Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
×
Hardware

Ten Lies About Microprocessors 59

Posted by timothy from the lies-they-tell-you-lies dept.
cloudkj writes "Processor selection too often turns into a religious war. Debunking the dominant myths is the first step towards making a rational choice. Embedded.com has an article highlighting the 10 most common lies and misconceptions about microprocessors."
This discussion has been archived. No new comments can be posted.

Ten Lies About Microprocessors More

Ten Lies About Microprocessors

Comments Filter:
  • Contrary to their names, the north and south bridges do not actually lie to the north or the south. Nor are they bridges. They're actually tunnels to the east and west.

  • INTEL Clarification (Score:3, Interesting)

    by trompete ( 651953 ) on Sunday July 13, 2003 @01:09PM (#6428936) Homepage Journal
    I'm glad that the article author clarified just how important Intel is to the entire processor market.
    Those of us who only use desktop machines have a hard time seeing past Intel/AMD/Motorola. Let's face it: the next processor decision I'm going to make is going to be whether I want to stick with AMD or go to Intel for my next gaming machine.
    • Did you happen to notice that the article was from embedded.com? And that possibly the article was meant about embedded microprocessors for the embedded engineering crew? This article was not about the next processor you should use in your gaming machiene, it was about the processors that may show up in your car, kitchen, etc.
      • Yes. The point of my post was that most of us never see outside of Motorola/AMD/Intel because we only use desktop machines and don't think about what's running our microwaves.

  • Top 10 Lies about Microprocessors (Score:5, Funny)

    by AtariAmarok ( 451306 ) on Sunday July 13, 2003 @01:16PM (#6428970)
    10. Tin foil helmets will protect us from rays and mind-control particles from microprocessors.

    9. Intel from Mars, PPC from Venus

    8. No, Porky Pig did not give the PPC its name when he tried to say "PC".

    7. Celeron was not named after Celery

    6. Go ahead, you can buy a Pentium 3 without worrying that the Blue Man Group will knock on your door and bore you to tears with their post-modern Bolian-hued Mummenschanz [mummenschanz.com] antics.

    5. "It's a chip, does this mean I can eat it if I dip it in bean dip?"

    4. "I paid $2,000 for this screamer back in 1987. It will blow the socks off anything you will put up against it"

    3. "Mine's bigger than yours"

    2. "Intel Inside"? Consider that label to be a warning.

    1. "Get me a microscope. I'm going to open up my PC and look at my micro-processor.

  • Heres my top ten (Score:5, Funny)

    by Froze ( 398171 ) on Sunday July 13, 2003 @01:22PM (#6429001)
    1. I don't want the latest processor
    2. I am perfectly happy with the one I have.
    3. Having a faster CPU would make me more productive.
    4. I bought the cheapest one because I want to support the underdog.
    5. Noisy CPU fans don't really bother me.
    6. If I get the most expensive one, I won't have to upgrade for a long time.
    7. My life is so much better now that I have CPU x.
    8. I don't envy you your brand new CPU, because mine has a years long proven track record.
    9. Nobody will ever need a 64bit CPU for home use.
    10. I only read newsgroups for the articles.

  • Huh? (Score:1)

    by Otter ( 3800 )
    #7: Price is proportional to performance
    Microprocessors are now sold like perfume: the price on the label has no connection to the cost of the ingredients. It's tempting to assume some meaningful relationship between cost and price. Save your time--there isn't one.

    I'm trying to find a meaningful relationship between the topic and that line of argument but don't think there is one.

    • Re:Huh? (Score:3, Informative)

      by ctr2sprt ( 574731 )
      There is one. The argument he's debunking:

      If performance is increased, the cost will increase; if the cost is increased, the price will increase. Therefore, a higher-priced CPU delivers higher performance.

      The problem with that argument:

      Price and cost are unconnected: the price will move up or down regardless of the cost.

      It's not terribly clear because he omitted a step from the argument, but I suppose otherwise the title for that section would be too long and unwieldy.

  • Missed the biggest one of all... (Score:5, Insightful)

    by smoondog ( 85133 ) on Sunday July 13, 2003 @01:26PM (#6429023)
    #3: Instruction sets don't matter

    I would say:

    #3. The clock speed is a good estimate for processor performance

    They implied it here, but even in this world today, there are competent people that think clock speed actually matters when comparing one processor over another. I had an IT person who controls a pretty big budget actually compare a processor in a 8way Sun server to a 21264 alpha chip using only the magnitude of the clock speed as the only performance benchmark. As most (should) know, clock speed works for ranking processors within a family, but mean very, very little in the real world. It's obvious, but as long as purchasers think this is true bad decisions are being made...

    -Sean
    • I used to work for a company that had two plants. One of them bought DEC Alphas solely because the CPU clock speed was higher so it must mean that they can outperform Sun machines. It didn't matter to them that Alphas are not as supported in terms of software selection like Sun, HP, or IBM. All they cared about was the clock speed. This plant also decided that VB5 was the best programming language to use on the back-end and that IT development should be handled by each division (Engineering, HR, etc) in
      • It was ironic that though we had as many staff as they did, they had 20x the number of contractor and consultants and took them years to do what we could do in hours.
        How is that ironic? Sounds pretty logical to me, given the conditions you described.
        • It was ironic that upper management (based at the other plant) never noticed and considered our IT department inferior. Later on, our IT director was promoted to be in charge of both plants, he was able to slash the budget (by getting rid of the consulants and contractors) and get more productivity.
    • Once processor speed was useful. If you had a 6 MHZ 80286 and the other guy had an 8Mhz 286, his computer was faster. Now when you get into cacheing, DDR vs RD ,and now hyper-threading it is all much more omplicated. Benchmarks need to look a to IO, Graphics, FP, and vector ops. The good old Byte Seive just does not cut it anymore.

      Embeded systems ofter are much simpler.
      When it comes to embeded systems you need a chip that is fast enough. You are not worred that you will have to run Doom3 on your latest wi
    • ... but even in this world today, there are competent people that think clock speed actually matters when comparing one processor over another.

      Of course there are: Intel think it matters, and AMD think it doesn't. ;-)

  • Interesting, but one nitpick (Score:4, Informative)

    by LizardKing ( 5245 ) on Sunday July 13, 2003 @02:17PM (#6429244)
    Definitely an interesting article, and I'm even inclined to read some of the authors stuff. However, he might need to clarify that bit about the Dhrystone benchmark being thirty or more years old. If it is, then it wasn't written to allow comparison against the performance of the Vax as he claims. The first Vax systems didn't ship until the late 1970's, a little over twenty years ago.

    Chris

  • Good read (Score:4, Insightful)

    by Anonymous Coward on Sunday July 13, 2003 @03:12PM (#6429489)
    Best quote from the article: There aren't many strong brand reputations in the microprocessor business but ARM enjoys one of the best. According to their reputation, ARM's chips are endowed with an almost magical ability to run on bright sunlight or the energy released by rubbing a cat. An ARM processor, two lemons, and some copper wire are all that's needed to build the latest PDA, it seems.

    Some of Slashdot's trolls would do well to pay attention to the cynical wit present in that statement. Overall though it's a very well-written, concise, and informative article. I'll be quoting his statement on MIPS next time a cpu discussion comes up.
    • I may be blind, and probably get modded down for this, but how is this trolling? It seems he's relaying an anecdote of sorts.
    • It's not two lemons and a piece of copper wire. If he's talking about creating a fruit-based battery [rr.com], he needs:

      * One lemon (or any compatible citrus fruit)

      * A chunk of copper

      * A chunk of zinc

      If I remember my daughter's 4th grade science experiment correctly, I'll need about a half dozen in series to run my Palm III.

      Of course, I wondered how the lemon (and the orange, and the potato) would *taste* after the electrical potential had been depleted... and what would happen if I tried to recharge the "batt

  • Not all Micrprocessor are reliable as each other (Score:5, Insightful)

    by plusser ( 685253 ) on Sunday July 13, 2003 @03:38PM (#6429603)
    Take a 486 processor and the latest pentium 4 and then run them for 4 or 5 years. Due to the way that modern processors are manufactured as the die features are much smaller, then there is a high probability that the Pentium 4 has now stopped working:- The 486 is still likely to be going strong.

    The problem is this, it is fine having the latest technology fitted in the a piece of equipment. However, if you expect that piece of equipment to be operational for an extended number of years, don't select the most powerful processor avialable. Fit one that is reliable and is likely to still be in production for the foreseable future.

    Ahh, some of you would say, surely you can emulate an old microprocessor in the future if you need to. Good premise, but if you are building for an application that is incorporated into a safety device on an aircraft or even a car, you will spend an absolute fortune re-qualifing the replacement software. An if you are only building a a few units for spares, it is one easy way of going bunkrupt.

    and this is before I start talking about whether the microprocessor will operate in the temperature, vibration environment and the effects of atmospheric radiation at altitude.

    Working outside the PC industry, there is a lot more to consider than you would think....

    Plusser

  • Great article (Score:2, Interesting)

    by Mensa Babe ( 675349 )
    It could be summarized as: "Myth #1: IA32 is good." All in all, a great article. Speaking about myself I might add that for me MMIX is the perfect processor (at least the most perfect designed so far) and I always look at every CPU architecture with the most important factor being how close it is to MMIX in certain aspects. I wonder what is the perfect processor for other people here on Slashdot. I bet for most of you it is Alpha, but I may be wrong.

  • -1, Troll (Score:4, Insightful)

    by Salamander ( 33735 ) <jeff.pl@atyp@us> on Sunday July 13, 2003 @05:16PM (#6430084) Homepage Journal

    The author seems to enjoy dispelling myths but, in this and one of his other articles (RISCy Business) that he links to, he seems to cling rather stubbornly to a couple of his own. For example, I just read twice about how "some RISC chips" don't have multiplication instructions, even though most do nowadays. But hey, it's a convenient club to bash with, just like the one about code density. The guy's a troll. He hates RISC for some reason, and perhaps he has some good points to justify that dislike in the embedded space, but in his zeal he just goes too far into exaggeration and misrepresentation. He also needs to read H&P to understand the real rationale behind RISC, instead of the strawman rationale he gives in the article.

    • Re:-1, Troll (Score:2, Insightful)

      by turgid ( 580780 )
      Quite right. The choice of the processor should be appropriate to the job in hand. Do you need high code density? Maybe go with CISC. Do you need to execute many simple instructions quickly? Go with RISC. Do you need low power? Choose something simple with a lowish clock frequency. Do you need fast floating-point? Maybe choose a DSP. Do you need to be able to program in a high-level language? Choose one with a good cross-compiler. As with everything else in life, there are no absolutes.
    • Bah - it's not really RISC unless it implements only one instruction [hawaga.org.uk]! I pity the compiler authors for OISCs, though.
    • Your own words:

      I just read twice about how "some RISC chips" don't have multiplication instructions, even though most do nowadays

      Well, it might just be me but when "most do", some don't. You're basically saying that he's right, and should be modded down for it?
      I'll grant that there might be a negative slant towards RISC in this article, but trolling? Nah.

  • Re: (Score:2, Funny)

    by account_deleted ( 4530225 )
    Comment removed based on user account deletion

  • ARM is the middle of the pack in efficiency? I didnt know that. I'm basing my designs on ARM for the reason of its power consumption/performance and its die size (core is 3mm^2 !!). That was the reason I bypassed PPC MIPS and the rest, because theyve always been big powerful chips not suitable for an mp3 player running on 2 AA batteries, but of course I never had the time or motivation to pull out datasheets and compare the numbers for these architectures.

    Does anyone know of the lowest power consuming 32

  • But This Goes Up To 11 (Score:2, Interesting)

    by Uosdwis ( 553687 )
    This was an interesting article. I enjoyed the 'There is no one answer' stance the article took. Being a mac user I know all about the holy wars and punishment due. Being a developer in the Aerospace embedded market processor choice is very important.
    This goes to show you what (even educated) people think because of TV/marketing. The CPU is the 'brains' & mark of a computer, Intel chips are the fastest, .. etc. I tell most people that there calculators have more computing power than the Lunar Lan

  • Thermodynamics (Score:3, Insightful)

    by bobbozzo ( 622815 ) on Monday July 14, 2003 @01:35PM (#6435833)
    From the article:
    the wee ARM6 consumed less total energy than the others gave off as heat.

    -1 Redundant

    • How is that redundant? Or that comment insightful? Sounds more like you didn't really understand what he was saying.
      Let's say, an ARM processor consumes x Watts.
      Others consume y Watts.
      Now, a significant percent of the consumed energy will be dissipated as heat. In the case of others, it would be in the order of (0.4 y) to (0.5 y) (I'm being conservative).
      What the statement says, is that x (0.4 y). Which, of course, means x y.
      • Now, a significant percent of the consumed energy will be dissipated as heat.

        Sorry, all consumed energy will become heat (and will be radiated as heat).

        The wattage consumption of a solid-state device is equal to its heat output (or heat+light output, in the case of an LED or lightbulb or CRT, but light becomes heat too).

        Of course, you may have some of the electricity going to memory writes or something, but eventually, it all becomes heat.

        10 150W lightbulbs (in a closed room with no windows for the

Slashdot Top Deals

The 11 is for people with the pride of a 10 and the pocketbook of an 8. -- R.B. Greenberg [referring to PDPs?]

Close