. What? You can't?

    I was referring to the scalability of the core and the fact that AMD has room to grow, while INTC doesn't. It is true that you can't run several Athlons on the same mobo. Yet.

    I also doubt that anyone cares how old a CPU's core is

    See above. Investors (which was the subject ;) should/do care about this stuff. Heck INTC was resorting to overclocking tricks to get their last chips to run at speed.

    Like I said though. Even in the event that AMD stumbles, this stuff is great for chip heads everywhere.

  • The physical slot isn't protected, it's the bus. Intel's Slot 1 share the same GTL+ bus with the socket 370 processors. Incendintly, socket 8 is covered under the same protection, even though it uses a slightly different bus. Haven't you ever wondered why AMD came out with an upgrade for all the socket 8 mobos? Because they can't.

    Slot A is physically identical to slot 1, it just uses a different bus archecture. Methinks the EV6 bus??? (I haven't studied that bus arch. yet)

    Joe Goldmeer

  • You're forgetting that Slashdot depends on the ads to generate its revenue... Enough said. (This isn't meant to be offensive, just an explanation...)

    --

  • Intel. That wasn't so bad now was it? At least it isn't a 4-letter word. :)

    Signed,

    SEAL - who is sick of ticker-symbols :)

  • I don't know about the rest of you, but I am VERY happy that AMD is giving Intel a run for our money.


    Personally, I was getting REALLY tired of the "new" and exciting "innovative" processors a whole 33Mhz / 50Mhz faster than last quarters every quarter.


    Since the Athlon has been out, P3 prices have been plummeting; I've been keeping track
    of the fall in prices [cpureview.com]; there is NO WAY prices would have fallen so low without the Athlon goosing Intel!


    I am very much looking forward to testing some Coppermine's, as soon as I can get my hands on them.


  • There's numerous advantages you can tout with the Athlon: Speed (in MHz - up until today), Price, sheer Performance, but let's please hold off on scalability until AMD can point us to a supplier of 2-4 slot motherboards.

    Well, considering the anti-competitive pressure Intel is putting on motherboard manufacturers to avoid distributing the single processor Slot A boards they've already designed and tested, I think AMD has an uphill battle to coax motherboard makers to design an SMP board.

    If I recall, the chipset designer that is working on a two-way Athlon chipset was talking about first quater of 2000 for release.

    ----
  • I mean, if speed was the only thing, we'd have had Alphas on all of our desktops for years, right? (Yeah, I've got *one*, but also seven x86s).

    The point is, I'm certainly not going to buy any AMD CPUs until their chipsets get stabler. This isn't intended to be FUD, but every non-USA chipset design I've used in the past (VIA, SIS and whomever) have just not been as stable (even when motherboards are from the same manufacturer) as their Intel brethren. I'm going to stick with Intel and 440BX until something stabler and better comes along -- and with features that I feel I need, unlike the 810 and 820 sets.

    -Chris
    Don't moderate this, bitch.
  • Why are CPU's considered more "cool" than other components? Why do people spend $250 more to gain a few MHz when the same money will get them more memory and a faster drive? I think it's because it's easy to quantify. People like that MHz is one simple number.

    So what if the Athlon is a tiny bit more value for the money - I'll get a lot more value if the price of a really good 19" monitor drops a couple of hundred bucks.
  • It's not much different now than when the PII came out, the K6-2 underperformed and he pretty much declared AMD to be in the dumper, all hail to Intel, etc.

  • Don't put to much trust in those Quake3 numbers. In true Tom fassion, he was dumb enough to use a geometry accelerated graphics card when testing processors. He is testing the card (and how well it happens to work with the processor) tenfold more than the actual processor.

    -
    /. is like a steer's horns, a point here, a point there and a lot of bull in between.
  • but you haven't really contributed anything at all have you. Maybe if you read the sources then you might have something interesting to say instead of just telling ppl that they are wrong without any evidence.
  • freedom is the by-product of economic surplus but hey, I've had a long day
  • If people pay 150 more for a sticker, it shows that we are all doomed. its like buying pepsi over coke, basically they are the same thing. Anyone that will buy intel for just the name; i have a penny i'll give you for $10. :)

  • Bad attitude. (Score:1)

    by Anonymous Coward
    Re: "I'm the kind of user who doesn't give a crap about who makes the chip,..."

    You should be ashamed of that attitude. Too many people share it. If, say Intel, is always putting out a chip 5% faster than AMD (or even 0% the way so many people buy only from the industry leader) and you types only buy from Intel, then within a few years, AMD will drop out and you'll see prices rise fast and performance rise slowly. And you'll have yourselfs to thank. Some temporary self-sacrifice from time to time can be a good thing in the long run. Do you really need that extra 5%? I'll answer for you. You don't.

  • I find this to be highly entertaining -- we've seen Intel basically dominate the market since it's creation. AMD the poor-geek's tool. I remember thinking back in the days of P6's (96 I think) that AMD's were simply a hack for people that couldn't afford the good stuff.
    I'm damn glad things have turned -- now we have AMD kicking Intel's ass in both price and power (For the most part -- I think that AMD is a better chip) and this really is shaking Intel up.
    I mean -- really, what type of respectable company would have the "Athalon Killer" anyway.
    $0.02 to AMD
    -= Making the world a better place =-
  • Did someone screw up the moderation here? This comment only contains incorrect guesses. It is definitely not insightful.
  • The benchmarks were odd, for everything else, the coppermine chips acted more or less like the other P3 chips, then in gaming stuff all of a sudden they jump way up. why?
  • Comment removed based on user account deletion
  • Did I misunderstand, or did Tom say there are a few PPGA socket 370 Pentium III's called 550E's. If so, would these work in the ABIT BP6 to give a relatively cheap dual Pentium III (or provide a good upgrade from a dual celeron system)?

  • Re:Fantastic news for AMD stock holders? (Score:1)

    by Anonymous Coward
    Come on, lay off the "Intel is pressuring them not to sell motherboards" crap and look at the facts.
    1: Both Intel and AMD have denied that this has happened. Only Tom Pabst has made this claim, and has never provided any evidence.
    2: AMD has announced that they only plan to ship about 1.5 million Athlon's this year. Intel should ship 20-30 million parts in the 4th quarter of the year.
    3: AMD has not commited to producing the Irongate once other vendor's release chipsets, which should occur in Q1 2000.
    4: The Irongate chipset requires a 6 layer motherboard, which is very expensive to produce. The VIA chipset (coming next year) works with a 4 (5?) layer motherboard.

    Now, given that you would have to produce an expensive, 6-layer motherboard, with a limited lifespan, and limited sales capabilities, would you spend your resources getting 20% of 1.5 million sales, or 5% or 20+ million sales with a motherboard that's probably cheaper to produce.

    In addition, AMD has repeatedly had production problems; one analyst once said "AMD has a long history of short term problems". Intel has generally delivered what they promise (not counting the recent i820 screwups), and is known to be a world class manufacturing operation.

    I highly doubt Intel would pressure anybody not to release product when simple economics would suffice.

  • Intel and AMD are going to soon milk the x86 design for all its worth, the moment either of them comes out with a true 64bit chip, I'll open my checkbook, otherwise I'll keep the PII 333 I bought a year ago ...
  • Every article I have read on that site has at least 1 point that gets on my nerves due to a low tech understanding of the issues. He claims that its a compiler difference that causes the big jump in performance on the Coppermine with the Q3 benchmark.


    Lets think about this for a minute! If Intel didn't change the SSE core then why would a compiler with SSE changes produce a binary that ran better on the same SSE core? The answer: Because its not the compiler! A better answer would be that the changes intel made to the cache result in significant performance gains in some situations. Tom dumps their tech doc's on what they did:

    1 they increased the associatively of the cache
    2 they widened the L2 data path to the CPU
    3 they decreased the latency
    4 Lastly they decreased the size by 1/2.

    In general 1 and 4 tend to cancel themselves to give similar performance (pick up an architecture book and read about caches if you don't know what I'm talking about here) So we are left with 2 and 3. Now 2 and 3 tend to allow you to get to the cache faster and get more data per cycle. Now quake is really an tiny engine (significant amount of the time it supposedly fits in L1) accessing a massive amount of data. Now lets assume that quake is so tight that it manages to fetch its data out of L2 cache a very large percentage of its time (as opposed to windows just randomly switching tasks, and using main memory like a big disk cache) now if suddenly your data loads which were always in the cache get to the processor faster keeping it from stalling a pipeline for 5 or 6 cycles what happens?


    Memory architecture is a __BIG__ deal with modern CPU's. A very large percentage of time on modern CPU design is spent trying to optimize data accesses. The intel engineers have done their homework. The PC market now considers games the standard benchmark (Quake being the main one, Celery-vs-K6! When was the last time you out typed Word? On the other hand when was the last time your Celery helped you kick that poor K6 owners ass because you were getting an extra 20fps?) so they discovered a way to help quake out while maintaining decent performance with data sets that were more sensitive to cache size rather than access times.

  • Um... AMD's 486 went into socket 7? Intel's didn't. Can someone with authority please chime in here. I know that old Pentiums went into socket 6 (I've got one).
  • Can anyone give any info on the relative merits (and accuracy) of the two different FPU benchmarks used by Tom's and Sharky. The 3D Rendering on Tom's shows a huge difference in favor of the Athalon, while the ZD FPUMark shows the new intel with a slight advantage. This is important to me because we are deciding on a machine to purchase in the next couple of months that will be more or less dedicated to floating point calculations.
  • Remember, this isn't like cheering for your favorite football team. I purchase whatever chip provides the greatest value/price ratio. I'm still a fan of AMD, but would switch to Intel in an instant if they matched their prices.

    Strict brand loyalty is a dangerous economic force. A history of quality products, more compatible 3rd party applications, and better tech support contribute to the value of a product, and should be considered. But factors like popularity (independent of compatibility) and company size/worth should be ignored. I honestly wouldn't care if Intel were the world's biggest, nastiest corporation -- it's the chips that matter.
  • Coppermine vs. Athlon? Am I the only person that keeps picturing a couple of Heavyweight boxers?

    Weighing in at 2 ounces...

  • For those less knowledgable amongst us, like me, could you explain the significance of Tom's oversight?
  • they didn't HAVE to go smaller to get more speed.

    Just a couple of niggles here. I'm not a hardware engineer, but I do believe that's wrong. At smaller feature sizes capacitors get more efficient and switching gets faster. Because your capacitor is more efficient you can use lower voltage. As voltage decreases so does power consumption, as the square. Less power consumption = less heat, so higher clock rates.

    They (Mot) did so they could reduce costs. The more chips you squeeze onto a wafer, the more money you generate from said wafer.

    Errr, somewhat correct. Yield plays a big part in the equation - as feature size goes down, so does yield, especially since new untried manufacturing processes have to be brought on line each time feature size ratchets down.

    The bottom line is that smaller features size is good - very good.
  • I'm not ashamed of it, and I don't really see why I should be. All I want is the best, and if AMD can't deliver it, then maybe they deserve to fail. And it's that extra 5% that keeps technology moving.
  • This isn't really news for AMD, they (and just about everyone else) knew that the Coppermines were coming out today. One thing to note, that is in AMD's favor, is that the .25 micron 700MHz Athlon is holding its own against the .18 micron 733MHz CuMines. AMD's Dresden facility is in the final stages of quality-assurance testing for the move to .18, and should be mass-producing .18 micron Athlons within 90 days. The fact that AMD's chip could easily hit 700MHz at .25 but Intel's couldn't says a lot for the Athlon design, and when AMD moves to .18 it should be able to scale up to much higher speeds, quite possibly beyond those of the Intel chips (and we already know that clock for clock, the Athlon is faster, especially in floating point). The smaller process ought to help out the Athlon's power consumption, as well.

    Not that Intel is taking this laying down, of course :) The enhanced core and full-speed cache on the CuMines is proof of this.

    Coppermine details and enhancements (note that these are pretty much ripped off of Sharky Extreme's coverage [sharkyextreme.com] of the Coppermines. Be sure to check out their site!):
    • 28 million transistors
    • 106 mm2 die size
    • 1.1 to 1.7 V operation

    .18 micron process

    Not only did they shrink the size of the transistors from the previous .25 microns (thus increasing the speed), but also moved to a fluorine-doped silicon dielectric for reduced capacitance, resulting in a further performance speedup.

    Enhanced L2 cache

    Rather than the 512KB, half-speed L2 cache that's been around in the P6 family for quite a while, the L2 is now 256KB running at full processor speed (and has been moved onto the same die as the core).

    Cache :
    • 8-way set associative, 1024 sets
    • 32 byte line (32 bytes data, 4 bytes ECC every 2 clocks, equals 11.7GB/sec throughput at 733MHz
    • 36-bit physical address space
    • 4 x reduction in latency versus Katmai P3 L2
    • Cache bus speed fully scalable with core frequency
    • 288-bit transfer width (256 data, 32 ECC)
    • 2 cycle back to back throughput

    Improved system buffering
    • 6 Fill Buffers (previously 4), increasing the number of concurrent non-blocking data-cache ops that can be done.
    • 8 Bus Queue Entires (previously 4) to allow more outstanding memory/bus operations.
    • 4 Writeback buffers (previously 1) for reduced blocking during cache operations.

    "Enhanced Power Management" for Mobile PIIIs

    While the smaller process results in lower power consumption, Intel has also added a technique they call Enhanced Power Management, or EPM. EPM essentially puts the CPU continuously into pseudo-standby mode, from which it can instantly accelerate to full speed when needed. This should lower power consumption further while maintaining the full capabilities of the chip.

    Packaging

    As well as the SECC2 Slot-1 catridge of previous PIIIs, the Coppermine will be available in a new format called "Flip Chip Pin Grid Array", or FCPGA, which offers lower power consumption and EMF interference, as well as being a less costly solution than Slot-1. Intel expects to move all the PIIIs to this format by late 2000.
  • Even then, the 667 Mhz CPU approxmates 666 Mhz.

    This is to say, the manufactor, while not being the evil one, approximates or tends to be so, which isn't that much different.

    Out of curiosity, do anyone recall any strange changes that took place at intel when people started dualing 333 Celeries?

    As observed something should be going on from there to make the company look what it is today...
  • It stands for "Flip Chip Pin Grid Array", and is the new format for the Coppermines (which will also be available in Slot 1). Intel expects to move the PIIIs over to FCPGA by late 2000 (and discontinue Slot 1 at that point).

    Advantages of FCPGA: lower cost, lower power consumption, and lower EMF interference.
  • I remember when AMD had quite a lot of respect a long time ago. I don't know what their big product was back then, but back in the mid-80s, when I was in grad school in the Bay Area, AMD was always in the news for their high-flying annual (christmas?) parties. They would rent out the whole Cow Palace and put out these huge spreads for all their employees. I was under the impression that this wasn't the only great employee perk they offered. What did they make back then that made them so much money?
  • I would just like to point out that:
    • If Intel ever sat back and decided they "have enough of the market share" their shareholders would sue them into oblivion... and remember it IS their legal duty in the US to maximize shareholder value.

    • Why should AMD give up the fight for the server market? Who says their engineers can't create a better server CPU than Intel? Especially when one considers that the server market has the highest profit margins. Remember these are not charitable organizations, they are corporations whose sole existence is to make money, and lots of it!

    • Last I checked CPU's have been making extraordinary strides in performance... but I guess these could be qualified as mediocre enhancements. In any case these companies will continue to do their best to satisfy consumer need. If you don't like what one is doing then vote with your dollars
    these just happen to be the realities of our capitalist soceity... and be careful when you knock it because although it's not perfect, the world hasn't seen a better alternative yet
  • why is there such a fuss in the x86 world over a 5 percent performance gain? If you really want a performance gain go get an alpha or a MIPS.
  • > So AMD had to stick with the Socket370(?) form (which was the non-proprietary Pentium Pro socket.)

    AMD has used the 'Socket 7' and 'Super Socket 7' sockets up until now.

    >AMD (or somebody) created the Slot A socket which is suprisingly(not!) similar to Slot 1 but not compatible.

    Slot A is dimentionally the same as Slot 1 (hence saving motherboard manufacturers money and resources), but uses the EV6 bus architecture, which was designed for the Alpha CPU's. EV6 allows for a 200Mhz system bus, scaling upwards of 400Mhz now on many Alpha motherboards. I think they're aiming for 600Mhz now.

    Keep in mind that Intel is still struggling just to get up to a 133Mhz bus.

    PS: This is only one bus, from one aspect..the entire system's buses aren't running at 200Mhz, but the capability is there for at least that speed between CPU and memory. Sorry, I'm not an engineer =)

  • The thing to remember (Score:1)

    by Anonymous Coward
    The one thing we all need to remember is that if you like Intel dropping it's prices, you better buy an Athlon. Otherwise, you can forget the dropping prices as this stint of competition will be over. Don't forget that AMD has bet EVERYTHING on the Athlon. They went deep into debt on Fab 30 and they have been losing money for the last several quarters. They have even announced they are going to sell one of their non-chip divisions which is profitable for cash to continue to fund their chip business. AMD is doing well on the technical side, but if they can't succeed with the Athlon, AMD is history. If it weren't for AMD there wouldn't have been low cost Celerons available to compete with their K6-2. If it weren't for the Athlon, Pentium 3 prices would still be much higher. This is not the first round of this fight. This fight has been going on for years, so many may be complacent that it will continue. You think AMD will always be around? Trust me, you will miss them when they are gone. If they lose this round, there won't be another one. Paul
  • That is SO true... Last time I read something that good was from classic literature.
  • ...has to do with how many memory different places in the cache each memory address can reside. This is not a measure of size, rather a measure of versatility. That is, in a direct mapped cache each memory address can be cached in one and only one cache line. This obviously leads to an overlap since main memory is alwasys larger than the cache. So, if a particular memory location is cached in a particular line, and another memory location that must be mapped to the same line in the cache is accessed, then the first must be flushed no matter how "fresh" it is.

    A two way set associative cache allows any memory address to be placed in one of two locations in the cache. A four way cache has four spots where each memory address can be cached, etc. Again this is not a matter of size, rather a measure of how flexible the cache is.

    The more "ways" the cache has the more flexible it is and this results in fewer flushes and overall more "fresh" data in the cache. This is what a cache is all about. That is, keeping the data that is needed right now, right here close to the CPU.

    These "X ways set associations" are expensive in terms of logic and chip space. Ideally, a cache would be fully associative and allow any memory address to be cached anywhere in the cache memory. Because this stuff is expensive, it is usually reserved only for the highest performance parts, that is, the level 1 cache which is the closest to the processor core and usually the smaller one. AFAIK, all mainboard caches are direct mapped. They compensate by usually being bigger and even thoough they are slower, they are still a good bit faster than main memory, but nowhere near as fast as a level 1 cache or register access.

    According to the previous poster, Tom got the two caches backwards. I don't have a data book on the new chip, but I'd really be surprised if they actually made an 8 way set associative cache that is 256K in size. No biggie, but it's an obvious error to those of us that know something of what the h*ll he's talking about.

    As far as his comment on the benchmarks goes, I have no idea where he's coming from on that one.
  • ...have been a "feature" of Tom's since the beginning. They are nowhere near as frequent now as they were a while back.

  • Ha ha (Score:1)

    by marcus ( 1916 )
    You took the bait.
  • It's this type of thinking that brings the world so many Disclaimers and Fine Prints. Nowadays everyone complains about anything, and if you can't find a fault with Tom's site or his articles, why not bash his spelling.

    No wonder there are so many Disclamers and Fine Prints in advertising and ... well, everywhere!. Even at the bottom of this Slashdot page!

    Note: the opinions expressed in the preceding statement are those of the author and do not necessarily reflect those of Slashdot or Andover.net.


  • Bingo! (Score:1)

    by marcus ( 1916 )
    I just bought a 19 inch monitor on the savings I got from buying the slowest proc I could find(350MHz for crying out loud!) instead of the fastest...I remember when PPro 200s were the fastest you could find. Sheesh!

  • I don't play slots. As I also don't play poker, blackjack, or drink alcohol. It's immoral. :) unity ps:the only way to fix it is to flush it all away. psps: :( I just ran out of southern comfort. damn
  • yeah but the coppermine is supposed to be the same design. so the optimizations should work for the other P3s too...seems weird.. ah well
  • If they suck so bad at making processors and are so damn good at manufacturing, maybe we should get apple to make AMD's chips. Think, a .15u Athlon right now. Immediatley scalable to probably 1Ghz. They'd trample all over intel.

    unity
    ps:the only way to fix it is to flush it all away
  • I don't play slots. As I also don't play poker, blackjack, or drink alcohol. It's immoral. :)

    unity
    ps:the only way to fix it is to flush it all away.
    psps: :( I just ran out of southern comfort. damn
    freakin' html.
  • Disadvantage of FCPGA: the name is extremely silly.
    --
  • Intel released a very disappointing profit report, which sent stock prices down as much as 10 points within a couple of days. Immediately following this, Intel makes several announcements about Willamette and Coppermine, and their stock goes back up. It seems that Intel may be pushing these chips a little too quickly in an attempt to make their investors feel better.

    While the announcement may have been shareholder-motivated, I don't think their schedule is. Their schedule may be advanced by AMD, but I doubt it will be at any detriment to Intel.

    Intel has always controlled the market and let the technology tickle out. I'm sure that if Intel wanted to, they would their next generation of IA32 chips on desktops now. That would be great for us, but bad for Intel. They want to get all of the intermediate upgrades.

    I mean, if you are the market leader regardless, why should you release the new chip until the current chip is sold through? Intel could step things up a notch, but they have had no motivation to do so (until now).

    I think that AMD's success is a great thing for the consumer. It should motivate Intel to get their chips to market faster, something that I'm sure they are capable of.

    For example, Intel has been producing 0.18 micron chips for many months. They were only available on portables until now. Why? It has to be a deliberate marketing decision to get all the money they can out of the desktop market.

    I hate marketing.

  • Re:Slot 1 vs Slot A? (Score:4)

    by hazydave ( 96747 ) on Monday October 25, 1999 @05:17PM (#1588370)
    Ok, first of all, forget completely about "Slot". That's just a hookup, a physical connector, it doesn't tell very much of the story.

    Slot 1 is the second physical delivery of Intel's P6 bus; Socket 8 (for Pentium Pro) was the first; Socket 370 is the third, and there's every indication Intel will market a Socket 426-or-thereabouts to handle some extra pins on a socket-based Coppermine chip. The reason you can't clone this without a license is simple: Intel has patents on the P6 bus. They originally didn't license, but now they are licensing, at least to chipsets (SiS, ALi, VIA) if not CPUs.

    AMD extended Socket 7 (which is once of the physical conventions for the non-patented P5 bus) to 100MHz, and ran their K6 family there. But they realized that K6 would have its limits. In part of their IP settlement with Intel, they promised not to make clones of things like the P6 bus, so legalities for everyone else aside, that wasn't an issue.

    Enter DEC (now part of Compaq). DEC designed the EV6 bus for the Alpha 21264, as a remedy for conventional CPU buses. In their earlier Alpha systems, even with L1..3 cache and all, they were so dependent on memory speed, the typical bus sharing in an SMP system (very important to DEC) was a problem. Especially when you wanted to make this bus (the cricital CPU to rest-of-the-world link) extremely fast. EV6 is a point-to-point bus; all you have is a CPU and some system chip on it, never anything else. This allows them to run EV6 very fast. DEC openly licenses EV6, AMD adopted it for their CPU. Since the PC market demands a socket or slot, they created Slot A, which (for cost reasons) uses the Slot 1 connector turned around. AMD runs Slot A at 200MHz now, 266MHz in the forseeable future. DEC runs EV6 systems up to 400MHz. This is a data rate; the bus runs DDR (a new data event on every clock edge, not just every clock cycle), which really doesn't matter; the data rate is indicative of performance.

    The important thing to realize about today's Athlon systems is that they're something like the first Pentium systems shipped out, retrofitted '486 systems much as the first Athlons are running based on modified Super7 chipsets. So memory is stuck at 100MHz, half the CPU interconnect's speed. You won't see Athlon reach its performance potential until 200MHz, or better, memory systems are delivered.

    And I do mean "or better" because of the EV6 architecture, there's no shared CPU bus. So a system chip (North Bridge) can actually use memory faster than the CPU can deal with. With switched interconnects, proper buffering, and super fast memory, one could imagine PCI, AGPx4, and multiple Athlons all rompin' along, each at virtually full speed. This can't be done efficiently within the current P6 architecture. That's why I find Athlon interesting. I hope AMD lasts long enough for it to realize its potential.
  • Xeon was originally made for three reasons:

    1) room for much more of the expensive L2 cache for full speed operation and modules up to 2MB.

    2) redesign to the AGTL+ bus, which allows four processors to run, even using modules (versus the normal P6 bus, delivering two way SMP with the module, four way with the older socket).

    3) more money. Intel knows that people who need 4-way or better SMP systems will pay for this, often to a foolish extent. Intel loves to milk some sector for high margins, and you can guess this won't happen when there's a direct (or near so) replacement like K6 or Athlon.

    Ok, so now consider PIII Xeon with only 256KB of L2 cache. Certainly, this is the same chip as you get when you buy a PIII-regular. But of course, it's on the Slot 2 module, and if you want a four processor (or better) SMP system, you have no other choice. Intel basically has you, and they like it that way.

    In fact, I'm surprised it only around $50.

  • Yes. They are. Though to be fair you have to also consider that they typically pick (if I remember correctly) names of streams and rivers to name the experiments by- apparently there was a Coppermine River or something, so naturally this name they're keeping.

Slashdot Top Deals

"Your stupidity, Allen, is simply not up to par." -- Dave Mack (mack@inco.UUCP) "Yours is." -- Allen Gwinn (allen@sulaco.sigma.com), in alt.flame

Close