Follow Slashdot blog updates by subscribing to our blog RSS feed


Forgot your password?
Intel AMD Technology Hardware

The Mother of All CPU Charts 176

||Plazm|| writes "Tom's Hardware has an entertaining read on the latest offerings from processor makers Intel and AMD. Not only does it contain a plethora of benchmarks on the latest Dual core CPU's, but it also includes benchmarks from over 60 other legacy processors. Better yet, they let the benchmarks speak for themselves and let you draw your own conclusions. You may want to fill up your 44oz mug before sifting through this one, though."
This discussion has been archived. No new comments can be posted.

The Mother of All CPU Charts

Comments Filter:
  • by strider44 ( 650833 ) on Thursday November 24, 2005 @09:47PM (#14110354)
    Summary: AMD wins every single result except the synthetic Sandra benchmarks, which Intel wins quite convincingly (all except one test). Something tells me there's something slightly wrong with that benchmark.
    • by ciroknight ( 601098 ) on Thursday November 24, 2005 @09:59PM (#14110391)
      This seems very odd to me. Intel's usually at least competitive, now it seems that Intel has almost stopped competing in raw performance entirely. This isn't because they can't build fast chips; everyone remembers how heated those battles used to get (both metaphorically and physically).

      But in all seriousness, where is Intel? Parts of me think they've almost entirely abandoned the race with AMD simply out of spite the Pentium 4 didn't work out as well as they had hoped, or that they're trying to move everyone into Mobile computing mode with their new chips which have been on the burner for the better half of the new century.

      When were the latest chips released by each company? It seems Intel's gone into hibernation mode kind of like they did right before releasing the Pentium 4 in the first place. (Allowing the P3 [and now P4?] market(s) to stagnate and die off?) Come on Intel, what are you up to???

      Not that I don't love AMD winning; it just seems AMD does their best when they're pushed excessively by Intel to produce. Now AMD doesn't even make chipsets and their mobile offering is still quite the joke in the face of the Pentium M.

      Eagerly awaiting the speed wars to start back up.. I'm ready for some bargains!
    • According to the article regarding the Sandra benchmarks:

      "Part of its draw is that users receive a performance rating for their computers within seconds. Since this is a synthetic benchmark, it sometimes yields results that seem out of touch with reality."

      I'm reading that as Tom's Hardware's way of saying, "Look, we know that these results don't match up with every other real-world test we conducted, and we suspect that it's the test that's wrong, not the real-world results. The only thing that confu

    • I would agree. Have a look at this [] page. Scroll down, and notice that there is a group of about 14 AMD processors essentially got the same score. These processors range from a Sempron 2500+ all the way an Athlon 64 3700+. Go up a bit an there is a simular group of about 18 Intel processors that get essentially the same score, where the processors range from the P4 2.4Ghz to a P4 EE 3.4Ghz. What gives?
    • But Intel has a special feature: The die of the 600 series Prescott processors takes up 135 square mm^2 [] making it the first 4-dimensional chip around.
  • The floppy (Score:3, Interesting)

    by afaik_ianal ( 918433 ) on Thursday November 24, 2005 @09:51PM (#14110363)
    From TFA: "One stalwart component has survived through all of these innovations: the 3.5" floppy. [...] The floppy is the only component that still remains in use today".

    Do people actually still have floppy drives in their PCs? I haven't owned one in many years, and wouldn't have a clue where to get floppy disks even if I had one.
    • 1. Yes.

      2. Office Depot
    • Oddly enough, people do.

      The school I work at requires all students have a USB thumb drive of at least 512MB though. It's nice to see a school environment devoid of floppies.

      • I haven't used one in months; the classes in which I use PCs let me use a USB drive too, and my PC has no floppy drive. It does have a floppy-drive-sized space, but I have a digital cam that uses the small SD cards, so I might fill that space with a card reader (or not, and buy a Dell 2405FPW [] with its readers, since I seem to read at least 2 more good reviews of it every day and I'd like me some flat HDness anyway).
    • Re:The floppy (Score:1, Interesting)

      Hell yes. Ever tried to copy some 1 meg network drivers from your networked computer to your test machine? Can't burn it(you could, if you were rich, but then you would have to burn one disc everytime you discovered your previous driver wasn't the correct one), so you use a floppy disk. Thoguh I now prefer USB keys for that job, sometimes the driver you need is FOR the usb, so obviously that wouldn't work. Floppys are still handy little buggers.
      • Can't burn it(you could, if you were rich, but then you would have to burn one disc everytime you discovered your previous driver wasn't the correct one)

        Enter the CDRW - problem solved!

    • Every college bookstore I've been in has had floppy disks. Any electronics store (Best Buy, Circuit City...) has them easily. And the vast majority of pre-packaged boxes (anything from Dell or Gateway, the ilk, not homebuilt ones and possibly not the $300 Walmart special) have drives. It's still by far the easiest way for technologically illiterate people to transfer files-easier than burning to CD, doesnt use a different system like email, and less expensive than a USB drive.
      • My friends slightly-smaller-form-factor computer currently has the floppy drive installed inside of the case facing backwards in a harddrive slot.

        You need a floppy to install windows onto a computer with an SATA drive and we decided it might be needed was an ugly-assed drive without a bezel that was going into a fairly pretty case so we stuck it in backwards. All you have to do is slide off the side of the case and you can insert a disk (if it is ever actually needed again).

    • We had a little optics lab "image processing" computer with: no active ethernet connection (within reasonable cord distance), no usb ports, and no CD burner.

      Aside from bodily moving the CPU (don't want to do that when it needs to be near the microscope), there wasn't really any other convenient way to get modestly sized images and text files off it than through floppies. The next challenge was to find a machine in the department that: was connected to the school's intranet, had a floppy drive, and wasn't
    • I still have a floppy, but then this PC is a few years old now.
    • Book publishers and agents still require you to snail mail your book on floppy. (I'm not talking AW or ORA, I'm talking publishers who do novels, and such.-

      • Is there any reason why they would not support CDs? It seems to me that CDs would be a lot safer to mail (CDs tend to be more resiliant to environmental effects than floppies).
      • Book publishers and agents still require you to snail mail your book on floppy. (I'm not talking AW or ORA, I'm talking publishers who do novels, and such.

        I'm surprised. I'm sure that most books or novels written in a modern word processor would very quickly exceed the 1.44MB that a floppy disk holds. Or do they make you send the novel as a .txt file too?
    • You can still buy floppy disks from office supply stores.

      I still add floppy drives to all of the systems that I spec (order from a company) or build myself. They only cost $9 or so and it's cheap insurance against the day that you really need a floppy drive.

      Mostly I use them for flashing BIOSs (motherboard, RAID card, etc). Some of the motherboard BIOSs now let you put the BIOS image on a CD-ROM, and there's a tool built into the motherboard that will read the BIOS image from the CD-ROM and install it
    • Floppy drive will always be there, until there's a new standardized removable media supported natively in BIOS.

      I still put floppy drives into every computer I build, since it is still the only reliable way to insert driver during OS installation, and upgrade/flash BIOS of your system. Although most system can be booted via optical drive, but there's no easy/simple way to produce a bootable CD or DVD..

      some MB can boot via USB/flash drive, but it's again not universally supported, and the media compatibility

  • Then whose it's daddy???

    Thank you, I'll be here all week.
  • I personally use an Extreme Gulp [].
    At 52 ounces, It convienently holds 4 x 12oz sodas + ice
    and it will stay cold as long as you could want.

    Even long enough for you to click through Tom's Hardware un-printer-friendly website.

    and i thought this was funny too:

    I Am Extreme
    Yesterday at work I drank an Extreme Gulp [] while doing some Extreme Programming [], and then I went home and ate Extreme Duritos [] while watching Extreme Sports [] on cable.

    Today every muscle in my body aches.

    Posted on May 01, 2002
    http://www.defectiveyet []

  • Toms Hardware chart that compares CPU's which double as heaters!! Wow - that will be great for those cold winter nights reading slashdot.
  • Does anyone else find it odd that in practically all the benchmarks, the single core processors beat out the dual core processors? Each core in the dualies has the same processor as a single core, so why would that be?
    • by Anonymous Coward
      Because the benchmarks are pretty much single threaded, so only one core is actually being used, so on these particular benchmarks a dual core CPU won't show any benefit, it might even suffer a little, if the fact of it being dual core allowed a small background task that might have otherewise stalled to run and was therefore fighting for the cache.
    • "Does anyone else find it odd that in practically all the benchmarks, the single core processors beat out the dual core processors?"

      No more odd than say a pickup truck beating a sports car in a race to move from one apartment to another.
    • I was looking at this very recently. First, most of the tests are single threaded, so the dual core is of limited use. Secondly, and more importantly, the individual cores run slower on the X2 processors than the single core does. For example, the AMD64-3800+ runs at 2.4GHz; the X2-3800 runs two at 2Ghz. For single threaded apps, like current games or benchmarks, the single core chip will beat the dual core hands down. The AMD64 3800+ is cheaper too.

      For general heavy use on a non-gaming rig, you're still be
  • by Barkley44 ( 919010 ) on Thursday November 24, 2005 @10:13PM (#14110446) Homepage
    I'd like to see a comparison of average cost against the speed, since the real question is what's the fastest speed I can get for the money.
  • Where are the Motorola and PowerPC chips? hehe
  • Moore's Law (Score:2, Interesting)

    by dirtyhippie ( 259852 )
    Processor speeds haven't increased much in the past 2-3 years... Are we hitting the end of Moore's law, or just taking a detour as CPU makers decide dual-core is more important? I've been wondering about this for awhile, but haven't seen much discussion of this. Do we have to wait for quantum computers before we can get more single-thread performance?
    • Re:Moore's Law (Score:4, Informative)

      by cide1 ( 126814 ) on Thursday November 24, 2005 @10:31PM (#14110520) Homepage
      Processor throughput has increased tremendously. Clock speed has not increased. Issue widths are wider. Larger, faster, and more effective caches are being used, in addition to the introduction of trace caches. Branch prediction continues to gets better, along with speculation techniques. More physical registers and larger lookahead windows allow modern CPUs to pull more parallelism out of single threaded programs than ever before.

      Features like hyper-threading and dual cores give a much greater system wide speedup than simply raising the clock rate, and avoid all the problems of power consumption. Even on single thread performance, having another core to run the OS, so your not constantly context switching, can make a differance.

      Reading this article made me sick, because they equate speed with clock rate. This is patently false, as the last two years of computer architecture have shown us.
      • Re:Moore's Law (Score:3, Insightful)

        by dirtyhippie ( 259852 )
        I didn't say that clock speed = performance, however, an increase in clock speed on the same architecture does indicate a probable increase in performance. I'm also sure you know that it is arguable at best that hyperthreading increases performance.

        I'm also not sure where you got the impression that the article equated speed with clock rate - if they did, why did they bother with all those benchmarks?

        As an example of the stagnation of the past few years, I have some code whose critical loop is unparalleliza
        • I have some code whose critical loop is unparallelizable.

          What is it? Is that unparallelizable in theory, or practically speaking unparallelizable? I've yet to see a real world problem that couldn't be parallelized, at least in theory ;-), and I'm curious to see one.

          Not implying this is the case here but it's been my experience that non-CS scientists (e.g. physicists and chemists), while they are programmers, sometimes don't have a good grasp of just how much computation can be parallelised. Not surpr

          • What is it? Is that unparallelizable in theory, or practically speaking unparallelizable?
            "This new processor will run your apps faster" is a who lot different than "this processor will run your apps faster if they're completely rewritten it in a much more complicated and error-prone fashion." It used to be, when you upgraded your P3 500MHz to a P3 1GHz, you got a big boost in all your cpu-hungry applications automatically, no ifs ands or buts.
      • Re:Moore's Law (Score:3, Interesting)

        by timeOday ( 582209 )
        No, the last 2-3 years have been a disappointing let-down in cpu development.

        Dual core processors cannot be equated with single core. You can always make highly parallel tasks faster by throwing more CPUs at them, so what? There's a reason we didn't go to multi cores until single-core development stalled. If you want to compare dual-core, compare them to an SMP single-core system.

        Now look at the benchmarks. For instance, on the 3dMark05 Futuremark, the fastest single processor is the Athlon 64 FX 5

        • I disagree with you, except for the dual core part.
          Primarily because you base your thoughts on shaky grounds.
          You cite a benchmark where the Athlon 64 X2 4600+ is rated higher than the Athlon 64 X2 4800+, which has more cache. This is nonsense for these CPUs.
          And obviously, these benchmarks have lots of problems, they surely are compiled with Intel ICC versions which produce slower code for AMD procs, as the benchmarks are completely out of touch with reality, putting Intel processors on top of AMD's fastest,
          • Well, the article contains a pretty wide range of benchmarks besides the one I quoted. Do some of them support the assertion that CPUs have continued to speed up at a rapid pace?
  • Certainly not ALL (Score:2, Interesting)

    by msbsod ( 574856 )
    It looks like the author has a bit of a limited focus. This chart hardly covers all CPU's. Where are the Alpha processors? Someone else mentioned Motorola. Or take the Cray processor. People would be surprised to see how slow PC processors are! This chart even does not cover all AMD and Intel CPUs. For example, processors like the AMD 29K, Intel i860, i960 and the Intel Itanium are missing. But maybe the narrow view of Tom's Hardware Guide is what PC users want, no?

    And how about the good old Intel 4004? Jan
    • his chart even does not cover all AMD and Intel CPUs. For example, processors like the AMD 29K, Intel i860, i960 and the Intel Itanium are missing.

      You missed a big one: The Celeron. How come not one single processor from Intel's budget line was tested?
    • Actually, given that I'm not really in a 'buying mood' right now, I'd love to see some more interesting/exotic processors thrown into the mix. The one I'd really like to see are some SPARCs, although I don't think they'd do very well on the benchmarks that are being used (they seem to favor single-thread performance). But some big iron -- especially old big iron -- would be fun for comparison value. We always hear glib comparisons between modern desktop computers and the mainframes of the past, but I've nev
  • I'm in the process of building my "media center" PC, and I know that processors can become a bottle neck if there are more than 1 or 2 capture cards running, as well as live playback or other activity.

    Is anyone up to speed with which processor to go with for this, but more importantly, which capture cards to use?

    • Usually in such a setup I/O performance determines whether you can do anything at all with the system, especially PCI implementation. Practically the only ones workable with one PCI hungry capture/editing cards are Intel and SiS, I wouldn't be surprised if the case was the same with few "normal" capture cards.

      Of course only if you'd want to do few things at one time.

      Honestly...I don't see a reason, so not much of a difference if you'd have only one capture card.
  • by Anonymous Coward
    Heh, you call that a list?

    SPECcpu beats this hands down. THG is great and all, but SPEC are a non-profit organization *dedicated* to measuring the performance of computing systems. Believe me when I say their "CPU 2000" benchmark is not only the standard benchmark, but the *best* standard benchmark out there. It's cross-platform: Windows, Linux, HP-UX, AIX, whatever: you name it, it's been tested. It's cross-compiler: GCC, Intel ICC, AMD/Pathscale, IBM xlC, they're all here.

    Here's the list. It's big. []

  • "You may want to fill up your 44oz mug before sifting through this one, though."
    Station operator> Do you want the 32oz or the large?
    Bill> How big is that large?
    Station operator> You're gonna wanna pull your truck up out back. I'm gonna go start the pump.
    Bill> Shit that sounds like a lot of coffee man. I don't know if I wanna be awake that long in Tennessee.

  • It's a nice list. Seems really lacking without the Pentium-M line though. THat's where intel is going.
  • How is this the "mother of all CPU charts" when it leaves out all the dedicated mobile processors? I'd like to know if that Turion 64 laptop I've been looking at will outperform the P4 unit sitting next to it, or what clock speed of Centrino or Pentium-M will beat my two-year-old desktop.

    Does anyone know where I could find a chart that compares mobile processors to their desktop equivalents? I really am about to buy a laptop, so it'd be quite helpful.
  • Gosh, my 1GHz Athlon MP doesn't even make the "legacy" category any more. I'll just go cry now...
  • I wasn't surprised to see my processor (1.4GHz Athlon) on the list... I WAS surprised to see that it's the OLDEST and slowest on the list though... especially since it still handles everything I throw at it with no problems.

    I'm willing to bet a large percentage of slashdotters still use processors that aren't even on the list anymore... and feeling no need to upgrade... am I wrong?
  • Here's some things i particularly enjoyed:

    -The charts showing "...performance per watt of processors from 1993 to today" that mention time on neither axis.

    -The picture and caption describing James though most of his readership had never heard of him.

    -The picture of a sticker with the caption that read "This seal guarantees that the box is unopened.". Duh.

    Good stuff...good stuff. I do however agree with the assesment that the floppy drive is not dead. Although I may use it once a year if i'm lucky
  • I put together this web-scraping chart comparing price/performance of available x86-compatible CPU families on []. The daily-run script collects web-advertised prices, and displays them as a series of lines showing price-vs.-nominal clockspeed data within the CPU sub-families.

    Note the logarithmic scale of the Y (price in US$) axis---in linear scale it's easier to see the knee in the curve, where additional speed increments begin to cost disproportionately more, but the linear graph

"The way of the world is to praise dead saints and prosecute live ones." -- Nathaniel Howe