Forgot your password?
typodupeerror
Intel

News on Pentium IV 309

Posted by CmdrTaco
from the cpu-for-you-cpu-for-me dept.
MotaK writes "Ace's Hardware and ARS-Technica has reported on PC Worlds article on the Willamette Processor, dubbed the P4. This proc. will apparently be only a 200Mhz frontside bus, and launch sometime in 2000. "
This discussion has been archived. No new comments can be posted.

News on Pentium IV

Comments Filter:
  • Come on Intel, isn't it time to retire the Pentium name? As if Pentium II and III weren't bad enough, now we have to deal with the Pentium IV. Even Williamette would be a better name.

    If it wasn't for the Itanium, they'd probably still be calling their chips Pentiums 10 years from now...
  • by StromThurmond (112168) on Tuesday December 14, 1999 @11:28AM (#1465068) Homepage
    AMD has had a 200 MHz FSB since they released the Athlon. Intel is only now getting around to it. I also heard recently that in AMD's Dresden fabrication plant that they are turning out 1.2 GHz chips while Intel is only /talking/ about getting over a GHz (w/o cooling). I think that Intel's star is falling and that the Athlon's superior performance over the P3 was no fluke. (Also that is one mighty unoriginal name)
  • Come on Intel, isn't it time to retire the Pentium name? As if Pentium II and III weren't bad enough, now we have to deal with the Pentium IV.

    They were, in all likelihood, just dodging the inevitable "Sextium." Can you blame 'em?
  • Um I could be wrong but I think I remember reading here that the next pentium is a 64bit processor. Can't find the story at the moment. Any one have more insight?
  • by jd (1658) <imipak AT yahoo DOT com> on Tuesday December 14, 1999 @11:31AM (#1465072) Homepage Journal
    Yet another outdated, underdone processor, that will probably cost far too much and sell because of it's name.

    If the money had been put into decent development, we might be seeing high-speed, 64-bit, dynamically configurable processors -today-.

    (As it is, we'll have to wait another 3 or 4 weeks for Transmeta to release theirs.)

    Alternatively, we could have seen, for the same cost as developing a new processor, Intel, or one of the other premier chip companies, experiment with preparing silicon or gallium-arsonide in microgravity. A slow-cooled batch, under those conditions, would be near flawless, and allow for some serious clock-speeds. And people -would- buy! I'm sure of it!

  • The new lines of processors focus WAY too much on the actual speed of the thing, mainly because that's the first (and sometimes only) thing people look at when they buy a computer..."Never mind that it only has a 33mhz bus, it's a 750mhz processor!"...The creators should focus on reducing power consumption, upping the bus, or making other parts more efficient.
  • I know it's not strictly on topic, but has anybody else noticed that the stories in the Slashdot Ars-Technica box haven't been updated in weeks?

  • Wow, news about news about news! Keep flying around the web like this, and it starts to resemble a sick game of Operator!

    "Tom's said that Ars' said that ZDNet said that Wired said that PCWorld said that Jane's said that... Corel is buying Amiga? Woo-hoo!"

    Seriously-- should Intel be concerned about the Japan market and their 4-aversion? Are they going to call it the Pentium III+?

  • Does this mean that my 999Mhz Craptronics i386 will not run my Microsoft BloatOffice faster then ever? ;)

    I think we should start asking computer salesmen to see the SPEC95 ratings of the chips and see what they do. : )

    -Sean

  • Yeah, who needs faster CPU's? I use a 4Mhz Z80, and it only takes a week to compile a new Kernel! Why would i want to do anything faster?
  • As much as it may be cool to have a 200 MHZ FSB, what will you do with it if your memory is 100 Mhz? I actually own a K7 (FSB = 200 MHz) and don't even think my board/CPU will let me use 133 MHz RAM (that is, without a soldering iron). Anyone knows about 200 Mhz SDRAM???
  • Intel is now playing catch-up with AMD. The Athlon should be the first to work with a 200Mhz bus, and they are working to put the cache on the chip, just like the old P-Pro's. This should be a big smack upside Intel's head. Intel is going to have to get in gear if they want to stay the leader in PC chip manufacturing.
  • This is the dreaded Athlon killer that is going to put AMD out of business. Sell your AMD stock now and enjoy the profits while you can, Chipzilla is on a rampage!

    Be afraid, be very afraid... of chip "errata"!
  • by billybob jr (106396) on Tuesday December 14, 1999 @11:35AM (#1465083)
    From Intel, analysts expect 800-MHz PIII chips during the second quarter, and the Pentium IV chip, code-named Willamette, by the end of the year. Willamette is expected to handle more simultaneous instructions than the PIII, and to break the 1-GHz barrier. The accompanying chip set will likely support a 200-MHz system bus, like the Athlon's.
  • by Anonymous Coward
    The AMD Athlon has Intel scared speechless. AMD now has the lead; Intel must announce something, anything.. Last night while surfing cable TV I came across a Home Shopping Channel which was selling complete AMD Athlon systems for $1500. Not only that, but they were really pitching Athlon as the world's fastest IA86 processor. Lot's of graphs, benchmarks, and carnival style barking. Ouch! That smarts!
  • that the PC World article made no mention of Motorola and IBM's processors (specifically the G4 and G5). I know, it's a Wintel-focused magazine, but an article that purports to be about "the processors of 2000" should be about _all_ the processors that are going into personal computers. That being said, I do applaud PC world for recognizing that the drooling masses are looking for nothing but more MHz, and that chip makers are capitalizing on this.

    I also must express my grave doubts that Intel will be able to ship a PIV in the year 2000. From what I've heard, they are really pushing CISC as far as it can go, and they're having a lot of engineering-type trouble (the PIV allegedly draws huge amounts of power). I really don't expect to see a stable system based on the Pentium IV until mid-2001. But that's just my opinion; I could be wrong. Anyhow, give me an Athlon or a G4 over a Pentium anyday. 'Specially when them LiuxPPC folk get some Velocity Engine support for the apps

  • by aUser (78754) on Tuesday December 14, 1999 @11:36AM (#1465086)
    I couldn't disagree more with what they write:

    "Itanium won't be relevant for PCs until 2003 at the earliest, more likely 2005,"

    Especially because their main argument is:

    "Software has yet to be written or recompiled to accommodate a 64-bit processor. "

    So, they really think that Microsoft will be able to hold back computer progress once again? Last time, there were no alternatives. You would run 16-bit dos, like it or not, but now we have choice and freedom.

    If Microsoft doesn't move fast on this one, they'll be losing market shares on the server end of things in no time, and increasingly rapidly on the desktop too.

    If they move fast, however, they will have to abandon their huge installed base and start at the same level as anyone else. It would be dangerous position for a company that has never excelled at quality and innovation.

    I think 64-bit architectures may very well turn into Bill Gates' Waterloo; and it will all be over before we know it.
  • by waldoj (8229) <`waldo' `at' `jaquith.org'> on Tuesday December 14, 1999 @11:37AM (#1465088) Homepage Journal
    I'm always in a bit of shock after I talk with people that are relative newbies (ie, less knowledgable than us folk :) about computers. They'll tell me something like "I got the 12.2 gig RAMs with a 15" drive and 256 of cache. Of course, it's a Pentium III. They're the fastest, you know."

    Explaining that the speed increase between a P2 and a P3 is negligible doesn't seem to help any. They just know that they've got to have that "III," they have to have that "MMX" on there.

    I wonder how much longer it will continue like this? Remember the rabid VCR market in the late 80s and early 90s? Maybe it was just my young geek mind, but it seemed to me that there was always some new feature, some great reason to upgrade to the next great generation of VCRs.

    Where has that gone? In part, DVDs have increased the upgrade crazy, to be certain. But, if we go back a couple of years, you'll notice a laid-back attitude about these simple devices. "Does it record? Good enough." Again, this might just be my perception.

    This same pattern has been repeated with radios, telephones, TVs (to some extent), coffeemakers, refrigerators, and just about any other overpriced gadget that you can think of. They become a commodity.

    Is the processor industry going to drive itself into commodity status? I assume so. No doub t that you and I will keep right on upgrading and getting the biggest backside cache and fastest motherboard speed that we can. But as computers are reduced to appliance status, consumers won't care about whether it's a P3 or a P4. They'll just want to know if it's "on the e-mail."
  • by Greyfox (87712) on Tuesday December 14, 1999 @11:37AM (#1465089) Homepage Journal
    Sounds like AMD's really planning on beefing up their chips, and Intel's playing catch up for the foreseeable future. Also sounds like Merced performance is going to be lackluster at best (We all knew that anyway, right?) If I were an Intel shareholder, I'd be selling...

    Oddly, no mention of the Alpha, which I expect will end up owning the 64 bit arena for the foreseeable future. Why wait 'til 2003 for 64 bit performance when you can get it today? And the alpha today will probably be faster than Intel's deliveries years from now.

  • And I never thought there was any problem with part numbers anyway. What would the P4 be, the 80986?

    I guess it ould be worse, they could use the Microsoft version numbering scheme. Pentium '99 anyone? How about the Xeon 2003?
  • Unless you're relying on your CPU to do all your Mesa rendering in Quake III (ouch), this kind of speed it overrated.

    What can the average person expect to see out of this? Not much, that's for sure. Maybe their MP3s will compress a little faster... I've got nothing against speed (a lot of places still need much more of it) but it's something that the Desktop PC market just doesn't have any use for any more of.

    What we need now is innovation. Speed could only take us so far, now we need a brilliant flash of insight into making computing different... Better. What's the next step?

    Use that speed. Better voice recognition. Artificial Intelligence systems that can figure out the difference between what the user says, and what the user wants. A machine that understands subtleties... a 3D desktop environment. Wireless T1-level access across the nation. WHY ARE WE STILL COMPUTING LIKE WE DID A DECADE AGO?

    -----------

    "You can't shake the Devil's hand and say you're only kidding."

  • by BranMan (29917) on Tuesday December 14, 1999 @11:37AM (#1465092)
    I think the author of the PCWorld article has Windoze on the brain. When the Merced chips hit the streets you can BET that there will be a Redhat distro for it the same month (if not the same day).

    It may be true that you won't be able to get Win200X (Specially built for Pentium IV! Upgrade today!) for quite some time, but the folks that just need to recompile like Linux, or BSD, will be there instantly. Slashdot itself has run the articles on Linux booting off a Merced simulator, and GCC being ported to run as well.

    With Transmeta, 1.2 GHz Athalons, and Merceds coming out, next year should be very interesting.
  • Ah, skip to 7 and call it Heptium. Then when people ask where was the sextium, tell 'em it was the Xeon or Pentium Pro.

    Come to think of it, wasn't the Pentium Pro the actual sextium then? The Pentium's are all P5, and I faintly recall that the Pentium Pro goes by P6...
  • by Anonymous Coward
    intels next processor is called itianium. It is 64 bit and all that other cool shit. But for these chips to be viable they need an operating system to handle them. I know that microsoft claims to have a 64 bit OS in the works but who the hell knows when we will see that. I know that BeOS is 64 bit, can someone tell me some others that are.
  • Maybe they're just holding out...
  • Really, I'm hardly an AMD supporter but this is like saying "ford will release a car in 2001 that may perform as well as last year's chevy". My next CPU is a 1ghz athlon... I'm done with Intel. For now.
  • the world's fastest IA86 processor

    Not to be pedantic, but i'm sure you mean IA32, or i386...
  • I REALLY wish I could convey to more consumers out there that MHz is not a true metric. I have seen friends and family wander down the path of: "I don't know what this means, but the numbers sound high and the price is good..."

    It is going to get really interesting soon even for the people who understand computers better, as we will find that are choices are opening up. How are we, the folks in the "know" going to convey to those not so technically minded what all of this really means? The human mind tends to grasp 5 plus or minus 2 strange things before it starts to overload. There was once a time (1992 or thereabouts) when I could fit advice in the form of:
    1) High MHz
    2) Intel-based CPU
    3) not Packard-Bell
    4) high RAM
    5) more hard-drive space is good
    6) get CD-Rom
    7) latest version of DOS/Windows

    Now there are so many factors to take into account that I almost have to walk my computer-illiterate friends through the process of buying a computer. It doesn't look like this is getting any easier either. In the "old days", the person usually got at least 4 of the above criteria right, and was OK. *Sigh*

    As we continue along this path, it is great for optimizing our tools, but we leave the general populace farther and farther behind. If only there WERE an unbiased metric that people could use to diagnose their needs compared to the products offerings... Maybe there is and I am simply not aware of it.

    B. Elgin

  • by Lotek (29809) <Vitriolic@gmai l . c om> on Tuesday December 14, 1999 @11:41AM (#1465100)
    Well, gosh guys, no wonder intel chips are slower than AMD chips. They run Roman Numerals for crying out loud. It probably takes a few more clock cycles to add XX, IV, M and then divide by V than to use that pesky binary stuff.
    Personally, I am waiting for the Pentium XXVII, which should be a real screamer.
  • Kind of a kooky statement from this article. Sure, there may still still be 16-bit code in Windows in 2003, or 2005... the rest of the world will be recompiling their distributions ASAP.
  • Linux runs on the Merc.. Uh, Itianium. I don't know about BSD. I had heard (Maximum PC) that NT is supposed to run on it as well. I have however, also heard rumors of NT not running properly on it. (I remember hearing about Intel showcasing Linux on this chip because Windows did not work. Sorry, don't have a verifiable link for this.)
  • What's next, "heptium"? what about "aquarium"? Next generation: "octopus"!
  • I think perhaps the Pentium V will be called the Pentipentium.

    I can't wait for the Dodecipentium. The chip will be fast!

    -m@
  • I've never heard of this before. Please enlighten me/us. :-)

    paranoid.android
  • Athlon really doesn't yet have a chipset to run with. Sure, there is the AMD chipset, but it is far from what could be done.

    Once the KX133 chipset from VIA is available (provided it's not a complete dud) and once KX133 motherboards start to appear, Athlon's lead over Intel would look much uglier (for Intel). That should happen by Jan '00. I would expect that with the proper chipset the whole-system performance advantage of Athlons vs Pentiums would be very very noticeable.

    Note to Intel: it's not time to panic yet, but recommend going on yellow alert.

    Kaa
  • Actually, I have it on reliable authority that in 2001 Intel will switch over to their Hurricane-style naming scheme. The first name on the list is "Abe"

    "Wow! Glenda over in marketing just got her new laptop. It's a 33 GHz Barry!!"

    -----------

    "You can't shake the Devil's hand and say you're only kidding."

  • I dont understand the fuss about 666. The number to worry about is 10314424798490535546171949056.

    For those not in the know, go read Heinlein's The Number of the Beast.

  • Hmm, I've already been using a P5 for quite a few years already. I'll hold out a bit longer.

    Yeah, and by that logic, I'll also assume that a Macintosh G3 is faster than an IBM G2 mainframe. Heh heh.

  • "...this addition to the PIII design is responsible for supercharging Microsoft Office apps."

    For some reason, I had trouble taking the rest of the article seriously after I read this line.

  • >>I think we should start asking computer salesmen to see the SPEC95 ratings of the chips and see what they do. : )

    Most likely they'll tell you no. It's the manufacturers job to put out specs and benchmark results, not the salesperson's.

    (Yes, I sell and fix computers for a living)
    LK
  • I know it's not strictly on topic, but has anybody else noticed that the stories in the Slashdot Ars-Technica box haven't been updated in weeks?

    Same with the segfault.org box...that's less important though, clearly, since the only reason to ever read segfault was for the weirdass trolls in the comments.
    --
    "HORSE."

  • by NovaX (37364) on Tuesday December 14, 1999 @11:50AM (#1465116)
    Well, lets see. Athlon has a nicer bus because they licensed it from Digital. How long were they on the socket-7 system? Intel converted over, and now will go to a 200, from the current 133mhz bus. Since Intel is just a tad behind, but AMD's Athlon is still quite new, its not something to get crazy about. The 66mhz bus was a real pain when overloaded, and so I doubt a 200 will be to stuffed by most of us. Still, don't believe AMD's bus is better by AMD's engineers, its because of DEC's, who already had EV7 designed and ready to be implemented when the licensed EV6 (whether they did, I don't know).

    Both Intel and AMD have talked or showed them off with cooling, never showing without. IBM has had the PowerPC at 1ghz for a long time, in lab, just like the rest of the industry. I'd be surprised if any of them were still needing to test 1ghz with massive cooling...

    Whether AMD has 1.2ghz chips cranking out, I'd be surprised. AMD has said before it could be at 900mhz right now, but why rush when you can *squeeze* the market? Intel did that, and AMD is just playing the same mhz game with us. Oh yeah, I love AMD, they play the same game as everyone else. When your small, you fight the bigger capitalist by creating better products and overshadowing them. You win. AMD believes they won, maybe they have, but they wont be saints, or demons (or daemons :).

    Superior performance is due to superior design, one thatcame out 3 years after Intel's. The P4, whatever it will be called, has been in the works for 3-4 years. AMD spent 1.5 years on the K7. Now, whether the P4 is any good (considering Intel spent all its engineering time on the Mercury.. ehh, name before "Merced" on Intel's media-news page), I don't know. I'd be more than happy to see an amazing deign come out again to rock the market. If Intel blows the P4, its in serious trouble.
  • by Anonymous Coward
    From what I've read/heard, Williamette has a 100 mhz QUAD pumped fsb. I thought originaly that it was 200 mhz Double pumped, but a friend of mine working at Intel was telling me that it's 100 mhz quad pumped. For you non technical people, thats the equivalent of a 400 mhz FSB.
  • Here is a good place [rcollins.org] to start looking for ideas -- notice that Itanium was registered a while back...
  • For immediate release:

    The Inhell Pentagram processor, 666 Mhz. Sure it runs hot, but there is no need for cooling (cold day in hell? Ever seen one?)
    You have never seen daemons dance so fast.
    (Holy water user protection not included.)
  • by Haven (34895) on Tuesday December 14, 1999 @11:52AM (#1465120) Homepage Journal
    The magazine is about Personal computing. Thats why they aren't talking about the Alpha 21364. The G5 has nothing to do with personal computing, and the G4 is already out.
  • by Bill Currie (487) on Tuesday December 14, 1999 @11:52AM (#1465121) Homepage
    Two banks of memory operating in tandem so that each bank is operating at only 100MHz. Will only give you the performance in burst mode access, but as caches tend to work that way (AIUI), that's not such a bad thing.
  • My Ghod! They must be writing the microcode in Intercal [tuxedo.org]!!!
  • This is a posting experiment, please ignore
  • Very good points: we are at the point where the speed of the processor doesn't matter, but now it's the amount of RAM and the speed of the media that will slow down the program.

    And what's very sad is that with faster and faster chips, application programmers think that they'll be developing for thoses really fast computers, and more and more of today's programs have a lot of chrome that probably is fine on a P3, but on low end machine, the chrome slows down the program. Sure, I expect that a billion FPU calculations will take a longer time on a slow machine, but there is no reason for the chrome to drag to a crawl - chrome-type feature should be slim, trim, and ideally optional.

    (And this isn't just happening with Wintel users, while both KDE and GNOME are great efforts, there *is* a lot of chrome that slows down their basic function on a low-end linux box).

    At this point, we shouldn't be getting too excited about faster processors; the next step in processor speeds will require nanotechnology to stay within heat sink and FCC limits. Instead, we should be focusing on optimizing the processor usage within programs.

  • I believe Alpha Linux is (I may be incorrect)
    But even if its not, it wont take long to switch it over.
  • aka, PentiuMM.


    Pope
  • Come to think of it, wasn't the Pentium Pro the actual sextium then? The Pentium's are all P5, and I faintly recall that the Pentium Pro goes by P6...

    Actually, Pentium IIs/IIIs/Celerons are just beefed up PPros. And Xeons are just beefed up (more cache) PIIs/PIIIs. So Intel hasn't really made a *new* chip since the P6, or PPro.

  • Why not just call it the "Hexium"? I think Intel stuck with the name Pentium because Pentium III sounds much more powerful than Octium ... (besides, those three I's stand for Internet Internet Internet)

  • by Woody (1159)
    Questions: What speed will this debut at? 600 MHz, 800 MHz, 1 GHz? Will it basically just be a PIII with a faster FSB and core speed? Is there anything new in the "P4" architecture?


    Alright now that that's out of the way, how about motherboards? Processor configuration - Socket 370, Slot x, something new? I'm really getting sick of Intel doing things like deciding that the new 370 Coppermines won't work in older 370 boards. The processor race has got to drop off sometime...
  • It has been this way for a while... The pundits are to blame.

    When you really don't know where to look to find out what empowers your computer to actually do what you want it to, you go to salespeople and advertising. The industry ads would like to convice everyone that this or that "wonderful invention" is necessary or the tool is garbage. Most people don't know how to upgrade their car, why should they know how to upgrade a computer? If you talk to salespeople, they want to sell you the most expensive item you will buy!

    As is the case with cars (my metaphor for the hour), people will buy features they will not need (like six cupholders for a two seated car...), simply because they are convinced that the features are useful or they want what comes with the package. How many people actually compare the horsepower of the cars they look at? Not that many. They go for the name brand of the hour. It will continue to be the same way with processors: "I don't really know why it is better, but the salesman told me it was."

    The P4 will be bought. People will not understand. People will be satisfied / dissappointed / upset / resigned / utterly confused. That is just the way things are likely to flow.

    B. Elgin

  • by SEAL (88488) on Tuesday December 14, 1999 @11:58AM (#1465134)

    As a consumer, I just can't WAIT to get my hands on one, Intel. I'm just seething with anxiety over new features [zdnet.co.uk]. Heaven forbid that they should f00f [ddj.com] this one up.

    Of course it's all simple mathematics [ddj.com], really. The more [ddj.com] Intel rushes things, the more likely AMD will come out ahead.

    :-)

    Best regards,

    SEAL

  • What I find odd is how consumers can stand using the crap they get from HP or Compaq or ... I hand-built my P2-350 and it comes up (when I run windows - 95) faster by 30-40 seconds than newer P2-450+ machines from the named companies.

    I think the "upgrade craze" will continue until the difference between this computer and that computer is not distinguishable except for the number of buttons on the front or how the case looks. Here at work I use a P200 and I'm happy as a clam, even with my 13" monitor (I have a fifteen sitting unplugged 3 feet away from me :)

  • by Anonymous Coward
    Intel reminds me of the american car companys during the 60's, They are increasing speed by bumping up the MHZ, kinda like bumping up desplacment, we need to see more efficient "import" processors.
  • So, they really think that Microsoft will be able to hold back computer progress once again? Last time, there were no alternatives. You would run 16-bit dos, like it or not, but now we have choice and freedom.

    Uh...there was a 32-bit alternative back then. It was called OS/2, and it could concurrently run 16-it Dos apps and native 32-bit apps at the same time.

    Hopefully history will treat 64-bit alternatives better.
  • by Rombuu (22914) on Tuesday December 14, 1999 @11:59AM (#1465138)
    What we need now is innovation. Speed could only take us so far, now we need a brilliant flash of insight into making computing different... Better. What's the next step?

    Great question, easy answer: 5 words -- ubiquitous wireless broadband internet access.
  • Voice recognition isn't gonna be a big event, no matter how good it gets. Do you really want to sit in your office, dictating code/email/memos to your computer while surrounded by 20 people doing the same thing? I certainly don't!

    But with the computing power we have now, you're right--there must be some ways to substantially change our computing environment.
    How about going back to monitors imbedded into the (physical) desktop, but using LCD displays now? Far more realistic. No reason that touch screens can't be brought into play, and in a more intelligent manner. (example: A desktop would know that it should ignore the input from a stack of papers laying across it)


  • Artificial Intelligence systems that can figure out the difference between what the user says, and what the user wants.
    Um, show me a natural intelligence system that can do that. Those sorts of things have been the root causes of many a fued and war. If anybody ever comes up with such a system, the world will definitly be a better place.
  • Yeah...but if they did that they might get suid by M&M (or should that be MM :) the official candy of the New Millenium.

    (How does one get endorsements from the millenium anyway?... do you have to go to the Vatican for it? They did set the calender after all [salon.com] )
    - Reunite Gondwana-land
  • Like AMD's IIRC amd's bus architecture was vastly different from Intels. Does this mean Intel is copying AMD? Wouldn't that make Intel the follower in the race?

  • Yup, they already exist. Well, I don't actually know how far along production is but there are chips being shipped around to OEMs and such. Some sites have had pictures (it's a chip, surprise surprise) and ArsTechnica has already gotten info on it. Like the fact that it won't work on current PPGA mobos for one. (There's rumours of converters in the works though.)
    What I _haven't_ heard about yet though, are the 'lower-end' Athlons I've heard AMD is supposed to be making.
    And then of course there's whatever little gem Transmeta is cooking up.
    Ah, I'm tired of playing the waiting game.
  • ...Intel's latest Coppermine PIIIs have high clock speeds plus a 256KB on-chip secondary cache--and this addition to the PIII design is responsible for supercharging Microsoft Office apps.
    What does this mean!?!? No other apps are supercharged by it? Only the Coppermine P3 supercharges M$Office?

    First quarter: 800-MHz AMD Athlon
    Second quarter: 800-MHz Intel Pentium III
    Late second quarter: Intel 64-bit Itanium processors for workstations and servers
    Late second quarter: Intel Timna for bargain PCs, Via chips for bargain PCs
    Fourth quarter: 1-GHz AMD Athlon, 1-GHz Intel Willamette
    Does it strike anyone else as odd that Intel is a quarter behind AMD now but they are promising the 1 Ghz processor at the same time? Sure, they could catch up, but I don't think they will so quickly.

    Me, I'll stick with AMD. They make the upgrades I'm putting into my old systems, and they are making the Socket 7 processors that are inexpensive as hell. How about $160 to replace a motherboard and put in a 400Mhz K6-2!!! Not a gamer's box, sure, but one hell of a system for my home office users to check e-mail and browse the web!

  • >Actually, Pentium IIs/IIIs/Celerons are just
    >beefed up PPros

    I dunno enough about CPU architecture to comment on the P2 or Celerons (well, beyond that they added MMX to the regular PPro instructions), but the P3 did actually add some new instructions above and beyond that. Not that they're really used by anything yet that I'm aware of, but it's more than just a "beefed up PPro".

    Besides, the primary (ie most significant to the majority of users) difference between the PPro and the P54 class CPUs was the location of the L2 cache (on the chip vs. on the board).

    I'm still more interested in seeing what other CPU makers do than yet another rehash of the ix86.

    -LjM
  • by Anonymous Coward
    In Japan, the number 4 ("shi") is unlucky because it is the same word for "death".
  • your kidding about "athlon's superior performance over the p3" right? in every test i've seen the p3 outperforms amd at the same clock speed. i have a computer with an amd in it and it just gathers dust, even though its clock speed is faster than my other computers.. it just doesn't run as fast. amd may be out of the group with crappy processors like cyrix, etc. but i still don't think its anywhere near intel.
  • The Pentium was developed because in 1993 we were in the depth of the worst recession since 1980 and engineers had two choices: work their asses off and get some major breakthroughs or join the thousands who were getting laid off. Since we don't have those pressures there's no need to develop anything new. Not until we have another crash like 1993 will there be anything phenomenal.
  • You know I'm really sick of all this hoopla over these processor races. Its like thats all that important. How about the recent shake up in the domain registration world. I mean the network solutions monopoly has been broken! If you don't believe me just visit this site www.npsis.net [npsis.net] and check out the going price of a domain name these days. Now that is NEWS!! Now guys like me (the poor folk who actually work for their money) can afford a lousy domain name. Come on people lets stop focusing all are attention on the processor race and look at other more significant events that are happening with the internet and the world every day.
  • I don't have the data for this handy, but it would be interesting to plot and extend trend lines to see when the roman numerals for Pentiums would pass the numbers for movie series like Star Wars, Star Trek, or Halloween.

    Of course, there could be some crossover -- The Revenge of the Pentium, Pentium in Love, The English Pentium, The Pentium Redemption, Indiana Jones and the Holy Pentium, perhaps.
  • Hmm... Your reasoning appears to be: Alpha is faster than Intel's CPUs today, so Alpha will be beating Intel for years to come. Well, forgive me, but I don't see the logic here. The EPIC architecture behind the Itanium is a radical departure from the RISC architecture that the Alpha is based on. EPIC allows for *much* more ILP than a traditional RISC design can achieve. I really don't see any way to significantly improve the performance of the Alpha other than further increasing the clock rate. Simply adding more integer and FP units is not going to cut it since the architecture does not allow the compiler to make parallelism explicit to the CPU. EPIC does. However, I suspect that it will take Intel some time before they have the compiler technology required to take full advantage of this architecture. I would be willing to bet that by the time Intel is on the Itanium II or III, that they will be trouncing all over the Alpha.
  • I believe that what you're referring to is the story that the Japanese words for 4 and 9 sound similar to the Japanese words for "sickness" and "death". Consequently, hospitals in Japan have no rooms with the numbers 4 or 9 in them. I don't know whether this is true, or just an urban legend, but that's the way I heard it.
  • Not anyone attribute does a good system make. However, the average joe consumer does not realise this. They are caught up in the malestrom of advertising and hoopla. Remember in the EARLY 80's when 300 baud modems were it, floppy disks were literal, and you were DAMN lucky to have color! Programer has very little to work with (looking back). They had to write tight efficient code or not at all. Now any fool can write a bit of code and processor speed makes it all even (on some levels). If MORE people who were writting usable code tighten it up, things would increase on their own. If people would take memory into consideration when writting, etc... I think the "I got a 12 gig HD for my PC" would also deminish. What we would be left with is an industry that would HAVE to make marked improvements. Instead of just jazzing up the existing material.
  • by um... Lucas (13147) on Tuesday December 14, 1999 @12:29PM (#1465171) Homepage Journal
    Try the engine metaphore.

    I always tell people that the MHz is more like RPM's and not horsepower...

    P-III = V6
    Athlon = V8
    G4 = V12
    Alpha = nitrous burning funny car engine! :)

    My saturns rpm-ometer goes up to 9,000 RPMs... My friends mustang goes to i think 7. Mine needs 3700 RPM's to get to 85, while the mustang needs 1900.

    It's a fitting match!

    Sorry if i went a little off topic though :)
  • Anything new? You mean new instructions that need more transistors that no one will use anyways? I hope not!
  • by JamesSharman (91225) on Tuesday December 14, 1999 @12:32PM (#1465174)

    For a start I was a little un-impressed with the article. The news item had me expecting some specific information but the article had few specifics.

    Several of the posts here I've read seem to imply that AMD is soon to get the upper hand and that Intel is playing catch up. This is unlikely to be true for the foreseable future, the lead Intel has will stay with us for a while yet regardless of the competition.Here are some of the biggest issues surrounding the market.

    Intel Brandname: Many of you will remember that not that many years ago only the total techies like us actualy new who made the processor inside your computer, the brand was the manufacturer, how many of you today know who made the chips in your mobilefone? your pda? your wristwatch? your microwave oven? Intel changed all this with an extensive and heavy duty branding efort (The Intel Inside stickers, the TV adverts and all that). AMD has not got a brand name anything like as big as intel, the general public 'Knows' that Intel is the best even when it isn't.

    AMD Will slash Athlon Prices: This is one of the biggest misconceptions about, for a long time AMD has been known for being the cheap option, not because they could design or makes the chips vastly cheaper than Intel but becuase they had to cut costs to compete. This all changed with the Athlon that for the first time put them ahead in the performance stakes, the Athlon price is now in the same bracket as the Intel chips and it can't cut the prices by mutch, the development costs have really cut into their revenue to the point that they have no choice but to compete on a level price playing field. That leaves AMD fighting only on the performance and branding front's and they only have edge in one of those areas.

    AMD Has The Performance Advantage: True, but for how long. Many people have said that AMD has now proved they can make better processors and will continue to have the edge. This is not necisarily true, Intel is very large with vast resources. Basicly AMD have cought Intel napping, for far to long Intel has has a clear lead in the field and they have got more than a little complacement. The Intel Brandname will only cary them for so long, Intel knows they will have to fight to regain the performance edge, and Intel knows how to fight. I am reminded of the over quoted Admiral Yamamoto after pearl harbor "We have awakened a sleeping giant and have instilled in him a terrible resolve". And I'm sure you americans will be quick to point out that not eveyone caught napping is the inferior side.

    Natalie Portman: Has nothing to do with this issue.

  • And what OS will run on this machine... You'd have to get rid of all your legacy software, pretty much revamp the linux kernel, etc...

    Unless, you switch to either the BeOS or Mach kernel... and then run Linux on top of Mach.... of course as i often say, I am not a programmer so i may be completely wrong.

    I'd see no advantage of switching from one fast processor to two slower processors... To two fast processors, yeah it's worth it if my apps care about the second cpu. What do i care about the heatsink on top of the chip? I usually keep the case closed, so it doesn't really matter all that much to me.
  • by roystgnr (4015) <`roystgnr' `at' `ticam.utexas.edu'> on Tuesday December 14, 1999 @01:00PM (#1465196) Homepage
    Enough with the oxymoronic names already! "Pentium IV"? Please. You'd think that Intel could make it's chip names correspond a little better to chip differences, too. The only significant difference between a PII and a PIII is the (little used) SSE support, whereas you have to shop carefully between PIII releases to make sure you get one of the uber-cache "Coppermine" (disclaimer: no copper included) models that actually can compete with an Athlon.

    It's a shame that Pentium IIIs, while invariably poor consumer buys, are still doing well because of marketing. Want a cheap computer? Get a Celeron. Want a fast computer? Get an Athlon, which will outrun any PIII of the same price.

    A neighbor of mine got their first computer last Christmas. Maybe my advice got there too late, because their overpriced, unstable Compaq PII system was coupled with a lousy ATI video card and a winmodem. I had the pleasure last month of trying to figure out driver problems with someone's dual PIII, 128MB (or was it 256?) RAM, Voodoo 3 3500, behemoth system... that was having conflicts with his ISA network card.

    Granted, there are a lot of CPU-intensive things (Quake 3 - intense!) out there I'd love to have an 800Mhz Athlon for, but those are the exception (Quake 3 - exceptional!) rather than the rule. (Quake 3 rules!)

    And even in 3D games, we're getting to the point where the processor won't be the limiting factor anymore. I mean, once you're getting 60fps at 1024x768, do you shoot for 90fps? No, you crank up your resolution, detail, or color depth. And once you do that you're worrying about your video card's fill rate, texture memory, or texture RAM bandwidth, not about your CPU. Hell, with T&L going on new video cards, we're going to be at the point where the AGP bus and human perception are the limiting factors on visual quality, not the CPU. Maybe games like Halo will have ultra-impressive physics and AI to use all those CPU cycles, but I'd like to see it.

    Granted, there are always going to be applications (if only Beowulf clusters) where CPU value increases linearly with CPU speed.. but these are niche things compared to the huge consumer market being sold to today.

    What I want to see isn't faster CPUs, it's broadband access! I have a K6II on a 10baseT college dorm connection this year, and if I have to trade it for a quad 1.2Ghz Athlon on a 56K modem when I graduate, it's not going to be a good trade. Whoever thought up Intel's "our CPU makes the internet faster!" campaign should be flogged.
  • You're exactly right.

    Just look at the VCR market (or actually, the game-console market :-) to see where the computer industry is headed.

    Tighter integration, cheaper parts, lower costs, name and model changes every 3-6 months to beat comparison reviews, and warranties, monolithic engineering to prevent do-it-yourselfer repairs and upgrades, to cause the consumer to remain beholden to the manufacturer through eternal forced-upgrade cycles.

    Is it Stereo? Dolby? Double Dolby? DBX? Surround? To me, that sounds just like the same questions asked in the CPU market today. Is it a PII? PIII? PIV? MMX? same garbage, different name stamped on the case.

    I wish I had a nickel for every time someone said "Information wants to be free".
  • What tests have you seen where the Pentium actually wins against the Athlon at the same clock speed? Every article in every mainstream magazine and every review on just about every website i've visited says basically the same thing: "You gotta hand it to them... AMD has outdone themselves this time..."

    Never before had AMD's CPU's actually bested intels efforts at the same speed (Well, the K6 did to the P5, but Intel had already moved on to the P-II and had left the P5 to die)

    Intel really is playing ketchup this time!
  • AMD Will slash Athlon Prices: This is one of the biggest misconceptions about

    The Athlon 600 is retailing for as little as $379 [sharkyextreme.com] now, about half of what they were at four months ago. Granted, it wasn't in one dramatic price slash, but it's a hell of a drop. AMD has kept the Athlons cheaper and faster than same-Mhz PIIIs forever, and I don't see that changing soon. Maybe Intel will get ahead in the Mhz war, but that won't change the market situation for people not buying the absolute fastest chip.
  • The marketting division is serious diluting the brand name "Pentium"

    Let's look:
    "Pentium" -- a decent 5th gen proc.

    "Pentium with MMX" -- Should've been Pentium+, it contained important L1 cache enchancements (P166 w/ MMX outperforms plain P200, etc).

    "Pentium Pro" -- Why not a new name, like P2? This is a 6th generation code.

    "Pentium ][" -- Finally, a proper increase in number. Pretty much the same core, but with cache slower.

    "Pentium ]I[" -- Well, there are two kinds of P3. Ones which have KNI, and no other changes. And the P3 which does have KNI, but which also has faster/less L2 cache, and other important tweaks. They rushed the Pentium 3 "name" out the door to get people buying what is essentially the same chip, which not as much as a performance grab as the Pentium w/MMX over the original Pentium. Not to mention serial numbers (oy vey).

    Now they have the "Pentium IV".. Does anyone else think it's just wrong to have 5 separate chips based on the same cores when, in the same span in the 1980s, Intel was actually inovating and took us through the 80186 (used on controller boards), the 80286, and the 80386 (with the prototype 486 chips just out the door 10 years ago). It seems that rather than reacting with the inovation they had of the 1980s, they are reacting with the Marketting Dodos who brought us "I was born to web!".

    If you'll let me indulge in a little metaphor, Intel is currently roping a nice little noose around its neck, with the word "Marketting" written on it. AMD is currently moving in to kick the chair out from under Intel. Anyone wanna buy some Intel stock?
    ---
  • by jayped (25281) on Tuesday December 14, 1999 @01:34PM (#1465218)
    Besides, the primary (ie most significant to the majority of users) difference between the PPro and the P54 class CPUs was the location of the L2 cache (on the chip vs. on the board).

    Not true. The PPro was a large design change with a heavy focus on 32-bit operations, hence the poor 16-bit performance. Yes it biggest change was integrating the cache into the same die as the core package (leading to a huge & expensive chip).

    Pentium (P54) - Original 586 Core with Good 16/32bit CPU with pipelined FPU, later added MMX

    PPro - New Core (P6) with Great 32bit, poor 16bit, even better FPU, full speed cache, but only 64kb, large memory support. No MMX.

    Pentium II - Evolution of PPro(P6), higher clockspeed, added MMX, moved cache into chips on a board in slot 1, running at 1/2 speed. Upped cache to 512Kb. Same bus as PPro only Slot design to reduce cost. Overclocked Pentiums were sometimes faster than the early 233Mhz PII on 16bit games and apps.

    Celeron - Same as PII Core only originally no cache, leading to poor performance and a bad reputation. Regular Pentiums often outperfomed it, clock for clock.

    Celeron A - New core added 128kb of cache integrated into the core at processor speed, different from PPro(cache was in the die not the core on PPro), different from PII(cache was in Chips at 1/2 processor speed on the slot1 board). Great chip with a low price. Later offered in Sockets for even lower price.

    Pentium III - Updated PII core with SSE instructions added. Later upped Bus speed to 133Mhz.

    PIII Coppermine - New PII core at .18micron allowing higher clock speeds. Integrates 256kb of cache into the core at processor speed (similar to the celeronA but with SSE and more cache). Also introduced FCPGA (flipchip packaging) and Socket that allows enhanced cooling of the core.



  • The portability between the various flavors of linux is source-level: you need to recompile to run on a different arch. Most programs that do things right and use size_t s instead of longs will compile on alpha with no difficulty.

    Microsoft has a lot more difficulty, because: (1) They are committed to binary backwards compatibility, and (2) there are a bunch of types that are specified as being 32 bits.
    --
    Man is most nearly himself when he achieves the seriousness of a child at play.

  • Let's take a look for a second at Intel's chip names...

    8086 -- Intel's first big 16-bit processor

    80286 -- Intel's next 16-bit processor - name is known also as 8x286, shortened to the 286 (easier to call it the 286 rather than the 80286)

    Pentium -- Intel finally feels that a name's better than a number. Many laugh at the name, but soon the term 586 becomes associated only with AMD.

    Pentium Pro -- Again, another attempt to use words to describe power

    Pentium II -- A legitimate upgrade to the Pentium in chip architecture (586 to 686) and in the name.

    Up to that point, Intel's had a good strategy for naming things. But look at the Pentium III! Ther'es no big change in architecture except for a few instruction extensions, but they upgrade it from II to III! Why? Promotion. No other reason. It's really pathetic.

    Pent- means five. It was named for the 586. That was back five years ago, and the name is just too old in technology years. It needs to be changed to something other than the "Pentium IV," but with their last name change from Merced to Itanium, I'm worried what they might come up with!
  • They have to have both product and marketing in order to, in your words,
    own the 64 bit arena

    I'm not convinced that this is likely; Digital has long had some pretty good products, but have lately had an inability to sell their way out of a wet paper bag.

    The recently announced Compaq/Samsung venture to put effort back into Alpha may be helpful, if they actually provide powerful product for decent value. It's not clear that they're likely to out-market Intel, which is a critical issue.

    This connects doubly to AMD:

    • AMD doesn't yet have a 64 bit CPU, and while there has been talk, there is little evidence to indicate what it would really be.

      I suspect their choice needs to be between third-sourcing Alphas or creating an "IA-64 clone."

      The former merely makes them a "me too" vendor; the latter runs the risk of running afoul of Intel patents and/or other "design license" restrictions.

      I am at a loss to decide which outcome is more likely.

    • The connections between Athlon and Alpha seem to be rather more tenuous than "Slashdot Discussions" try to suggest.

      The common thing is the memory access protocol, which implies very little about there being any other interoperability between Athlon and Alpha.

  • i was just in the bookstore yesterday looking through several computer gaming magazines that were discussing building the ultimate gaming system and they all showed tests between amd and intel, and intel was ahead. i have yet to see any tests show otherwise, maybe you can give me some url's? i have nothing against amd or intel.. i'll buy whatever performs the best and priced well.. i don't understand why everyone on slashdot seems to be anti-intel and praises amd. every post about amd being good gets moderated up.. i know intel has invested money in linux (red hat, and suse i think and probably others) so i don't see why people here would be against them. what has amd done for linux? i never see any articles on slashdot about them investing millions for the development of anything in linux...
  • The AMD 200MHz FSB used on the Athlon is a point-to-point bus. The Intel 133MHz FSB used on the Pentium III is a shared bus. They each have their own advantages and disadvantages. Point-to-point busses are inherently faster, while shared busses enable easy SMP. It's easy to say that "Intel is only now getting around to it [a 200MHz FSB]" but in truth, AMD's bus is not as amenable to a quick SMP solution - where are the multi-processor Athlons? If this supposed 200MHz FSB for the Willamette is a 4-stub shared bus it will be quite an achievement since these are quite tricky to design.
  • Yes, but that was more for bus width than bandwidth. This was because memory only came in 8 bit (or 9, but thats irrelevant, the extra bit was for parity) modules and the processors had 16 and 32 bit busses respectively. Pre 72 pin simms, the 486 also needed memory banks. and pre 168 pin simms (dimms?) pentiums also needed banks (64 bit bus).

    However, as a side benefit of requiring banks to get the bus width, it was possible to use slower (thus cheaper) memory and still get similar performance so long as the memory was accessed sequentially. Other tricks like row pre-select were also used to increase the effective bandwidth of the memory. It was clever tricks that kept the 60-70ns memory useful for so long.

  • Sure, but with Roman numerals, you'll never take a divide-by-zero exception. :-)
  • CPU speed needs to catch up with graphics video card 3D accelerator fill rate, which is increasing a la Moore, with CPU falling behind until K7.

    High fill, and future and present hardware transform and lighting which will lead to even higher polygon throughput, will lead to the "ability" to draw greater detail. But so what?

    When the bottleneck of many 3D engines is the per dynamic entity cost?

    When there is not enough CPU power to handle sophisticated physics and AI of a lot more entities now that are capable of being rendered? Do remember the algorithmic complexity rise quadratically.

    If you don't want to see new 3D games make "meaningful" use of extra polygons, except prettier and smoother and "static" geometry, then you don't need the speed to catch up with fill.

    Without faster CPU to catch up, there will be too many technical barriers beyond pretty but dumb 3D gaming.

    Console conglomerates like Sony and Nintendo are already learning their lessons of fast polygon and dumb CPU, and are reverting their trends in their PS2 and Dolphin designs. PC definitely should not fall behind.

    With due respect, if you are happy with the current state of PC gaming technology, or you don't play games on the PC at all, this may not apply to you.

    Corrinne Yu
    3D Engine Programmer
    3D Realms
  • Absoultely! Intel and AMD moveing heaven and earth to beat each other into the ground in the performance war can do little but increase the fps in quake and the RC5 turnover. I've the battle for supremancy last long enough we may even see a processor that can run win9X at a decent rate.

  • If there is a 64 bit AMD CPU, called the K8, why isn't it listed on the AMD website as a product that they sell?

    Reality is that There is no such product. "K8" has not been announced. Do a search at the AMD website and you will not find it.

    AMD has announced something codenamed "Sledgehammer" but that was announced as a FUTURE PRODUCT.

    And as far as the Athlon/Alpha motherboard interoperability, I'll believe it when I see it.

  • Ah but you forget one thing about dual processing, each of the 800mhz chips are sharing the same bus and clock. Four 800mhz chips are still running at 800mhz they just have four times the Int and FP power than just a single 800mhz chip. Forking over more money for a 1ghz processor would be prudent in many aspects because it's likely the 1ghz processor will have much increased Int and FP capabilities than the older 800mhz. When you have to share the same bus you lose a good deal of speed between the processor and the rest of the system. The actual speed difference between a 450 PII and a 500 PIII is only 50mhz but in actual processing power the PIII is almost twice as powerful.
  • If SGI is still relying on Intel processors in 2005 I will cry my little eyes out. LONG LIVE MIPS!
  • While it is more difficlut to make Athalons multi-processor, the bus allows better multiprocessor. You can do it quick and cheap with the PIII, but with the athalon bus you spend more money and end up with more scaleability. The PIII design does not deal well with more then 4 processor in theory, and practice is worse. You get around this by doing what the athalon requires all along. And the Athalon scales to 16 processors easially in theory. (If someone would ever do it)

  • >Hum seem to rember working on a 1 Meg cache PPro system today

    Yeah, there were a few sizes of PPro processors done. Smallest were 256k of L2 built into the chip, then 512k, then 1 meg.. the vast majority were 256k, as the larger cache sizes (obviously) commanded a significant price premium.

    Other than that, yeah, he was correct (and usefully informative, I woulda moderated that one up myself).

    However, I still kinda self-consciously defend my geek points by pointing out that I said the PRIMARY difference that most people saw. I had basically forgotten the whole fuss that went around when Win95 came out about how it was slower on a PPro 200 than a regular P54C running at 200 for many operations due to the lingering 16-bit crud.. his notes reminded me of it. "oops".

    -LjM

What hath Bob wrought?

Working...