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

 


Forgot your password?
Close
typodupeerror
×
Intel

Intel Introduces 1 GHz Chips 169

Posted by Hemos from the the-pissing-match-continues dept.
sheck was one of the first people to write about the release of Intel's 1 Ghz chip. Beating Intel to the punch, we already covered AMD's 1 Ghz Athlon. If you want more coverage check out C|Net. This corporate peeing match about who can release these machines first is pretty funny to watch.
This discussion has been archived. No new comments can be posted.

Intel Introduces 1 GHz Chips More

Intel Introduces 1 GHz Chips

Comments Filter:

  • Nice chip, but .... (Score:1)

    by Anonymous Coward
    Putting cost and evangelism aside, I'd take the 1gig Intel rather than the Athlon. AMD's 2/5 L2 multiplier is really detrimental on performance, at least the P3 has a 1/1 and 128bit wide L2 bus.

    However, you've probably got a better chance of getting hold of a 1gig Athlon, the number of 1gig P3's could probably be counted on your hand. Without question, both have been rushed to market, which isn't a bad thing at all, it's certainly better than performance increases being introduced whenever Intel decided to enlighten us mere mortals.

    Personally, I'd look out for the Athlons based round the Thunderbird core in SocketA packaging, rather than the current Athlons with the L2 hack.
  • While I realize some people just like to say "peeing match", let's reserve that term for petty conflicts along the lines of the RealAudio vs. Microsoft Media Player exchange, where each program tried to take over everything the other program did, a pathetic effort to grab market share that only annoyed the end user. AMD vs. Intel in a battle to provide the fastest chip is what it's all about: a legitimate drive to have the top product, in an arena where the consumer ultimately benefits from the competition.

  • Re: you are wrong (Score:1)

    by Anonymous Coward
    Athlon 1 GHz L2 cache runs at 333 MHz on a 64 bit bus.

    P3 1GHz L2 cache runs at 500 MHz on a 256 bit bus.

    And RDRAM sucks big time by the way, did you read Tom's latest article ?

  • Re: Cars, trucks, and computers... (Score:1)

    by Anonymous Coward
    But it's true that cars have not advanced as far as computers--if they improved at the rate that computers have over the past 15 years, today's car would deliver 1000 hp, get 100 mpg, and cost less than $1000.

    To quote an auto company executive's response to this claim: "Yeah, and they'd burst into flames at random intervals."

    I can't help but suspect that the relative lack of concern for reliability on the desktop has made it easier for hardware to race forward. Less concern for reliability means less time spent testing and refining, and therefore shorter product development cycles.

    I wonder if the rate of improvement of truly high-reliability systems is empirically slower than that of PCs. Any mainframe people care to comment? How does the mainframe of today compare to that of yesteryear, performance-wise?

  • Benchmarks (Score:1)

    by Anonymous Coward
    lots of benchmarks are available here [anandtech.com]

  • Cooks bacon, too! (Score:1)

    by Anonymous Coward
    Just slap a micro-grill on the heatsink and watch that meat fry, baby!

    These x86 houses are all playing catchup to
    the Alphas anyhow. The 667Mhz EV6.7
    is what.. 3 times faster on fp than the
    800Mhz Pentium III?
    Hah.
  • I will be very interested to see what happend when AMD gets enough marketshare that they can leave the x86 instruction set behind. The k8 does this to some extent (new fpu and such) but there is still a great deal of bagage being towed along. The powerpc processors are a great example. They have a much smaller die, fewer transistors and very good performance.

    What I think that AMD needs to be able to do is create a new processor without regard to legacy hardware. They could build a smaller/faster/cheaper chip.

    Binder
  • I have a fealing that we aren't going to see 10GHz processors. At least not for a -long- time.

    I think that IBM's concept of putting two cores on the same die is a great idea. effectively doubling the performance at the same clockspeed.

    We will see multicore cpu's and parallel optimizing compilers before 10GHz.

    Eventually it is going to become too expensive to continue the MHz race. Instead it will be cheaper to make a more efficient processor. All those engineers figuring out how to push the current die faster will be designing a processor that does more per clock and does it with fewer transistors.
    ( the better algorithm vs faster computer debate )

    just my 2c

    Binder
  • Comment removed based on user account deletion
  • the main point is that people keep missing is no matter how fast you build the processors (and related hardware), the computer is never going to go any faster than a human will allow it. a person who has the top of the line everything (mobo, maxed ram, processor, vid card, harddisks, etc) is still limited by his own ineptude at typing. :)

  • Yes, I suppose we *are* totally off-topic. ;-) I have been thinking about this a good bit lately and frankly, I have no sympathy for those car owners caught by surprise by the price increase. Although the federally mandated fleet average for *cars* is 27.5 MPG, many (most?) people have side-stepped this requirement by buying trucks and other vehicles that must only average 21.5 or so, as I recall. Hopefully, this will make an impression on someone, but I rather doubt it. I'll continue to sick with my trusty '88 Chevy Sprint Metro, rated at 55 MPG City, 60 MPG Highway, the highest rating ever for any production car sold in the US (although manufactured by Suzuki) and also the most reliable car I have ever owned, and currently at ~274,000 miles.
  • Now that we are in the New Dawn of Gigahertz processors, the following questions must be answered:

    Which is faster? An Intel gigahertz processor or an AMD gigahertz processor?

    Which is heavier? A pound of Intel chips or a pound of AMD chips?

  • Gah! "...find and dandy..." Strike that! Make it "...fine and dandy..."
  • "Features like extended cab, 3rd/4th door, etc. weren't available in trucks 10 or 20 years ago. "

    Sure they were -- Your average DOT has a fleet of ancient orange four door trucks. Besides, putting a truck bed on the back of a 4-door SUV body is hardly a significant change.

    Trucks have seen improvements, just not as radical as cars have. Compare the brand new 2000 Chevy Suburban with the 1970 model, then compare the brand new 2000 Chevy Impala with the 1970 model -- you'll see the difference.

    (However, note that trucks and SUVs are not covered by the same fuel economy and emission rules as cars - a big reason cars had to change is government regulation, and the popularity of SUVs is to some extent a reaction against this.)
    --
  • Cars are enormously better than they were 15 years ago -- much higher quality, much safer, much higher efficency (although the efficency is usually applied to get horsepower rather than fuel economy). They are also relatively more expensive than they used to be.

    Of course, the American public doesn't like buying cars as much as they used to. Instead, half the market is trucks, which are usually much more primative in design and efficency.
    --
  • A gigahertz is a nice, round number.
    At some point the wires on the chip become just
    a few atoms thick and too choppy to shrink.
    I was hearing numbers around 3-4 GHz for a new
    insulator.
  • OK, but it's 1 000 times faster than my 1 Mhz C=64 !

    Not really. I'd be willing to bet it was more than 1000 times faster than your 1MHz C64. You see, you can't really compare MHz ratings to fairly compare two processers of different architecture in terms of speed. Hz is just the number of clock cycles per second. What really counts is what gets done in each clock cycle. A 500 MHz Alpha will not be the same "speed" as an 500 MHz Pentium III as an 500 MHz G4.

    In fact, it is not even really fair to compare the Athlon to the Pentium III in the MHz game to be able to say one is faster, even though they are both x86, because the Athlon can do much more (on average) per clock cycle than the Pentium III, making it an overall faster processer clock per clock.

    -B
  • ...who gives a fuck? This whole thing about MHz is just propaganda from intel and AMD. Sure, more MHz == more speed - but instead of spending paltry amounts on bigger fans, these companies should fscking stop ripping off consumers with their huge margins, and make some decent technologies.

    MHz != innovation/evolution/better

    pffft...
  • Again as the previous poster mentioned, go to www.gw2k.com, click there little "Introducing the Worlds First 1GHz PC" blurb at the bottom. Damn, only $3k, I wonder how much that HP system is going to be. I am guessing $4k at least with that insanely expensive RDRAM they are using.
  • Intel pretty much HAS to undercut AMD now, the only officially supported platform is the i820/i840 which is Rambus only. That is $800+ for 128 MB of RAM. That lower price Intel is charging for the processor vanishes very quickly when you put the system together.

    An interesting note, if you look at this article [tomshardware.com] at Tom'sHardware you will see that the best platform for the copermine is actually an old BX chipset overclocked to 133 MHz bus.

  • does AMD's 1-ghz chip, while having the same high clockspeed, have the same level of Innovation?
    for example does it Make the Internet More Fun?

    **walks away snickering**
  • <I>What better way to give feedback to OPEC then to tell them "no thanks, we don't need much oil anymore"</I>

    Not very likely. Sure - we'll get hydrogen-powered cars. No carbon dioxide or other pollution. But the hydrogen will mostly come from hydrocarbons, with the carbon removed chemically.
  • <I>Athlon is a superscaler processor that is capable of issuing 9 instructions per cycle, the Pentium 3 according to the intel document is not superscaler, it can only issue 1 instruction per clock. </I>

    Something's wrong here. Even the original pentium had an advantage over the 486: It could issue 2 instructions per clock as long as they didn't conflict. The athlon has a clearly superior floating point unit though.
  • <I>Yeah, I remember reading that my 1 Mhz C=64 ran it's code as fast as a 4.77 Mhz IBM PC... something about a more efficient design in the instruction decoding section of the chip. Where the the Intel chip used a ROM look up table, the 6502/6510 chip used combinatorial logic? Is that right?</I>

    It isn't just MHz. It is also "how many of these fast clock cycles are necessary for doing the job?"

    I had an assembly programming manual for the 6510. Most of the common instructions took 2 or three clock cycles to complete. There were some rare instructions that could take as much as 9 clock cycles, (if they used the seldom used "indirect indexed adressing mode" and also stumbled onto a 256-byte boundary while doing so. That was the very worst case.

    The 8088 were a lot worse off - any instruction requiring memory access had an extra cost of address calculation and could easily use 30 or so clock cycles. So the 6510 won on speed, but the 8088 had nifty stuff like 16-bit registers and could access more than 64kB.

    The c64 was interesting in another way too - main memory was <I>faster</I> than the processor those days. No cache needed! The c64 video logic accessed memory between the processor's memory cycles - without slowing processing at all. If I could get that kind of memory for today's processors...

  • They're canning the K6-2+ in favor of the Athlon Select, a socket-based Athlon core with 128KB on-die L2, for the low-end market. It will compete very well with the Celeron, as you can imagine.
  • Check out the volume of AMD stocks being traded and that'll tell you the big deal -- It's not important to most of us, but to marketing folks, it's a big deal. AMD *had* to beat Intel to 1 Ghz to get all this attention from the mainstream media.
  • Well, that's leaving out the one number-crunching-intensive application that would be really useful on PDA's, etc: voice recognition. Now, the quality of voice recognition isn't too useful at this point, but while it's mostly a novelty on desktops, voice recognition on a PDA would be EXTREMELY useful.

    I don't know anything about the algorithms for voice recognition, but I wouldn't be surprised if some of that could be moved into specialized hardware which could run at a much, much lower clock rate, so you probably won't need 1GHZ anytime soon anyway.
  • Yeah - remember when the 80486 CPU was a server component, and desktops wouldn't be able to take advantage of the CPUs?
  • Hmm, same here... Has the wait for SMP mobo's been so long with all new CPU's?
  • If it's as stable as the new i820 chipset then good luck to intel....

  • ... in real world benchmarks. If you're talking pure CPU measurements, the AMD *can* be *up to* 40% faster, but it's not usually... they're about even with Intel, a little faster. But because of their 1/3 speed level 2 cache, they get bogged down in most real-world tests, falling slightly behind an Intel chip of the same speed.

    Overall, the two chips are just about equivalent... there is no real major advantage that either has over the other, other than price and availability. That is where AMD really hits its stride, because they are far easier to find than Intel chips.

  • The good news isn't the speed, but the price, intel are undercutting AMD by a couple of hundred dollars a chip

    Currentley AMD is undercutting Intel by a couple of hundred

    and with the onboard cache things are comparable.

    Actually as the Athalons got faster the cache divisor got bigger, leaving Athalons with less of a performance gain each time. My guess is the performance between intels and AMD's at the high end is finaly starting to get close. I personaly cant wait for AMD to release the Thunderbird with full speed on die cache.

    Problems with supply will dog them as they are only doing a limited run in the first instance.

    cant agree with you more, by my calculation I'll be able to purchas a 1GHz P3 by the time I'm 40 ;->

  • What are you, four?
    -jpowers
  • I love seeing newer faster processors being released because it gives me the ability to by the old ones cheaper. In about 2 months the windows/zdnet/cnet users are gonna say 'oh my god my processor is _only_ 600 mhz' and I'll be there to say 'here is 300 dollars for your system'. -- the joys of running an efficient operating system.
  • See here [sharkyextreme.com]. From the people who brought us the review posted w/ the Athlon article earlier this week.
  • It has been said before (in different context) clock rate isn't everything. There are other factors. Like the fact that the AMD has 128k of L1. Clock rate on L2's
    doesn't mean shit if it has extra latency associated with it.

    10 minutes running W2k and your 10 favorite games on equivalent Athlon -vs- PIII will tell you the Athlon blows the doors off the PIII for nearly everything except a few rare cases.

    These people who pound the one or two cases where the PIII is faster have obviously forgotten the noise surrounding the 286->386 and the P5(pentium)->P6(PPro) upgrades because there were a number of cases where the newer processor didn't keep up with the old.

    The bottom line is the K7 will be the death of the existing P6 core. Intel knows this and is just attempting to keep their head above water while they work out the kinks in their next core. This time next year the Athlon core will still be around and fighting the fight while the PIII will be a memory just like the P5 is today.
  • Yes, your phone does have a very fast processor. The processor is designed with a very specialized instruction set that is an extreme form of risc called a Parallel Logic Ladder (PLL). These processors are typified by the PLLatinum line [national.com] from National Semiconductor [natsemi.com].

    PLLs have an instruction set that is sparse even compared to conventional risc chips. Most PLLs only have one instruction, integer divide. This makes programming a bit cumbersome at times but modern compilers make the job easier.

    Over the last decade PLLs have worked their way into almost every wireless communications device. Their rapid growth has been attributed to the massive computational demands of the NSA's voice recognition software.

    Ryan
  • > The processor speed is not a frequency.

    Exactly right. When applied to processors (or other computational components) MHz refers to inverse time. This phenomenon is an obvious consequence of Gates' Law and Moore's Law.

    Moore: Speed of hardware *2 every 18 months.
    Gates: Speed of software 1/2s every 18 months.

    As you can see, there is an inverse relationship between hardware and software speed, that is, software = INV(hardware). This is where the term 'inverse time' comes from.

    Ryan KE6FFQ
  • ... and the idiot is you.
  • OK, but it's 1 000 times faster than my 1 Mhz C=64 ! [Snip] wow my god ! 1 000 times faster (not talking in MIPS, of course) than my first computer.

    Yeah, I remember reading that my 1 Mhz C=64 ran it's code as fast as a 4.77 Mhz IBM PC... something about a more efficient design in the instruction decoding section of the chip. Where the the Intel chip used a ROM look up table, the 6502/6510 chip used combinatorial logic? Is that right?
  • Well, a nice side effect of finally releasing 1GHz processors is the next step will be in .1 GHz increments, instead of 33 & 50 MHz increments.

    I can't wait for 1.5GHz chips. By that time all these 600MHz chips will be dirt cheap.

    -Adam

    "Consider the two levers for moving men - interest and fear" - Napoleon
  • AMD stock is also climbing due to them making the cpu for Microsoft's X-box.
  • AnandTech Review of CuMine 1Ghz [anandtech.com]

    This article has "interesting" benchmarks, using the i820 chipset. It shows that the Athlon can beat the crap out of the Intel chip in alot of the cases.
  • C'mon, what red-blooded American male technophile with a keen sense of penis envy wouldn't buy a 1-Ghz CPU? Then they can go to their office, harrumph, hitch up their pants and proclaim to their co-workers in a loud, booming voice that they have the latest, greatest hardware on their desk at home.

    Of course, if you really want to be the big stud on the block, get a dual processor 1-Ghz machine. Then the chicks will just *flock* to your doorstep.

    Moore's Law is getting blown out of proportion. Personally, I think it sucks if my system bus runs at 10-13% of my CPU speed. When someone comes out with a 1-Ghz motherboard, then I'll get excited. I'd rather see these companies working more towards widening bandwidth to memory or some other more beneficial pursuit. Intel's made enough of a mockery of Rambus already.

    ** Programmers that do not comprehend the Tao are always running out of time and space for their programs. Programmers that comprehend the Tao always have enough time and space to accomplish their goals. How could it be otherwise? **
  • Beleive it or not, many OEMs "sell" bleeding edge Intel machines, but they rarely ship them in volume right away. Many of these sales take forever to actually ship. They take long enough, in fact, that customers cancel their orders in droves (ask dell how many millions they lost due to lack of chip supply from intel).

    "Perspective is lost in the spirit of the chase."
  • Sharkey Extreeme [sharkyextreme.com] implied in the AMD 1000 MHz article [sharkyextreme.com] that he would now have less work to do since the speed increments would be more significant. 150MHz was a big deal when compared with a 100MHz chip. 750-800MHz? *Yawn* I don't expect people will even notice the difference between a 1.2Ghz chip and a 1.25GHz. People can now focus upon more subtle things like quality. Just give me a chip that doesn't run HOT...thankyouverymuch

    In 18-22 months we should have 2GHz chips!
  • Okay, I just caught a couple of mistakes in my own message.

    One, everything you said *wasn't* truthful, according the to guy above. (However, my post didn't pertain to technical speficics.)

    Two, I understand now that there are motherboards for the CuMine that use SDRAM that aren't really too shabby.

    Three, I said SIMM when I meant RIMM. And PC100 comes in a DIMM package, usually.
  • While everything you say is truthful, you're missing a few key ideas here.

    If you have a *really* fast processor, it does you no good at all to have a shitty motherboard, IDE hard drive, slow memory, etc. I personally have an Athlon 750 system, with SCSI hard discs, PC100 memory (Irongate 750, blegh), and a GeForce video card. This is one butt-fast system, but I had to pay out the ass for it. But the components do exist.

    Now, look over at Intel. As I understand it, to get top performance out of a CuMine CPU, you need an i820 chipset (buggy), and a stick or two of RAMBUS. RAMBUS, for crying out loud, is like $600 on pricewatch right now for a 128MB SIMM. Compare that to SDRAM, which is around $90 once you add in shipping.

    RDRAM is supposed to be technically superior to SDRAM, which I don't doubt. But it just reeks of conspiracy. I mean look at the prices. Yet Athlon benchmarks continue to edge out P3 ones in all the reviews.

  • But it's true that cars have not advanced as far as computers--if they improved at the rate that computers have over the past 15 years, today's car would deliver 1000 hp, get 100 mpg, and cost less than $1000.

    I'm no conspiracy theorist, but don't yout think that just maybe OPEC/the oil companies had more than a little part in holding back progress & discouraging innovation?

    ...or maybe I'm just bitter that gas prices are 72.9 cents a litre ($CAN) despite the fact that it's produced here in Alberta.

    Price-fixing pure and simple... maybe I should have posted this as an AC I'd hate for anythkdflidysff ad,a;dsolkdladflshd.3

  • Compaq has st ated [compaq.com] that they will begin selling a 1 Gig system before the end of this week as well -- HP will be starting next week. And in v.limited quantities too, I might add.
  • Handhelds tend to be used as personal organizers, and furthermore won't have the room to handle the heat dissipation of the current generation of 1ghz chips, but disregarding that, let's think about where they're going. The tendency right now is in the direction of communication, so let's assume we suddenly get cameras on them that imply a need for compression. Well, that still doesn't require 1ghz, but beyond that, to transmit moving images like that requires again a decent amount of cellular bandwidth, which is going to be harder to come by than even the new Intel chips. So how about number crunching? Generally that requires large sets of data, and that requires miniaturized storage, which isn't there yet either.

    I have talked to numerous people who used slide rules for math, and used three pages front and back to do one math problem. Now I just plug it into a faithful TI-86. Think how in a few years, we could all be doing analysis of major equations. Even now it takes about a minute for a 20 degree probelm (don't ask me how I know). Each generation of chip gets cooler. Think what one of those K6s would feel like if it were majorly OCed. HOT! As these developements come, we get better technology in everything. It is only a matter of time before I have a TI-1000 (not an advertisement) that has a lens on the end and soem way to transmit info out the bottom. It will be able to take pictures, do math, and have say infrared gaming. Now that is what I call study hall.

  • A 300 watt power supply should be more than enough.
  • What are these going to cost on the street, huh? $1000? What's the point? If you want realpower and you don't care about price, why don't you go out and buy a Cray or something? This is rediculous. I think that Intel just decided to announce when they got five or so coppermine 800mhz's that rolled off and could do 1ghz - not because the chip is actually there.

    "The romance of Silicon Valley was about money - excuse me, about changing the world, one million dollars at a time."
  • The whole 1 Ghz thing is "only" pure marketing, of course. It's not a lot faster than 900 Mhz, OK, but it's 1 000 times faster than my 1 Mhz C=64 !

    In my mind (call it perverted with the metric & decimal system if you want), this looks like a milestone : wow my god ! 1 000 times faster (not talking in MIPS, of course) than my first computer.

    Psychologically, it's a bigger step than the adoption of a 64 bits CPU. There's nothing rational here.

    Stéphane

  • perhaps someone can fill me in on this, since the link wasn't too technial.

    when the athlon was released, everyone complained about the 1/3 core speed of the L2 cache. how does the new intel chip compare in terms of the L2 speed?

  • would that by any chance be mr. mac of P.C.C.C.?

    q


  • If you can run Q3A on the Alpha platform, then I'm wrong right in the category where I was making my case, lol.

    Way to go :)

    I was also educated via email that Alphas can use some PC addon cards. Very interesting!

    As for the implied Alpha vs Athlon thing...isn't the Athlon the brainchild of some Alpha engineers?
    ========================
    63,000 bugs in the code, 63,000 bugs,
    ya get 1 whacked with a service pack,
  • Geez guy, you need to lighten up!

    How can you read these comments and have to be all nit-picky and such? Its just a little fun!

    Thats why it was rated as funny and not something else, because non-nitpickers find things like this humorous.

  • Hmmm... Question is: Can they deliver? (Score:2)

    by Anonymous Coward
    Seeing as how Intel has been having MAJOR problems ramping up their Coppermine chips recently, even on their .18 micron process, I would really have to wonder if Intel will actually release their 1GHz chip within the next 3 months.
    Note that it took 2 months from the time that Intel announced their 800MHz coppermine and when it actually appeared on the market. And even still, its in very limited quantities.
    This is just a "Me Too" action by them, I am willing to bet a lot on it.
    That would be wierd if I got a first post ;)

  • Go ahead! (Score:2)

    by Anonymous Coward
    Throw your money away!

    Well, you will unless you buy the chips in the 600-800 MHz range...

    Those are sooooo January, though..

  • Sorry for replying to myself but this is the URL for ordering your 1GHz Athelon with 30GB drive and 128MB RAM for $3,000.



    http://www.gw2k.com/prod/hm_sel_Matrix .shtml [gw2k.com]

  • The funky thing about having stock holders is that you owe them a load of explanations for your actions. AMD's 1 GHz costs US$1299 (1000 units) vs Intel's 1 GHz at US$990. I'm sorry, but the demand for such beasts is high and Intel has supply problems, their price must go up.

    I really hope Intel won't be able to keep up with this.

  • Dateline, March 3, 2000 [slashdot.org], Hemos:

    "I do have to say that the corporate peeing match between these two is pretty amusing..."

    Dateline, March 8, 2000 [slashdot.org], Hemos:

    "This corporate peeing match is pretty funny to watch about who can release these machines first."

    Some kind of fetish, there, Hemos? :)

    ---

  • The deal is speed == big_shiny_object so it sells.

    However, I agree with you. Now that we've hit the 1000 MHz level, maybe the industry can concentrate less on clock speed and move onto cool architectural enhancements. The next big step is not 2 GHz, but 10 GHz, IMO, since "factor of" jumps are more impressive than "multiple of" jumps.

    Maybe they could come up with a line of chips with interesting features besides clock speed, keeping clock-speed oriented stuff the consumer line. However, that could fragment things, but it would still be kinda fun. I'd love to play around with something like that.

    Woz
  • At the moment, Intel's price is $990 each and AMD's is $1299; it's pretty much unprecedented to see AMD charging a premium over Intel, especially for what is not as good a chip except for FP work.

    Neither of those prices bear any relation to the cost of manufacturing the chips; I have a feeling that the yield at these speeds (at least for Intel) is not high, and that these might even be loss-leaders ("I've got a GHz chip. OK, we get three working chips out of each $5000 wafer, but think of the press release").

    I expect dual 1GHz Athlons to be in the same position this Christmas that dual Celerons are now, and by this time next year I hope to be using a dual 1500MHz Willamette box.
  • This is all find and dandy, but it's just incremental improvements in existing stuff. What I get the hots for are the quantum leaps: fundamental changes in technology that give you huge leaps in performance, reliability, portability, usability, scalability... You get the picture.

    We need both. We need the Intels and AMDs shaving off a few nanoseconds here and there by upping the clock speeds and improving the caching, etc. But we also need someone in the skunk works somewhere trying for the "Now for something completely different" stuff.
  • > HP is already planning to sell systems within a week.

    For my money, a chip is on the market when I can go to pricewatch.com [pricewatch.com], find some prices, call the seller, and find someone who actually has it in stock.

    Special arrangements with OEMs are exactly that: special arrangements.

    p.s. - By the above definition, the fastest x86 chip "on the market" today is the Athlon 850, and it is about three times as available as the PIII 800, if you measure availability in terms of number of sellers. It's also cheaper.

    It will be interesting to watch and see when the G's show up, but right now I don't think there's any possible spin that is going to let Intel come out on top on this one.


    --
  • My cordless telephone says it's "2 gigahertz". Since 2 is greater than 1, that means my phone is faster than those new chips! Top that Intel and AMD!

  • And the frequency at which a processor operates has about as much to do with its speed as a handsets signal frequency does with its cpu frequency. Which, I believe, was the point.
  • Many standard PCI cards will run in several different Alpha motherboards - both under Linux and NT (maybe Digital Unix, too, who knows). A Matrox G200 in a 533MHz Alpha 21164 was pretty decent at quake (before my board died and Compaq refused to even *sell* me a new one, much less replace it under warantee)...
  • One of the two Ghz chips will probably suffer from "Better 75% right and in the market than 100% right and too late." We'll see who blows up first. (can of Splode anyone?)

    Now if we could just get those G4s over the .5Ghz hurdle... (POINT five!!!)

    You're all obliged to scrub the zeros off your processor speeds and add a decimal in front.

    joe maller
  • Anandtech [anandtech.com] has a really good review of the Intel (and AMD) chips. The difference in performance between the two is pretty interesting. While the 1/3 speed cache on the AMD hurts it in some benchmarks, it still whoops up on Intel in others. Very good piece on the technical and business aspects of the speed race.
  • I'm no conspiracy theorist, but don't yout think that just maybe OPEC/the oil companies had more than a little part in holding back progress & discouraging innovation?

    ...or maybe I'm just bitter that gas prices are 72.9 cents a litre ($CAN) despite the fact that it's produced here in Alberta.

    To be honest, I'm of split mind about the recent huge increase in gas prices. It's horrible having to spend to much to fill up the car, but perhaps if OPEC continues to use it's monopoly power to driver up prices, people will start to consider alternative fuel sources more seriously, instead of just as a novelty.

    What better way to give feedback to OPEC then to tell them "no thanks, we don't need much oil anymore" and watch their fortunes (and maybe the political instability of the area) dry up due to greed.
    ---
  • Technically, all Coppermine procs faster than the 600 are overclocked because there is no real difference in the cores between a 600 MHz coppermine and a 1GHz coppermine. There are some tweeks, but mainly the only difference is that the 1GHz procs are lower yield/higher quality parts. All procs are made at the same. If a particular core can't hit 1GHz, its sold as something lower. If there is a really high yield, and no cores are failing the tests, then even cores that pass at 1GHz will be sold as something lower. Overclocking is only when a chip is run faster than the manufacturer sold it to run. (not necessarily what speed it passed at. Many 600MHz procs actually have cores that passed at a higher speed, since Intel is getting really good yields on the coppermine chips.)
  • Okay, I noticed there's a lot of misinformation/total crap being thrown out on /. right now...so I'll clear a few things up.

    Correction: Clock for clock, according to SharkyExtreme's [sharkyextreme.com] and AnandTech's [anandtech.com] benchmarks, the Pentium III takes a majority decision against the Athlon while using the i820/RDRAM and KX133/SDRAM chipsets (with the notable exception of professional CAD/CAM), which is useful for the money-is-no-object department. Interestingly, Anand also benched the P3 with a Apollo 133A/SDRAM chipset revealing a give-and-take tie relative to the Athlon, for those of us that are a bit more price conscious.

    Correction: The P3 L2 cache is 8-way associative, 256 bit wide, 256KB in size, and runs at full clockspeed. The Athlon L2 cache is 512KB in size, running at 1/3 the clockspeed. The Athlon also has a 128KB L1 cache compared to the P3's 32KB L1 cache, both running at full clockspeed.

    Correction: There is NO yield problem at Intel. There is, however, a supply problem, due to management mispredicting what quantity in chips they need to have supplied, as well as reallocation of resources as Intel prepares its fabs for Willamette and Itanium. Gotta love management. For proof, check out the amazing ability of Intel's 500E-600E chips to overclock to 700+ MHz. That's not a characteristic of a chipmaker with yield problems.

    Correction: Why on Earth are people deciding what processor is superior by the supply of said chips? Like most sane people, I happen to judge performance on the basis of performance alone. Or maybe it's because I'm not a brand-name zealot. Either way...unless you're talking price/performance (in which case why even talk about GHz processors?) please can the supply arguments.

    So who wins? The consumer does. Hopefully with the introduction of Cyrix's Joshua processors, the chipmakers will be squeezed even harder to cut both profits and prices. If you really desire a God Box, go take out a student loan and treat yourself to an SMP Alpha platform.
  • While I can give you no great arguments that you're wrong, (other than the fact that there are already --I think -- 500 mhz crusoe chips -- you think it'll take more than five years to hit one ghz?) when has industry-wide development on component of a computer ever slowed down because they had produced the hardware ubercomponent? Who really needs a 1 ghz chip? I would guess that less than a quarter of the computer market needs really fast chips (as opposed to really fast hard drives, or whatever)? Everyone knows the answer : chips are pretty, and so they'll keep making faster ones.

    I think the problem with your argument is that you're trying to figure out why one ghz processors would be needed in handhelds. I say that they'll happen regardless of whether or not they're needed. One ghz chips are sexy.

    Plus, who knows what cockamaimy scheme somebody'll dream up next year that'll start chewing up mad mobile cpu time? Just because you can't envision it (and neither can I, because if either of us could we could potentially grow rather wealthy -- please let me know if you think of anything) doesn't mean it won't happen.

    Besides, the ambitiosity of software is usually directly related to what kind of hardware is out there. People just keep making bigger and bigger software.

  • ...saying that its 1 GHz processor is 15% faster than its rivals (AMD?) when the Athlons are approximately 40% faster than the P3s of the same clockrate? Are they referring to the speed of the internet on the P3 1G vs the Athlon 1G? :)

    But seriously. Maybe they consider the Athlon 700 to be their competitor? (1.4 * 700 = 980)

    Ah well. I personally can't wait for the SMP DDR Mobos to start falling out from AMD so I can run dual 700s. Awwwiyeah.
    --
    "A mind is a horrible thing to waste. But a mime...
    It feels wonderful wasting those fsckers."
  • What does the #1 chipmaker title entail? I noticed the article referred to Intel as such... is it because Intel sells more chips, is more popular in the minds of empty-brained managerial types? Its obviously no longer because they lead the industry in producing the highest performing chips and hasn't been for several months. I'm just wondering what AMD needs to do in order to wrest this title from Intel...

    Or is it a totally unofficial definition-less title the author bestows on Intel in an attempt to give some deep loving tongue to their ass?

    Esperandi
  • Sorry to break this to ya, bub, but if you actually read the article about Intel, you'd see that HP is already planning to sell systems within a week. Have you seen any packaged systems with an Athlon yet?

    Brad Johnson
    --We are the Music Makers, and we
    are the Dreamers of Dreams
  • This is somewhat ironic. The first week in December, I ended 15 years of employment at Intel Corp. One of the major frustrations over the years was that their entire I.T. organization was set up to support someone who runs MS Office and that's it. It was very difficult for engineers (particually software engineers.) to get decent machines. So it seems funny that they will be giving employees new machines for home but can't seem to give their engineers decent systems for their desk at work.
  • Please...I can't take it anymore.

    "1 GHz" this and "700 MHz" that...I'm still paying for my PII-350, you insensitive fucks!@#$

    Reminds me of when I paid $480 for a 540M hard drive years ago. Excuse me, I have to go sit in the corner and cry now. Make the bad people stop. :P

    -Legion

  • vaporware alert:

    Your best bet between Intel and AMD for a 64bit cpu is AMD's implementation, aka the Sledgehammer architecture, which is an extension of the current IA32 instruction set to 64 bits. Without sacrificing 64bit quality, the 32 bit apps will run far quicker on a Sledgehammer architecture than on Intel's 'hard core' Itanium.

    Then again if you're using an Alpha, I guess you're not even using x86 based software. Alphas make great server machines, but I wouldn't count on any leisure applications coming out for it :)
    ========================
    63,000 bugs in the code, 63,000 bugs,
    ya get 1 whacked with a service pack,
  • Intel has been having problems ramping the CuMine because of the on-die cache, something that AMD has been having great troubles doing as well. They cannot get good yields of their K6-3 above 450mhz because of that very problem.

    The on die cache makes a huge difference, look at the benchmarks, the 1Ghz Athlon with off die cache is not that much faster than the 800mhz CuMine which has it's cache on die.

    With current processors being 10 times faster than their memory, the cache has become increasingly important... AMD must realize this but has decided to create a even bigger problem by rushing these 'faster' chips to marked by using a devious trick, they have changed the divider on the off-die cache to much lower than the ½ on the original Athlons thus making the speed gains of all their chips that run at higher than 700mhz lower than their clock speed indicates.

    Intel on the other hand has solved the problem with on die cache, which has cost them dearly in the short term but they have been rapidly improving yields. They are also proving that on die cache is the only real solution in the long term by nearly equaling a chip that is 200mhz faster than the 800mhz CuMine even though the architecture of the CuMine is greatly inferior to the Athlon.

    Lately, I have noticed that the supply problems are starting to subside and the CuMine at all speeds but 800mhz are readily available. Nevertheless, I have serious doubts that Intel can provide a steady supply of 1Ghz CuMine chips without some process tweaking and a few of their infamous microcode and stepping changes.

    Of course, all indications show AMD is having great success with their Thunderbird chips that have on die cache. When the Thunderbird arrives, in all probability, they will give the CuMine the same spanking the original Athlon gave the original P3.

    Of course this is a "Mine is bigger than yours" kind of thing but it is giving us faster chips at lower prices, that is all that really matters to us mere mortals. ;-)

    -----------------------------------
    Jeff Coulter
    Geek in the clouds
    Virtuoso - Smart Personal Agent
    jeffcoulter@users.sourceforge.net
    ICQ: 33011156
    -----------------------------------
    "He who will not reason is a bigot; he who cannot is a fool; he who dares not is a slave."
    - Sir William Drummond
  • The MHz race is boiling down to nothing more than a marketing warefare. The overall impact from speed this or speed that printed on the box makes a huge difference in sales, especially with the big manufactures who they have had partnerships for years. Technically, for Intel this is really not all that big a deal, just another step in the longer term strategy of Intel. They have to announce "Break through this or break through that about once a quater to keep shareholders happy. Used to be about once every six months or so, but with AMD biting at their heels with the Athalons selling like hotcakes, they need to keep up. The down side is that addind x MHz every quarter or so takes time and resources away from development of the next architecture. Newer architecture on processors will make a significant impact on my buying decisions, not 50 MHz a quarter, It just doesn't give me enough to make it worthwhile.
  • and someone will still overclock it

  • I think two drunk guys with a ruler and their pants around their ankles is a better description.

    "Peeing Match" is just easier to say.
    -

  • I'll bet it runs real hot (Score:3)

    by Wansu ( 846 ) on Wednesday March 08, 2000 @07:42AM (#1217591)
    It's going to need a wombat heatsink and a 3&1/2 inch fan. I wonder if the power supplies out on the market can adequately power a system with one of these, a DVD, a ZIP, a bleeding edge video card, etc.

  • Whoopee (yawn) (Score:3)

    by RayChuang ( 10181 ) on Wednesday March 08, 2000 @07:57AM (#1217592)
    So finally Intel has gotten the Pentium IIIE CPU to run at 1,000 MHz.

    There are a few problems, though. First, FINDING a 1,000 MHz PIIIE CPU is going to be just about impossible. Secondly, when it comes to pure FPU performance, the Athlon 1,000 MHz is still better because the PIIIE is still heavily based on the original P6 core from the Pentium Pro some five years ago!

    Now that motherboards that use the VIA Apollo KX133 chipset is now becoming available, there's no incentive to use the PIIIE instead. In fact, if you have a graphics card that uses the nVidia GeForce 256 chipset and also run the latest Detonator 3.76 driver, the Athlon in many tests will run rings around the PIIIE 1,000 MHz.

    I think the Athlon's advantage will increase even more when the second-generation Athlon (code named Thunderbird) with its CPU-speed L2 cache becomes available in a few months. I think a 1,000 MHz 2nd gen Athlon may perform as much as 20 to 25 percent faster than a PIIIE 1,000 MHz, mostly because the 2/5 L2 cache speed restriction will be gone.

  • Processors passing memory price (Score:3)

    by CAPSLOCK2000 ( 27149 ) on Wednesday March 08, 2000 @06:46AM (#1217593) Homepage
    For a long time processors have been the most important part of a computer. At school I was taught that some day in the future RAM would become the most expensive part, not processors. When this occured a major change in computing would happen, using more processors instead of more RAM. I think this will happen very soon.
    Prices of processors are dropping so fast because of this speed race. RAM OTOH stays expensive. We are allready seeing a steady increase in dual processor boxes.
    Does anyone have any ideas about how this change from lots of memory to lots of processors will look like.

  • Why Intel is priced lower (Score:3)

    by SpinyNorman ( 33776 ) on Wednesday March 08, 2000 @07:42AM (#1217594)
    Dell 1GHz PIII: $5999

    Gateway 1GHz Athlon: $3199

    Those prices are for otherwise identical systems: 30G HD, GeForce, 19" monitor, 256K RAM (gateway base price is $2999 with 128K, so I added the $200 that their configurator adds for a 256K config.).

    The Dell/Intel system is almost **DOUBLE** !!!! the price of the Gateway/AMD system!!!

    The Rambus memory used by the PIII is of course the reason, and is why Intel is forced to price the CPU itself under AMD. If you check my history I predicted this yesterday, and stand by my predition that AMD will not drop their price in response - they have no need to!

  • Real release or Symbolic release (Score:3)

    by Forty-two ( 40787 ) on Wednesday March 08, 2000 @06:58AM (#1217595)
    The last I heard when AMD and Intel were demoing their newest Ghz chips is that AMD was going to beat intel to the punch again and have a Athlon out much before Intel had their PIII. Sure enough:

    Intel will release Pentium IIIs running at 1 GHz or faster by the second half of the year as well as the next-generation Willamette chips running at the same speed, Yu said. Quote here [slashdot.org]

    Now isn't the second half of the year starting around July, August? Intel hasn't been doing well meeting its deadlines much less breaking them by months. I seem to remeber something simmiliar happening when both AMD and Intel were comming out with 600Mhz processors, AMD demoed theirs and then Intel came out a close second with their chip that, while it did run at 600Mhz did it using a little more voltage the usual and didn't seem to be as stable and their regular batch of PIIIs.

    Remembering my comments about failures of PIII 600 CPUs, actually also reported by several other publications in Germany and the UK, should give you an idea how hard it was to run all the benchmarks with an even overclocked PIII 650. Quote here [tomshardware.com]

    It will be interesting to see tests on these 2 new processors to see how good they actually are, but this just seems to be a release by Intel to show that they arn't lagging AMD even though they really could be if AMD can produce good 1Ghz chips in mass when Intel is suck with declaring that there ARE 1Ghz PIIIs but if you actually wanted to find one it would be as easy as finding a Athlon and motherboard when they were`released'.

  • It's great... (Score:3)

    by Saige ( 53303 ) <evil...angela@@@gmail...com> on Wednesday March 08, 2000 @06:23AM (#1217596) Journal
    I hope Intel and AMD continue their little battle - I'd love to see more companies get involved. After all, we're the ones that benefit from it from faster chips and cheaper prices.

    Too bad car companies don't put as much effort into improving over each other instead of just advertising better - we'd be driving much safer and fuel efficient things...
    ---

  • Re:Processors passing memory price (Score:3)

    by LinuxParanoid ( 64467 ) on Wednesday March 08, 2000 @12:40PM (#1217597) Homepage Journal
    This is fundamentally false. Processor prices and memory prices drop at essentially the same rate since they are based on similar manufacturing technology. There are bubbles over time in pricing if Intel gets lazy or agressive or if the RAM manufacturers don't build enough factories or conspire to keep prices high, but overall, for the forseeable future both microprocessors and RAM will follow Moore's Law.

    It is true however that CPU performance improves much faster than *memory latency*. For this reason, for about the last decade, academics have speculated and examined the possibility that (simplistically speaking) CPUs might be built around RAM, rather than RAM built around CPUs.

    To some extent, this thinking and today's reality match the scenario you outline; on-chip and off-chip RAM caches are taking up a steadily increasing percentage of chip real estate; for some chips, 3/4ths of the processor is transistors and paths for the cache memory. For economic reasons however, it will continue to make sense for quite some time to have full system memory implemented separately from the CPUs.

    --LP

  • official prediction to be laughed at in ten years (Score:3)

    by karb ( 66692 ) on Wednesday March 08, 2000 @06:42AM (#1217598)
    The one-gig milestone may be more symbolic than functional. Despite the ongoing race between AMD and Intel, many analysts say the future of home computing is in limited-function Internet appliances and handheld computers that simply do not require the processing might of these new 1-GHz processors.

    Ha!

    Reasons for laughing:

    • In a little while, the internet appliances and handhelds *will* have 1 ghz chips (I don't know exactly when, but it won't be long.)
    • PC's will go the way of dinosaurs like the mainframe and the minicomputer. Oh, but wait, they haven't gone away, have they?

  • Re:whats the limit? 5 GHz? (Score:3)

    by daVinci1980 ( 73174 ) on Wednesday March 08, 2000 @07:27AM (#1217599) Homepage
    There was a /. article a few months ago about this. IBM has the capability now to produce circuits that can handle 90 Ghz. Here's the link...
    slashdot.org/articles/99/12/06/ 0823227.shtml [slashdot.org]


    --
    "A mind is a horrible thing to waste. But a mime...
    It feels wonderful wasting those fsckers."

  • Grrrr - yet more marketing BS (Score:3)

    by el_guapo ( 123495 ) on Wednesday March 08, 2000 @06:53AM (#1217600) Homepage
    Sheesh - go look at this [spec.org] and see that Mhz nowdays (has it ever?) is irrelevant. A 733 Intel CPU gets a rating of 336 while an Alpha 667 get 413!!! And I forgot where I saw it, but Intel says the 1Ghz rates a 410 on this. STILL lower than the 667 Alpha. (on par, but lower)Total marketing BS. I'll admit the speed wars are great, but let's remember what Intel ISN't telling you - that they HAVE to run their chips faster because they're inherently slower designs (sorry for the rant)

  • Good for me (Score:3)

    by Yaruar ( 125933 ) on Wednesday March 08, 2000 @06:37AM (#1217601)
    The good news isn't the speed, but the price, intel are undercutting AMD by a couple of hundred dollars a chip, and with the onboard cache things are comparable. Problems with supply will dog them as they are only doing a limited run in the first instance.

    Although what it should mean is the drastic lowering of Athlon 700 chips, which I'm thinking would make a good system for me. All hail competition. Price wars are good, price wars are our friend.

  • Sorry, little too late. (Score:4)

    by hyoo ( 155460 ) on Wednesday March 08, 2000 @08:39AM (#1217602)
    The PC world finally catches up with the Macs.

    Check out this Mac Plus (circa 1980s) running at 1 GHz.

    http://socrates.berkeley.edu/~schrier/plus.html [berkeley.edu]

  • Subtle Vaporware (Score:5)

    by Forge ( 2456 ) <kevinforge&gmail,com> on Wednesday March 08, 2000 @06:39AM (#1217603) Homepage Journal
    AMD's idea of "introducing a 1GHz chip" is that you can go to gw2k.com , order a PC with that chip and expect it to be delivered in a few days or weeks.

    iNTEL's idea is that there are a few sample chips for manufactures to practice tweaking motherboards.

    AMD has sent chips to the larger retail stores and they should be on the shelf at Comp USSR soon ( if not already ).

    iNTEL will be selling Gigahertz chips retail in a matter of months at best.

    These people define release in vastly different ways and it will take your typical PC user a few more years to work out the difference. As for me personally, I am just happy that this will hammer the prices of the Celeron or K6-2 I can actually afford farther into the cheap range.

  • Interesting development overall (Score:5)

    by Patton ( 70344 ) on Wednesday March 08, 2000 @06:28AM (#1217604)
    What makes this siutation interesting in two reguards is this:

    First AMD now has corperate attentions. It produced the 1GHZ chip first. That demonstrates that it is a very serious player. My boss didn't know what AMD was a few weeks ago. He does now.

    Second Intel is now having to dance to the beat of someone elses drum. How long has it been since they've had to do that?

    I'm hoping AMD can keep this up. If they can I could be able to convince upper checksigners to start letting me put in AMD powered servers and such very soon. Trick is they have to keep delivering.

Slashdot Top Deals

Been Transferred Lately?

Close