Please create an account to participate in the Slashdot moderation system


Forgot your password?

Pentium III hits 1Ghz 190

Frac writes "somebody has finally overclocked a Pentium III to over 1 gigahertz. Here's the translated version of the original French article. "
This discussion has been archived. No new comments can be posted.

Pentium III hits 1Ghz

Comments Filter:
  • The big deal about 1GHz+ processors is now I can use the microwave radiation from my cpu to make popcorn, cook Hotpockets, etc...
  • by Raul Acevedo ( 15878 ) < minus city> on Monday November 01, 1999 @07:27PM (#1570898) Homepage
    Are you kidding? Gigahertz speeds are essential to seeing Netscape crash faster than ever.
  • The fastest G4 you can get currently is a 450Mhz, and I have a hard time believing that the g4 is does twice as much per cycle then an Athlon Especially since it's a 'RISC'. The original Idea behind RISC is to do less per instruction, and run many more cycles per second. (The benchmarks apple touts on TV comparing a g4 to a pIII are done with a faster clocked g4).

    For a long time, Apple had Intel beat in MHz game, but they don't anymore. I'm not saying that an 900Mhz Athlon is necessarily faster then a 450Mhz G4, but I'll believe it until someone can point me to some real, hard numbers that prove otherwise.
    "Subtle mind control? Why do all these HTML buttons say 'Submit' ?"
  • Bear in mind DPCE was based on Thinking Machines' C* language. The Thinking Machines boxes could have up to 65536 CPUs, and so it's reasonable to expect C* could generate code to make use of that. What's nice is that the programmer still writes in nice, linear code. The explicit loops they had to write before to do, say, a matrix multiplication, now become simple math operations. The compiler can then parallelize as it sees fit. The programmer actually sees nice linear code which is actually simpler than the looping code which he would've written before.

    The people who want major parallelism (and can actually benefit from it) are numeric processing goons running weather simulations and other large number processing that is the traditional province of supercomputers. The other space where it is useful is in signal processing and graphics. The desktop crowd isn't going to care much, unless the parallelism gives them an extra frame per second in Quake III. :-)

    As for knowing multiple paradigms: Yes, it's useful, just like learning foreign languages is beneficial, even if it only serves to help you understand your native tongue better. The problem is that not all programming paradigms are immediately useful to the task at hand, so many of us just don't have the time or inclination to get around to it. (I fall into this category.)

  • What exactly is the big deal about speeds in the GHz? May be a mile stone, but it's only logical progression.

    Well I think the big deal at least with this test is that it actually scaled very well, I personally found it impressive the amount of performance gain that it actually acheived. It basically means 1) Intel probably can do Ghz soon w/o much problem 2) the Coppermine architecure is pretty solid 3) We don't know if AMD can scale as well, which means Intel might catch up sooner then AMD would like
  • What exactly is the big deal about speeds in the GHz? May be a mile stone, but it's only logical progression.

    It's the same big deal as with 10MHz (those 12MHz AT clones were "way cool" :-), 30MHz, 40MHz, 100MHz, 200MHz, 300MHz, 400MHz, 500MHz and 700MHz. 2GHz will also be a big deal :-)
  • What I'd like to see is AMD's Athlon technology (micro-ops, along with all its weird Funky massive parallelism and the like) applied to a decent architecture such as PowerPC. Sure, the clock speed would only be a 'measly' 700MHz, but it would be able to do so much *more* at that frequency...
    "'Is not a quine' is not a quine" is a quine.
  • I thought I heard about this when thew first PIIIs came out.
  • by Frac ( 27516 ) on Monday November 01, 1999 @07:44PM (#1570916)
    Note that they also make absolutely no mention of any method of cooling this thing.

    Did you read the article?

    They used the ASETEK Vapochill [] system to cool down the coppermine.

    Now, since the multipliers are locked on Coppermine processors, there's no way that these people could get a 7.5 multipler. The highest is a "7" with a front-side bus of 100 mhz

    Intel engineering samples are not multiplier-locked. I guess we can throw your "simple facts" table from Tom's right out the window.

    it'd be wise to check the facts first.

    I agree. Do you plan on following your own advice though?

  • People are questioning the validity of this, and some funny quotes. 1. Don't forget that this is translated... 2. Intel Engineering samples are not multiplier locked. They could've put that thing at 1.066 GHz, too. Wouldn't matter... 3. I still think that those copper Athlons (not coppermine, it's still made of aluminum) are going to rock. A 1.2 GHz Athlon? Oh, yeah...
  • and hence you try to use SCSI with the more demanding systems! try to reduce the CPU usage!
    I just find it funny that ppl. will have a PIII with a old IDE drive running at 5400Rpm and having a nice 128K cache!
    Ohh but i forgot, its the CPU that matters for everything!
    I would hope that the HIGH end system from Gateway anmd others would have started using SCSI by now... but hey, its just the clock speed that matters! And lets not even talk about RAM, how in the earth can you justify having a K7 with 64 Megs of RAM!
  • Someday Intel chips will get so hot and so powerful that only the five richest kings of Europe will be able to cool them.
  • What exactly is the big deal about speeds in the GHz? May be a mile stone, but it's only logical progression.
  • > What exactly is the big deal about speeds in the GHz? May be a mile stone, but it's only logical progression.

    So is Y2K. Look out, Moore's law says that your processor will revert to DC in about 18 months.

  • ...of 1 GHz on a Win98 machine? What's the reason you'd want to speed up a single-user system like that? Surely those extra Half-Life frames aren't going to make that much of a difference? (And if you have a chip that fast to begin with, is your processor really a framerate bottleneck?) Maybe I unfairly associate overclocking with mysterious gcc crashes, but I just don't see why it's worth the instability.
  • > What exactly is the big deal about speeds in the GHz? May be a mile stone, but it's only logical progression.

    Because its a baby step to TeraHz computing.

    Us graphics programmers can ALWAYS use the extra horsepower. :-)

    Imagine 4096x 4096 x 32 bit @ 120 Hz
    (Which on a 17" monitor (16" viewable) is close to 300dpi! Hey, I can dream, right? ;-)

    (yes I KNOW a gpu doesn't need to run as fast a cpu)

  • This is the kind of talk that reduces the credibility of LINUX.!
    being the flagship platform for the greatest OS every written, namely Linux!

    What are you talking about! i am a Linux fan, but realize that Linux is not even close to coming near some OS's like IRIX and Solaris!

    SlOW SPEEDS LIKE 400MHZ!! I am guessing you have learnt your computers from the corner store of the latest version of computer gamer! I would recommend reading a book on computer architecture! Learn about the difference between clock speeds and how much use is made out of them! Heck man intel has had a rough time getting a 64 Platform!! I wonder why ?

    SGI, Sun and IBM (they still make some monster machines) have some of the fattest machines! I mean for them to have a few Gigs of Memory and butt load of processors is common!

    Finally....... but pathetic poor performing unscalable machines form sun and sgi
    What are you talking about.... adding a new graphic card or sound card is not what constitutes scalability! Look at some of the SGI;s, hell you can just drop in a few more processors when you feel like it! And where can i find a motherboard that will allow 64 processors ? IBM, SUN, HP and SGI all have em! Hmmm but they are not used for games!

    sun/sgi/hp will be out of business within a year, they just cannot compete with such superior technology.

    I will not even comment about this! i mean you just have to be out of it to make such a statement! please if you dont know what you are talking about dont say anything..... its a shame that you have the .sig that you do!

  • This isn't exactly hard numbers, and I think they're referring to older PIII's (not the Coppermine) []

  • are you REALLY that ignorant?
  • yep.. exactly. I always wonder why laptops never went scsi.. cheap scsi. The ide-to-scsi convertion chips people seem intent on claiming all scsi is, and is thus slower.. except for the extremely pricey drives. Not much slower, and with the scsi bus, no extra weight or really lack of power.. laptops should be scsi!

    My old P5-200 SCSI desktop usually outperformed friends P2s on anything not cpu driven, as windows is disk driven.... they used to claim firewire would bring a better standard than scsi at a cost for home and corperate users.. killing the pesky ide. Would hae been nice.. since all of the IDE plusses (ease of installation) has always been nicer on scsi (just stick on a unique ID.. plug in.. make sure your card is detected and usable..)
  • (a) Yes, if you have a special-purpose task, especially if you can fit it in the cache on the CPU, sure you'll benefit.

    (b) If you have a special-purpose task, why are you using general-purpose hardware?

    It seems to me that for something like this, some extreme 3D and floating-point power is needed. Like maybe a new, updated coprocessor that either retires multiple FP and "Multimedia" instructions, or implements a new architecture geared towards this kind of software. Real-time rendering on generic hardware would be awesome!

    (c) Oh yeah, and of course this has hack value. If anyone "doesn't get it": you probably need to reread The Jargon File and FOLDOC, and if you still don't get it, ask yourself how you ended up on slashdot.
    pb Reply rather than vaguely moderate me.
  • I compared our companies G4 450 to my Athlon 600 using Matlabs "bench" utility. The Athlon was much faster in the "Integer" (0.33),"Mixed Integer/Floating" (0.33), "2D" (0.77) and "3D" (0.71) catagories and the same in the "Pure Floating" (0.27) catagory. My Athlon is much slower when looking for ET but at least comperable if not better when running our companies propritary software.
  • A Hertz isn't a byte. 1GHz = 1000MHz = 1000000kHz = 1000000000Hz EXACTLY. A Hertz is simply a periodic event measured as "the number of events per second". A byte is made of 8 bits each having two states and is most easily described with the binary or base 2 numerical system.
  • The x86 instruction set is such as hack job its not even funny. Go take a look at some of the Motorola CPUs to see a nice orthogonal layout.

    x86 can't die soon enough.

    The problem is, does Intel have enough influence so that the massess will want a Itanium?

  • Drunk? Please think before I post? This is a little over the top. The truth of the matter is that my argument was sound. MY DATA was incorrect. Nobody with false data, no matter how correct the logic, arrives at a correct conclusions. Now a sound argument is usually based on clear thinking. Therefore, two things;

    I need to doublecheck my data or information better.

    You need to relax.

    I was wrong, true. Next time just point that out. Otherwise, your reducing a rather intelligent past time to verbal buffonery.

    Rock hard, ride free
    Big Din K.R.

  • Wouldnt it rather be a contest to see who can design the most efficient CPU. From what I've been led to believe the x86 architecture is not very efficient, and has a very low Hz : performance ratio, vs. other competing chip architectures.
  • by Microlith ( 54737 ) on Monday November 01, 1999 @08:28PM (#1570959)
    Here. []

    This was done a few months ago with 2 P3 500s by Hardwarecentral. They said it wasn't terribly stable at 1055 Mhz (1.055 GHz), but it DID go over 1 GHz. In fact, they did it with DUAL CPUs.

    Results. []

    (Yes, their CPUs were unlocked, they modified the CPU for the cooling, but it does count)
  • Score:2?!? - more like Score:0 (Troll/Flamebait)
  • "While waiting, greediest in MHz can fold back itself on the systems KryoTech Cool Athlon 800 and 900..." - greediest in MHz? Fold back itself while waiting? Sounds like a contortionist that wants all the MHz for itself...
    Actually, I think it sounds like PCs that get too fast start to travel back in time...
  • Why do it yourself when you can make the research student do it for you (especially since they can't say no...)
  • Actually, AMD has demoed K7's running at 1ghz at shows and such using Kyrotech.
  • Giga in software terms is considered a power of two (because the computer works in base 2, or one's and zero's). It's easy for the computer to work with.

    In most scientific fields, it is very inconvenient to work in base two, so people often use a number system called "base 10"; don't know where they came up with it (something to do with the number of toes a person has or something rediculous like that ;).

    As a result, MHz is a power of 10, GHz is a power of 10, KHz is a power of ten, etc. If you're really curious, I'd suggest getting an old physics book (the one you chucked out the 10th floor window after finally finishing that last required physics class) and look at what they define "mega" and "giga" as. It should also list what powers "nano" and "um" are as well, to give you a general idea of how really SMALL (yet, not small) the stuff inside your computer is.
  • This guy is apparently an uneducated 'normal' computer users who makes decisions with the 2% of the computer knowledge he has... I wish these kinds of people would educate themselves a little further before speaking. This is how the market thrives on gimmicks, small increments in technology, and hyped products. Please! Think, then write!
  • forget those previous replies, this is a funny comment! good laugh...
  • You know, it *does* take engineering effort to make something good. Intel's *problem* is that they're made to be cheap. If you want something good, it comes out of someone's pocket. What's so unreasonable about $2000 for a kickass G4 system anyway? Sure, if you want to really crank it, you can spend a lot more, but you clearly want a compromise between cheap and good. Or you could get a low-end G3 and overclock the fucker till it's blue in the face. Oh, wait; it already is. My old beige G3 233 runs happily at 300MHz. I'd overclock it more, but a) I don't need to (compiling is more or less instantaneous), and b) I like it a whole lot and don't want to fuck it up.
  • It is not the same thing to go over 1 GHz with the combined frequency of two chips

    If understand the article correctly they managed to overclock 2 Pentium III 500 MHz to 1055.79 MHz *each* but I dont't know if this is believable or not.
  • The SH3 processor is a good example of nice code density. The StrongARM is also a good example. Unfortunatly they trade off a FPU for smaller die space and lower power comsumption.

  • no biggie - I get 552Mhz from my Celeron 366s. Am I a hoax too? The bus speed isn't always _exactly_ 133Mhz or 100Mhz. Mine runs at around 100.4xx , and changes all the time. with a multiplier this big, what's wrong with getting a 4-5mhz extra?
  • Apple says a lot of things, in this case its "G4's are twice as fast as pentiumIIIs*" if you follow the astrick, it leads you to the statement "in 'cpu mesurement' and Photoshop tests". I'm not sure, but I'd be willing to bet that photoshop is optimized for the Mac harder then the PC (though I could be wrong). In any event, they didn't mention what photoshop tests were done, some have usualy done better on x86 with mmx.

    as for 'CPU mesurement', well that dosn't really help. What I was asking for was acutal SPECint and SPECfp for the CPUs, and real numbers not someone say 'up to twice as much' or other meaningless nonsense.
    "Subtle mind control? Why do all these HTML buttons say 'Submit' ?"
  • I was thinking of chaning it, but you've convinced me to keep it. :)
    "Subtle mind control? Why do all these HTML buttons say 'Submit' ?"
  • Ok, so you can super cool it, and get it running at >1Ghz; but where is the beef?

    ALL of the current fast processors are crippled by a lack of memory bandwidth; all of these high multipliers and slow main memories make for great benchmarks when the code fits in the L1/L2 cache but the speed goes into the toilet when the chipset hits the screeching breaks and must talk to main memory.


    Or if RAMBUS can ever get the initial latency down, a four or eight channel RDRAM rig would rock.
  • Gawd, reminds me how junked PCs are. Maybe someday prices of workstations will reach our grasp :P Then we can observe the real power of computers, not some silly gaming machine.
  • yeah the name sure fooled lots of people. Shame on Intel.
  • intel hits 1GHZ, but pathetic poor performing unscalable machines form sun and sgi have slow speeds like 400mhz and less. that says it all really.

    A 3-inch tall man with a hundred-pound rock on his back can move his legs ten times faster than I move mine and still get nowhere. [For our slower readers: hundred-pound rock == ancient architecture produced without forethought and a refusal to destroy back-compatibility]. Clock rate is fairly meaningless, and people who do not understand this are the problem with the computing industry today.

    in todays enterprise market you cannot top intel for price/performance.

    Well, I'm sure it was unintentional but you let a nugget of truth slip in there. Of course, this only applies to P/P of actual CPUs, and only for CPU-bound applications, and finally it should be noted that, strictly speaking, AMD not Intel is the leader.

    it leads the industry in scalability and reliability as well as being the flagship platform for the greatest OS every written, namely Linux!

    What a joke. Scalability? Of what? The only area I can think of in which the PC architecture scales well is heat production. Reliability? I suppose there's a sick truth to this: PCs can be relied upon to offer low quality, terrible scalability, and atrociously bad design all around, from the processors to the cases to the i/o subsystems. That's reliable, I suppose, in the same fashion as Microsoft produces reliably bad products.

    this is great, sun/sgi/hp will be out of business within a year, they just cannot compete with such superior technology.

    One might. Not all three. Have you ever used anything other than a PC? If so, you'd know why such companies are still in business. Hint: it's the hardware, stupid. I'll agree that (for instance) Solaris can kiss my ass [happily running Linux on a sparc], but there's no way you can find me a PC-type machine that I could use as a drop-in replacement for, let's say, an Origin 2000. Not gonna happen. I'm truly sorry you've never had the opportunity to work with real hardware. And even more sad that people with your attitude will eventually make sure that nobody gets that opportunity. We've just seen the end of it all, folks, all wrapped up in one brain-dead post.

  • I definitely hope so. For many tasks parallelism is the easier and more logical way to gain speed, for instance, neurocomputing, computer graphics, games, signal processing etc. Many of the problems could be reduced to separate problems that can be solved concurrently.

    Hopefully we will also see transition from programming languages that don't support parallelism (like C/C++) to better languages that are easier to parallelize even implicitly by the compiler (like the functional and logic programming languages).

    The programming tools are not really up to massive parallelism yet and it may be one factor affecting development; the change requires a paradigm shift. I don't think many people would want to program a 64 threaded program in C.

    I also hope the parallelism will become fine grained where individual statements in procedures can be executed in parallel, which of course puts the burden on the compiler.

  • Not quite relevent, but I'm curious:

    I have never like overclocking because I was under the impression chips burn out faster. Since I am a starving student, lifespans are important to me (the 3 boxen I am running are a k5-75, Cyrix 6x86 P200+ [150 MHz, what a gyp!] and a P-100, all of which have been in use almost 24/7 for 2-3 years).

    My question is: assuming the computer stays on 24/7, is the total number of MHz*hours greater with or without overclocking? In other words, taking into account both speed and lifespan, do you get more for your money with or without overclocking?

    Obviously, this has a large number of simplifing assumptions: a) the mobo/chip burns out first, b) lifespan is shorter than total upgrade cycle, etc. Still, I'd be curious, does anyone have any sort of numbers one way or the other?

    (The email is real. I don't check it though)
  • No I truly don't give a rats a$$ about games. Sorry. Coding is far more fun. :)
  • There are a lot of things running at those frequencies. Take GSM-mobile phones. They run at 1900mhz in the US and 1800 in Europe. If you just crank the output up a bit you'll be able to fry chicken with it. As of today we just fry our brains...

  • thats not that cool.... i could hit 1ghz by overclocking my dual p3 500s to 533 each. but then that would be on two chips and not one... =P
  • ... Intel would allow itself to slack-off and stay with the same CPU lines for much longer. AMD are excellent competition for Intel, and that can only mean one thing: great CPU's for us the consumer at tug-of-war prices.

    I only run AMD, but what's pushing AMD to come up with great chips like Athlon? It's Intel. What's forcing Intel to come up with 733 MHz chips? AMD.

    Competition is good. Flattening out other companies is baad.
  • Not that I think anybody reads posts to old stories here at Slashdot, but one point is that
    copper interconnect on both of the current techniques, doesn't really lower resistance. The resistance stays the same - the technique has difficulty making tall thin wires (which are the way current lower-level wires look like), so instead the copper process reduces the thickness/height of the wire by a little less than half. Thus the capacitance of the wires drops while the resistance essentially stays the same.
  • "I WANT LOW-LATENCY DDR PC266 OR BETTER SDRAM!" well, dell now makes 400mhz ram. not sure how, or much of anything else about it, just saw it in a pretty brochure they mailed me. ;) in any event, screw rambus, screw ddr, if you can get 400mhz main memory. sounds like it would rock on athlon 200mhz fsb, I wanna know why no one has done it yet.
  • Uh, d00d, Hz (and MHz and GHz) is metric.

    I guess the non-metric version would be "cycles" or "cycles per second", as I recall from old books and movies.

    Which are just Hz under another name, but sound kinda cool - "Zounds! Professor, this interocitor is running at a speed of one million kilocycles! Surely it's the product of an advanced civilization!"

  • Actually, there is a bit more headroom available.

    Some boards using the chipset from Apollo have an 8x multiplier for the 133MHz FSB. So with some very insane cooling, an engineering sample, and a board that uses the new chipset (i.e. DFI's PA-61), you could actually push the CPU to ~1066MHz.

    Why push it that high? Why not? We try to be the fastest at everything else, so why not CPU clock speeds?
  • Close. Better said the RISC philosophy was to reduce the number of cycles per instruction by simplifying the instruction set without increasing the instructions needed to execute a program too terribly much. This also ment smaller and cheaper CPU's. It worked pretty well but the cost of silicon real estate dropped so far that former RISC shops have been loading their CPU's with fancy features to the extent that I've heard the G3 has a more complex instruction set than the PII! The RISC/CISC distinction is outdated and misleading. See ml
  • Wasn't that picked because it is the resonant frequency of H2O?
    If so, then it would stand to reason that anything off that frequency would be a very inefficient microwave heater.
  • Your implies possession as in "Your problem is that you talk before you think."

    Use "You're". It's the contraction of "You are" as in "You're looking pretty stupid right now."
  • I stand corrected. The main point is still valid: Copper is faster because RC went down. Instead of R being less, it sounds like it was C. Thanks for the info! :-)

  • >> 1) If I'm not mistaken, the L1 cache in the PIII is at least 64k (32+32). According to the screenshot (which can easily be faked) it's 16 and 16.

    The L1 cache on a PIII is identical to the L1 cache on a PII, 16k+16k.

    >> 2) The L2 cache of the 733 PIII is 256 kb ONBOARD! Why is this important? Well first of all, there is nothing listed under the L2 setting. Did they have to DISABLE THE L2 CACHE to get it to work? If this is not the case, then why isnt' the L2 amount shown? If it was indeed disabled, then it's extremely doubtful that those benchmarks are reliable.

    The L2 cache on CuMine is new and (slightly) different. Many programs and some BIOSes that were released before the CopperMine don't know about the new L2 and missidentify it. It's no big deal so Don't Panic.

  • by BDKR ( 86989 ) on Monday November 01, 1999 @10:24PM (#1571009)

    1) If I'm not mistaken, the L1 cache in the PIII is at least 64k (32+32). According to the screenshot (which can easily be faked) it's 16 and 16.

    2) The L2 cache of the 733 PIII is 256 kb ONBOARD! Why is this important? Well first of all, there is nothing listed under the L2 setting. Did they have to DISABLE THE L2 CACHE to get it to work? If this is not the case, then why isnt' the L2 amount shown? If it was indeed disabled, then it's extremely doubtful that those benchmarks are reliable.

    Performance of todays processors is almost as much dependent on cache performance as archetecture and design. By increasing cache performance and memory throughput through the chip, you are decreasing the amount of time that the processors instruction pipelines are left empty. With no L2 cache, this would prove an incredible crippling of the chips ability to execute instructions.

    The only other possibility here is that the L2 is disabled, but the test scores are real. This being based on the idea that perhaps, all of these benchmarks are running within the L1 cache, which if true, would mean that the tests are not indicative or real world performance.

    But then again, I could be wrong,
    Big Din K.R.
  • Hmmmm I run UltraIDE-33 and it gets around 3-4 % cpu usage on sustained random seeks (win98). It's all about drivers ... the big problem is that a lot of Win systems come with DMA disabled by default. Get rid of that pesky PIO and all things are good again.

    Oh and as for those spare cpu cycles...I could really use a few hundred million when I'm doing video editing & mpeg compression. Leaving the PC on overnight to chomp files is something I thought I had got past all those years ago when my 486 got replaced with a P133!

  • I really don't think you understand how this works. An R12k SGI Mips processor running at 250MHz on a typical O2 will absolutely blow away a 1 GHz PC. Even the older R10k at 180MHz is much faster than a 500MHz PC. One reason is that they don't use a bus. All memory is shared instead, which gives you over 2.1Gb per second data transfer!!!! And then we go up to 1000+ processor machines with over 1 Terabyte of RAM (That's 1000 Gigabytes - roughly)
    We like the Cray T3E
    There is no contest. Believe me! PC's are currently baby food.

    On the super PC note, you must have seen Slashdot's article about SGI's new Linux Supercluster. Should be the fastest machine in the world.

  • I don't have any hard facts to confirm this either, but I've had my two systems (AMD 133 o/c'd to 166MHz, and a celery 366 up to 458MHz) since February, w/ no problems. It's my understanding that you have nothing to worry about unless there's heat in excess of 60C(140F). I could be wrong here. Basically, I say those numbers, because I've heard that some(older) AMD chips run that hot out of the box. Buying your own thermometer [] to measure the heat on the chips would be a good idea.

    As far as just leaving them on 24/7, I also remember hearing that turning your computer on is about the worst thing you can do to it (cycling power spikes, etc), so leaving it on instead of turning it on/off everyday or several times a day is actually much better for it.
  • I wonder how their translation system managed to translate 'front end bus' into 'drunk face side'
  • x86 chips are actually quite efficient in terms of ops/cycle.

    No, they're not.

    For example, loading a 32-bit immediate value to a register takes one cycle on a pentium or K6, but on a MIPS RISC cpu for example takes 2 instructions and 2 cycles due to the smaller instruction size.

    So? Generating 32-bit constants is not what interesting programs spend most of their time doing. Optimize the common case.

    The problem is that the complex instruction decoders use a lot of power.

    To sum it up, I'd have to disagree on several points:

    • Complex instruction decoders are a bottleneck to issuing large numbers of parallel instructions. It's more than just a power issue.
    • The 32-bit constant generation example you have is not a typical bottleneck in most code. Control flow is more likely to be.
    • x86 is less efficient than alot of other machines due to being a register-starved architecture. It has to rely on tricks such as store-to-load forwarding and micro-caches to turn a small cache of memory addresses into a pseudo-register file.
    • While Intel has generally been the "Floating Point King" in x86 world (with Athlon recently taking the title away), x86 floating point as always sucked compared to architectures that have traditionally taken FP seriously. RISC wins squarely here, period.
  • Nothing on that page indicated that Intel "clarified the situation." I saw no reference to the theoretical possibilities of running at 1GHz at all. What you (and all the other hoax-callers) are forgetting is that engineering samples sent to hardware reviewers do not have multiplier locks!

  • These milestones are completely arbitrary - we only use base 10 because humans like to count using their fingers. We use decimal because it's easy.

    If we had 8 digits on our hands instead of 10 we would have been celebrating when cpu's reached 512 Mhz (1000 octal). Woohoo!

    Also - we'd only be in year 3717 so we wouldn't have to worry about Y2K!! Of course we probably would have had to worry about the year 3700 since programmers were still trying to save those precious bits 17 years ago ;-)

  • by Anonymous Coward
    they realized that the performance bottleneck was never the processor. Doh!
  • That pesky constant 'c' must be holding back faster CPUs by now.

    Not exactly, but close. Signal propogation delay is getting worse and worse with respect to clock cycle. In the early days, you could consider wire to have zero delay, since transistors were so slow compared to the propogation rate on the wire. As wires have gotten smaller and smaller, their resistance has gone up. Meanwhile we've packed them closer and closer together, so we have tons of capacitance between wires. And finally, transistors have gotten orders of magnitudes faster.

    End result: It takes bloody ages for a signal to get anywhere on the chip with respect to how long it took to generate the signal.

    Most pieces of silicon nowadays operate with various "domains", each of which has its own local clocking. Depending on how fast you're running, it can take several clocks just for a signal to travel from one end of the chip to the other, so designers tend to subdivide problems into domains that aren't more than a clock-cycle wide. Pipelines and replicated hardware help a little, but physics really starts to bite you in the arse. Copper helps a little here (since it lowers resistance), but it's not a cure-all either.

    Our friendly constant 'c' is a couple orders of magnitude above the propogation rate on the chip, so it's not the main limiter. To put it into perspective, light travels 30cm every nanosecond, and chip dimensions are usually closer to 1cm on a side. But we're getting uncomfortably close. :-)

  • Awhile back, the X3J11 group responsible for the ANSI C standard was looking at some rather nifty data-parallel C extensions that retained the otherwise "serial programming" nature of the control code. I suspect that an evolutionary approach such as this is likely to gain more ground than forcing people to think about programming in a completely different paradigm.

    Postscript and text files containing the Data Parallel C Extensions draft is available here: []

  • Don't go so far..maybe a dual though. ;)
  • Oh, man. that sounds nice, but I seriously wonder what kinda of cooling we need. KyroTech or above? At the lab here, we have Duel PIII-600's in Mini-Tower [had to go Mini-Cause we have 20 of thse suckers] and 2x9 GIGS SCSI HD and SMC 1211TX NIC and they generate heat. [Each of them runs a program called GAUSSIAN which somehow runs for like two weeks each time and generates scratch files totallinng 8 to 12 gigs each run] and it does get damned hot.
  • So 733MHz goes to 1GHz.
    Next year we get an Athalon 1.1GHz, which we overclock to 2GHz. And so forth.

    But why is this such a big deal?

    Probably the same reason as going from 700 mph to breaking the sound barrier in the air was, or having the first car exceed 100 miles per hour. Its a logical progrression, but milestones are important in our culture. Why do we celebrate our 50th birthdays or 50th anniversaries? Well, because they are milestones.

    I'm still using a p/233MHz and it still offers me instant gratification for anything I want it to do. What's the rush?
  • What that means is that running your microwave will generate noise in your processor. Fortunately your processor is small enough this won't be a signifigant issue... but when the memory bus starts reaching frequencies where it can hear your household appliances... hardware debugging will get fun!

    Imaging going through your kitchen and turning off all the appliances so that your overclocked system will be stable enough for running Quake 3!

  • Okay, if that's the Athlon Killer, where's the Athlon Killer Killer? I'd like to have that.
  • >>It's no big deal so Don't Panic. OK ripper, I'll try not too. Later on, Big Din K.R.
  • Now the big question is -- where the hell did they get a PIII CPU with a multiplier of 7.5? Coppermine 733 has bus speed 133 and multiplier 5.5. Since the multipliers are locked, there is no way to change them. The PIII would have to work at 750MHz (100MHz bus) to have a multiplier of 7.5, but such CPU does not exist. The only CPU that has a multiplier of 7.5 is Celeron 500.

    So does that mean it's a hoax?
  • Wasn't that picked because it is the resonant frequency of H2O?

    I don't remember exactly, but it's something like that. Maybe the resonant frequency of the H-O bonds, or maybe one or other resonant frequency of the H atom itself (in that case microwaves would not only heat water but also all carbohydrates).

    If so, then it would stand to reason that anything off that frequency would be a very inefficient microwave heater.

    That's right. But who knows what other atoms or molecules vibrate at other gigafrequencies... Let's hope silicon doesn't have a resonant frequency at 1.5 GHz, for example. Those CPUs wouldn't last very long.

  • Heheh it would be funny if they were using the metric system, and had really just cranked a 333 up to 500 or so :o)

    Yeah yeah i know that's not it.

    I'm just saying it would be funny :o)


  • Isn't 900Mhz the frequency at which microwaves ovens operate... will PCs really fry brains?
  • Well, my condolences for your bedroom, but really, I'd just like to see either one of them trash the other. Because, to do so, they'd have to come out with a line chips that was *much* faster than the other (the differences now really aren't that great, relative to each processors speed) and it would have to have a viable pricetage. Cheap speed? Sounds good to me.

    Then, after the solid whooping, I'd like the other one to come back and do the same. A nice see-saw of continuous ingenuity that allows me to maximize my utility, while one company or the other maximizes it's profits.
    Let the best man win!


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

  • by Anonymous Coward
    might be an engineering sample...
  • According to an article I read in New Scientist, isn't 1GHz around the frequency one uses in domestic microwaves? So as well as yer average Pentium running hot enough to fry eggs, will it be that the inside of a computer is going to resemble a combination oven now...?
  • WOW WOW You should know that Intel always send special samples of their Cpu to the press. They are not locked and you can change both the multiplier and of course, the bus clock. Therefore, this articles doesn t represent what the man in the street can do with a P III 700 but what the cpu is able to do whitout these (stupi....) locks. I am a writer for Hardware-fr (the site which DID IT : ) and i saw it whith my own little eyes..promised.
  • The 2 is due to karma. The post hasn't been moderated.
  • Posted by cookieman.k:

    Hi! Does this mean that my system will (c)rush in to the BSOD whit greater speed ?
    Or is it Microsoft presuring INTEL to make the csip faster due to the great time requirements of booting Win2000.?
    How much time do we have left ? Our UPS wont resist more than 3 hours , sir...
  • (b) If you have a special-purpose task, why are you using general-purpose hardware?

    eer.. because it is what the bean counters agree to pay for?
    And at home my $2K self assembled box beats the crap out of $15K+ workstations of a few years old.. How's that bad?
  • Close, the idea behind RISC is to do more per cycle. This means you want instructions that are easier to decode and which are pipelineable, which means you may be doing a register write, register read, instruction fetch, instruction decode, and ALU operations all in the same cycle.
  • This Pentium III was cooled, as was the Athlon that AMD and Kryotech demoed about 6 months ago. I read about it right here on Slashdot. []

    ?syntax error

  • we're probably going to see multiple cores on one die with AMD's next gen sledgehammer
  • It's fake People there is no way it could be done:

    First: The program recognizes the CPU as a coppermine witch it should not be able to do yet. Since this is a new CPU

    Second: The chip was tampered with Intel set the lock multipliers at 5.5 AKA they fucked arround with the insides.

    Third: The board does not officially support the chip. Meaning they're roasting the board.

  • I do believe that the 733Mhz Intel CPU only came out last week. Could the programs already be updated to recognize Intel 733Mhz chips? I seriously doubt it. And I looked at the specs for the mainboard they used for the 1Ghz chip, although it does support up to 8.0 mulitplier, it only mentions support up to the 500Mhz chip. (This doesn't mean that they can't be used for the board, but motherboards required a BIOS update to support AMD chips, I am sure something similar would be required for a 700Mhz+ chip) And it is likely that the Intel 733 chip used in overclock could not be obtained by anyone as the mulitpler was obviously unlocked for these tests and will not be sold that way. Something is not right about these tests. A more reliable source is needed.
  • Posted by cookieman.k:

    Hi !
    I haven't observed that chips burn out faster, tough I've overclocked all my processors till now starting with the old 486Cyrix DL 40 -> 66 Mhz, Mhz 486DX2/66->80. AMD 5x86 133-> 150 Mhz, K6II 266->333 Mhz, and much more(I can't remember how manny of them).I must tell you that i've done some pretty hard thing with some of them : ex: Cyrix 6x86 133 -> 200 with 2.8V to 3.2. An it works right now very well.
    Be sure that they were used hard. My Brother used to put the machine at long rendering session (8 hrs or so) with 3DMax and he never had any problem.
    I think that technlogy evolves faster enough to use the overclocking technologies, CPU's wont explode on you, peopels shift through CPU's faster than they burn out. You should get the maximum performance while you can.
  • Subject says it all
  • >Surely those extra Half-Life frames aren't going
    >to make that much of a difference?

    Well if 5 years ago you'd said about 100mhz processors "Surely those extra Wolfenstein 3D frames aren't going to make that much of a difference?" we wouldn't HAVE half-life.

    Games will easily be able to consume way more CPU than we're giving them today. Better physics, better modeling, better lighting, etc.

    - SteveX
  • it's rather interesting how analogous the semiconductor market is to the manifest destiny belief of the US 'progressive' movement in the early 1900's. Some may argue that it is an innate desire or urge to 'strive for more' or 'further our development', but *enough* never tends to suffice. i await the day that i awake desiring absolutely nothing.

    in the meanwhile i'll crank up that voltage setting...

    -raj jr
  • It's a fairly run-of-the-mill "Hey I overclocked my processor to Ludicrous Speed!" article. However, I nearly fell out of my chair laughing at some of the truly odd translated phrases...

    "If no processor given rhythm at this frequency..." - these processors have rhythm baby! This line reminds me of the dancing bunny-suit Intel engineers.

    "While waiting, greediest in MHz can fold back itself on the systems KryoTech Cool Athlon 800 and 900..." - greediest in MHz? Fold back itself while waiting? Sounds like a contortionist that wants all the MHz for itself...

    "While waiting, history to dream a little and to see what will have in the belly the processors of demains..." - no comment. Too good for words.

    "...on a chipset BX the AGP is to 88 MHz when one uses a drunk face side of 133 MHz..." - my question is, if it's drunk then is it still greedy? Does it still have rhythm? (I know that I lose any rhythm I have when I'm trashed)...

    "...the profit of 39% is a little less significant but remains impressive, even if it is also with the read-write memory given rhythm to 133 MHz." - ok it's getting dumb now (the joke, not the article) so I'll end it here... but I like to think that my read-write memory has rhythm too, since my Celeron 300a doesn't shake its stuff too well yet...
  • by Palin Majere ( 4000 ) on Monday November 01, 1999 @07:10PM (#1571067)
    The simple facts (from Tom's Hardware [])

    Coppermine processors:
    Rated Speed Bus Speed Multiplier
    733 133 5.5
    700 100 7
    667 133 5
    650 100 6.5
    600 EB 133 4.5
    600 E 100 6
    533 EB 133 4
    550 E (PPGA) 100 5.5
    500 E (PPGA) 100 5

    Now, since the multipliers are locked on Coppermine processors, there's no way that these people could get a 7.5 multipler. The highest is a "7" with a front-side bus of 100 mhz.

    Note that they also make absolutely no mention of any method of cooling this thing. The only thing presented are the benchmarks, all of which are directly proportionate to the increase in mHz rating. Coincidence? I doubt it. There's no mention _at all_ of the hardware this was tested on, except for one screenshot that mentions an ABIT BX6 (Rev. 2) motherboard.

    Before you go expounding on the wonders of 1ghz Coppermines, it'd be wise to check the facts first.
  • Well, the good news for AMD is that this is just an Overclocked 733 (although, I can hardware junkies trying out the 153mhz bus speed on some 'overclocker' boards like the bx6-2), so right now AMD still has the fastest chip. Of course, if the yields are high enough over at Intel, they could hit 1Ghz quicker then AMD.

    I personally hope that AMD can keep ahead of Intel; their competition has dramatically lowered the cost of PCs in the past few years. When I purchased my p75, it cost me $1500, and that was about as far down the line as you could get in terms of CPUs (The chips out were the 75, 90, 100, and 120 (little did I know at the time that my little 75 could be overclocked to 133). This Saturday I saw an Athilon 600 system for the same price (made by IBM no less). Cheap PCs are good for everyone :)

    Of course there are probably other reasons as well, such as the emergence of the PC as an 'internet appliance' But if it weren't for AMD, they'd all be using Pentium 200s, instead of celeron 466's (witch by the way, was selling for about $500 in a Compaq that day.)
    "Subtle mind control? Why do all these HTML buttons say 'Submit' ?"
  • The way I see it, we'd have had retail 1GHz P!!! processors even before now if only Intel had its mind on the right technologies instead of pie-in-the-sky technologies that no one really wants. How many of us really care about Rambus, considering that we haven't nearly reached the limits of SDRAM? Think about what graphics cards manufacturers are using in their next-generation (or for the GeForce, current-generation) cards--Double Data Rate SDRAM, much as the Athlon FSB is really 100MHz but transfers data twice in the time it takes a normal FSB to transfer once, DDR SDRAM would effectively double memory performance whereas Rambus will start out more expensive than SDRAM and perform worse than 133MHz SDRAM.

    We could have DDR SDRAM 200Mhz memory right now if Intel had supported it instead of Rambus, and DDR SDRAM would quickly reach effective speeds of 300MHz--150MHz x 2 times the transfer rate.

    But what does this have to do with the 1GHz mark? Intel pushed Rambus for its own agenda, not caring about customers' needs; likewise, development has been almost completely shifted to Merced (oops, make that stupidly-named-Itanium) instead of pushing x86 to its limits first. Were it not for AMD--and this is supported by Intel's own "development roadmaps"--we wouldn't even have seen 700MHz Coppermine this year (and we still won't get it in quantity this year). We have AMD to thank for 700MHz P!!!, which is reason enough in my book to buy Athlon--Intel simply does not care about the consumer, they care about pushing the unnecessary and too-expensive technology of their Rambus partner, and they care about finishing their high-end server processor Itanium; they do *not* care about making their soon-to-be-low-end-compared-to-Merced P!!! run as fast as it can for their customers. Why do you think Intel is suing VIA, whose 133MHz SDRAM chipset beats Rambus performance to Hell and back? This is not off-topic--this is why we don't have 1GHz retail processors by now, which is as on-topic as it gets. Intel, we want x86 at well over 1GHz before we even want to think about Merced. We want you to care about what we care about. But since you don't, a lot of us are going to start thinking AMD Inside instead of caring about Intel, and by the time your prized Itanium rolls around we just might drop it in favour of AMD's 64-bit x86-on-PCP offering. Think about it.

Time to take stock. Go home with some office supplies.