Please create an account to participate in the Slashdot moderation system


Forgot your password?
Intel News

Intel's Per-Chip Cost Averages $40 423

Fedorpheux writes "According to a report by the analysts at In-Stat, Intel's average cost per chip is about $40. These same chips, such as the Pentium 4s, can cost consumers up to $637. This $40 average cost has remained rather steady since 2003. This cost does not include money spent on marketing or development, but it does explain how Intel can continue its profits even in this era of quickly dropping prices in computer hardware."
This discussion has been archived. No new comments can be posted.

Intel's Per-Chip Cost Averages $40

Comments Filter:
  • Grr (Score:2, Funny)

    by serenarae ( 154753 )
    I wish they'd pass these low costs on to us peons who pay out the ass for their processors.

    Wishful thinking.
    • Re:Grr (Score:4, Insightful)

      by DrEldarion ( 114072 ) <[moc.liamg] [ta] [0791uhcsm]> on Wednesday September 14, 2005 @10:49PM (#13563210)
      Considering that a HUGE amount of money gets put into research and making/updating the facilities to manufacture these processors, the actual costs are much higher.

      Not to mention they have to pay salaries, benefits, etc...
      • Re:Grr (Score:5, Insightful)

        by k2enemy ( 555744 ) on Wednesday September 14, 2005 @11:02PM (#13563308)
        additionally, the cost quoted is an average per chip cost. i'm assuming this isn't limited to pentium 4 chips, but includes all chips that intel makes.

        they contrast this $40 average cost with the consumer price of a p4, which is probably one of the most expensive chips. the average cost of making a p4 is probably much higher than $40.
        • Re:Grr (Score:3, Insightful)

          by b0r1s ( 170449 )
          Exactly. This post ignores salaries, R&D, and the fact that many (most?) of the chips Intel makes are NOT Pentium class chips, but rather, Cell phone and embedded processors.

          Hype article, no real news value.
        • Re:Grr (Score:5, Informative)

          by pla ( 258480 ) on Thursday September 15, 2005 @07:51AM (#13565194) Journal
          i'm assuming this isn't limited to pentium 4 chips, but includes all chips that intel makes.

          I don't think you fully appreciate the size of Intel, or the sheer number of chips they make...

          P4s count for less than 1% of their sales, in terms of volume. CPUs in general almost certainly make up less than 10%.

          The vast majority of Intel's output consists of things like opamps, ethernet controllers, simple logic chips, and other trivial (compared to a modern CPU) ICs that mostly cost well under a dollar (to buy, not to make) each.

          No, their average cost per chip, over their entire product line, does not come out to $40. Not even close. That would bankrupt them in a week, selling X chips at $600 while selling 95X at $1.

          That said, the $40 figure certainly does not take the total cost into consideration. Perhaps the raw materials, electricity, and immediate labor to produce them once everything has fallen into place. But just the cost of building a new fab (in the low billions), or retooling an old one for a new process (hundreds of millions) far outweighs the ongoing per-unit production costs.
          • Re:Grr (Score:5, Informative)

            by networkBoy ( 774728 ) on Thursday September 15, 2005 @09:50AM (#13565899) Journal
                They don't make simple logic chips, nor do they make op-amps.
            Intel is not in the business of making discreet components.

            If you anre interested the Adjusted gross margin of the company (based on previous SEC filings) is roughly 48-52%. That would indicate that a $600 wholesale chip cost the company about $300 to produce, with the other $300 going to expansion, investment in new tech, shareholders dividends, into a bank account for a rainy day, etc.

            I would tend to think that most of the profit is mature tech for which the R&D has been ammortized, such as the P3 chips, Xscale, etc.

            I have a fairly reliable way of thinking that the P4 division, while a profit center, is not where the big money comes from. Most of that $400-$600 you spend on a CPU is covering other people's costs. (remember retailer markup)

      • Re:Grr (Score:4, Insightful)

        by bhirsch ( 785803 ) on Wednesday September 14, 2005 @11:05PM (#13563321) Homepage
        In other words, from the accounting standpoint, the variable product costs are $40/unit. There are also fixed product costs, period costs, overhead, etc. Intel certainly has profit margins, but they are sure as hell nowhere near what the blurb or article would lead one to think.

        I would actually think that their margins are lower than AMD's, though their profits would of course be higher due to shear to volume.
        • Re:Grr (Score:3, Interesting)

          by multimed ( 189254 )
          Not to mention the costs of employee healthcare. This is by no means meant to be a boo-hoo for them or others but if $1500 of the cost for every car GM makes is employee/former employee healthcare, it's pretty reasonable that those costs contribute a significant amount to what it really costs Intel to make their products as well.
      • by EmbeddedJanitor ( 597831 ) on Wednesday September 14, 2005 @11:50PM (#13563548)
        Any analysis of fab costs is just pointless. The costs of CPUs are largely development costs.

        $40 is actually a hell of a lot for a chip. That explains why x86 really is not going to become a contender in low cost devices. OMAP parts etc cost sub-$20 to the customers.

      • Re:Grr (Score:4, Insightful)

        by dvnelson72 ( 595066 ) on Wednesday September 14, 2005 @11:53PM (#13563563)
        You stinking capitalist pig! How dare you point out the obvious.

        I can't believe how people are so quick to react to this with some neo-socialistic view that the capitalist corporate scum are somehow raping us.

        What really boggles my mind is that slashdot is, supposedly, a computer geek oriented user base. Software developers should understand implicitly how much money is spent trying to develop products before one is actually profitable.

        Maybe it is the Seinfeld-Kramer idea of "write-offs". Could everyone think that all losses just get written-off into the ether and never affect the bottom line? The fact is the world, hell the universe, is governed by net calculations, not gross.
      • Re:Grr (Score:5, Interesting)

        by _ph1ux_ ( 216706 ) on Thursday September 15, 2005 @02:50AM (#13564211)
        when I worked at intel I was told that the per chip cost was $48.00 - this was in '99..
      • Everyone is talking about other costs - and they're right. This metric is next-to-useless, and extremely difficult to analyse.

        However, even without the extra costs - it's a free market. This means that the company can charge what they like. If they are not a monopoly (and intel may have tried their best - but at least there's some competition now) - then they charge what people will pay, if it's easy to enter the market (and I know it's not), then someone will and outdo them.

        That's the beauty of a monopol
    • Re:Grr (Score:4, Insightful)

      by ErikZ ( 55491 ) on Wednesday September 14, 2005 @11:15PM (#13563381)

      Just curious, what do you think they do with all that profit?

      R&D and building costs for new, cutting edge fabs go past a billion dollars. Do you think their new 64nm fabs come from the magic wishing fairy?

      They're going to have to figure out how to make 64nm chips. This is hardcore applied physics to make the most advanced CPUs in the history of mankind.

      Where do they get the money to pull this off? From us. We pay them.
    • They do, you just have to wait. Go take a look on the net, you can find P4s with respectable amounts of power for $40-$60, they're just a generation or two old.

      If you're paying out the ass for your processor then you're falling into the latest and greatest trap. Intel banks on these people the same way that the graphics industry does. They're going to sell their processors for the highest price that they think consumers will pay, and they'll cycle them down to continue the perception that the next process

  • So what are we paying for with that $600 or so above the cost price? Are marketroids so expensive?
    • Research and development. Chip fabs. Engineers.

      Copying something once it's made is a lot cheaper than figuring out how to make it in the first place.

      • by lullabud ( 679893 ) on Thursday September 15, 2005 @12:00AM (#13563609) Homepage
        You left out that one time I left the pod on the nanoscope while I went to check the SVG tracks and the arm failed to retract before proceeding from wafer #1 to wafer #25, breaking the lot before I could get back to hit the emergency stop button. Sucks too, that was right after a fresh coat of polyimide, two or three steps before the end of the line. Must've been $150k worth of product in that pod. So, yeah, some of that money just goes right into the trash, or becomes a souvenir refrigerator magnet.. shhhh... ;-)
    • The first couple cpu's of a line cost millions and millions to make, they need to charge for the money lost on the first few.
    • How about the cost to actually *design* the chip.

      Or do you think that Intel should foot the multi-million-dollar bill out of the goodness of their hearts?
    • The actual manufacturing of most electronics products is very cheap. The costs stated don't cover R&D or advertising.

      They said "average chip cost", I imagine P4s cost a lot more to make than flash memory or the chipsets that Intel makes, as well as the miriad other chips, such as microcontrollers.
    • Remember that the $40 is an average. The $600 dollar chip might cost $350 to manufacture, while several other chips cost $10 to make, bringing down the average.

      I remember this particular fable from a book about mathematics. Imagine a person at a job interview:

      Prospective employee: What kind of salary might I expect if I were to work here?
      Owner: The average salary here is $85,000.
      PE: Sir, I will accept your offer for employment.

      Then, two weeks later:

      Current employee: I have a problem with my paycheck.

    • Look at Intel's profit margin, it's below 25%. []. There are lots of expenses in running a company. Only the fastest P4s are in the $600+ region, many are below $100 retail. []. The cheap ones cost every bit as much to make as the expensive ones; it's just a matter of supply and demand.
    • It's more than marketroids, although they do seem to excede the scientists in cost. If you look at the last reported quarter, Intel grossed $9.2 billion of which $2b was profit. Of note $1.17b went into R&D (1.34b into "selling and adminstrative"). That's still a decent profit -- about 22%, but it isn't like they profit $540 on every $40 of expenses (1350%). Intel Balance Sheet []

      Compare that to AMD's rather ugly results []. Only $11 million in profit (and two of the last four quarters were losses).
  • With tech... (Score:4, Insightful)

    by Anonymous Coward on Wednesday September 14, 2005 @10:46PM (#13563180)
    ...R&D costs will almost ALWAYS top manufacturing costs...
  • by ScrewMaster ( 602015 ) on Wednesday September 14, 2005 @10:46PM (#13563183)
    1. Profit!
    • Back in the PLM/80 PLM/86 days, when the assembly instruction 'DI' was replaced by the PLM 'Disable$Interrupts' or 'INT 10' was 'Cause$Interrupt(10)', we had a saying that all they really cared about was 'Remove$From$Customer'.
  • $637? (Score:2, Redundant)

    by ravenspear ( 756059 )
    I don't think I have ever seen a P4 for that much. Looking at pricewatch [], it looks like almost all variety of P4s can be had for under $300.
    • Re:$637? (Score:5, Informative)

      by Feyr ( 449684 ) on Wednesday September 14, 2005 @11:00PM (#13563298) Journal
      $719 - xeon 3.6ghz 604

      straight from your site. and while it doesn't say, it's likely that you can find higher end xeons (with gobs of cache) for a few grands
      • And Sun is selling an entire Opteron server for only $745.*

        * Note that retail OEM CPU price can be radically different than retail, and for all we know AMD may be giving away the chips to Sun for free.
    • Re:$637? (Score:5, Insightful)

      by drmerope ( 771119 ) on Wednesday September 14, 2005 @11:59PM (#13563606)
      The thing is this comparison is extremely misleading.

      Look, they averaged the costs overall production. They excluded development.

      As a engineer ing this field let me tell you development costs are *huge*.

      Moreover, some processors might cost $637 but those process cost a hell of a lot more than the average to manufacture... and that is the price you pay for the processors that come off the line with performance in the second or third standard deviation from the mean. There are not many of those processors (hard to make) + lots of demand => no shortages require high prices. The point being those those applications that can justify the cost are the ones that get the chip. This is about not wasting those chips on grandma's email computer while some scientist needs them--and to make that allocation in keeping with liberalism, i.e., without coercing people + corruption.

      Anyone who was moved by this article should read
      "Economic Calculation In The Socialist Commonwealth" []
  • by bloggins02 ( 468782 ) on Wednesday September 14, 2005 @10:47PM (#13563197)
    does not include costs for marketing and development

    Which, given that a product's true cost includes not only the per-widget cost to make the item, but also the amoritize costs of slaries & benefits, facilities used in production, third party contracts, marketing and advertising and probably a lot more that I'm too tired to think of right now, makes this number pretty useless, no?
    • The chip-making facilities themselves are not R&D, so they should be included in the $40 figure (the sheer cost of constructing the plants is considerable). Salaries and benefits too (not for R&D people but production people). This is not a completely useless number (if it is accurate). Of course, the only *really* useful number is the price you have to pay to get one, but it's interesting to see the actual costs of production and compare.
      • The chip-making facilities themselves are not R&D, so they should be included in the $40 figure (the sheer cost of constructing the plants is considerable). Salaries and benefits too (not for R&D people but production people).

        The article is not very clear about this and my gut reaction is that the $40 only includes the operational costs. Fab plants are UBER expensive (measured in the billions of dollars), by far the largest expense of the total cost of making a chip. Even assuming that all of the
        • Actually, I'm about an order of magnitude off.. it would be around 30-70 million chips until break-even point. Still a butt load though.
        • Well not only that, but according to the Inquirer [], it's cost per die, not cost per chip. That means that you have to add the cost of all the failed chips. You can only sell the ones that work, and yield is never that high for processors using the latest process. Also, the die price may not include packaging the chip (i.e. adding all the pins, heat spreader, etc. Seems cheap but it actually adds a noticable amount to the overall cost.

          Finally, average is just a stupid metric to begin with. Intel probably
    • by Phat_Tony ( 661117 ) on Wednesday September 14, 2005 @11:56PM (#13563584)
      I agree.

      In other news, an analyst has determined that Adobe's cost per copy to manufacture Adobe Photoshop is about $0.35. This leaves out all research and development costs, such as writing code.

      What are they including in this $40 cost, the price for 1/10 of an oz of silicon? If you don't want to include developing x-ray laser lithography or designing the circuit layout for 55 million transistors, I'm sure they are cheap to pump out.

      There are lots of companies for whom the marginal cost to them of providing an additional unit is often near 0 - software companies, airlines, informational databases, etc. For such companies, knowing their actual marginal cost of production doesn't give you much useful information.

      It should be a dead giveaway that the marginal cost of production on a processor is very low, because the same processor always costs a fortune when it's introduced, and then is sold new for a small fraction if its original price a few years later, just before they pull it from the market. Clearly, they wouldn't lower the prices that much if they were losing money on each one.

      Merely building a new chip fab represents a significant amount of money compared to the aggregate marginal manufacturing costs of every chip to come out of the entire plant for it's whole life. The semiconductor industry runs on huge R&D costs and small individual unit manufacturing costs. Pointing that out isn't really news.

  • I'm not sure why this surprises anyone. Intel and Microsoft make hefty profits on the products they sell. People can understand the software side of things a bit more easily than the hardware, I guess.

    Sure, there's a lot of money poured into product development for Intel, but at the mark-ups they've had it wouldn't take that long to recoup the costs.

    I imagine AMD is doing the exact same thing.
    • Well, software is a different animal than hardware. For the same product, hardware decreases in price over time. For example, when I purchased my laptop about 3 years ago, a 512MB DDR266 SODIMM stick cost over $250. Now a pair will get you $10 back from your Benjamin. However, Windows XP still costs the same $299 for a full Professional installation CD that it did in 2001 when it launched. And Windows 2000, which the OfficeMax near me still stocks, costs $259. Software is IP while hardware is mostly "nuts
  • What??? (Score:4, Insightful)

    by Eightyford ( 893696 ) on Wednesday September 14, 2005 @10:49PM (#13563208) Homepage
    Do you have any idea how much these manufacturing facilities cost? Here's something else you might not know: it doesn't cost apple 300 bucks to make an iPod either! Gasp!
    • Ya, everyone seems to forget that a chip factory, which is good for about 2 years before needing to be retooled, costs upwards of $1,000,000,000. Yes, one billion dollars. Look it up. Thats where your $600 goes, not to mention the huge R&D and advertising departments, and all of the umpa lumpas that actually make the chips. Plus, not every chip Intel sells is $600. I bought a brand new Pentium D 820 about 3 months ago for a little over $300, and it was pretty darn near top of the line. Look on New
  • Trivial Overhead? (Score:5, Insightful)

    by goldspider ( 445116 ) <> on Wednesday September 14, 2005 @10:49PM (#13563214) Homepage
    "This cost does not include money spent on marketing or development"

    Yeah, that would have been too... Honest? Thorough?

    So what's the per-chip cost WITH all of the overhead?
    • Re:Trivial Overhead? (Score:5, Informative)

      by shark72 ( 702619 ) on Thursday September 15, 2005 @07:45AM (#13565167)

      "So what's the per-chip cost WITH all of the overhead?"

      No need to ask. We can deduce this using some basic Internet research skills and some junior high-level math.

      If they amortize overhead equally across all products, you can guesstimate it this way (we'll use a part that costs $600 at retail for an example):

      1. Take the price you pay at retail and subtract the margin the retailer makes. For example, if Fry's makes 20%, multiply $600 by 0.8 to get $480.
      2. Then, subtract the margin made by the distributor (assuming Intel uses two-tier distribution for chips). Distis typically take 5%, so multiply that $480 by 0.95 to get $456.
      3. Intel's profit margin last year was 22.45% []. For each product they sold, they made an average of 22.45%. Again, assuming that Intel amortizes overhead equally across all products, multiplying $456 by 0.7755 = $353.62.

      So, for that $600 part you buy:

      1. Intel makes (nets) $102.
      2. The distributor makes $24 (gross -- before similar overhead has been applied).
      3. The retailer makes $120 (again, gross, not net.)
  • Does that $40 per chip include fixed costs or development? I'd suspect it's rather pricy to design a microprocessor and set up for producing it.

    (I don't work for Intel or anything. They probably are pulling a huge profit margin anyway, but I do suspect that this is more /. sensationalism.)
  • It probably doesn't count the cost to build the fab, either. The last time I checked, building a sub-micron fab cost roughly eleventy billion dollars....
  • by wigle ( 676212 )
    The 1000% profit from higher-end server and Xtreem gaming CPUs go toward research and development. This is good because eventually those CPUs will be priced at the consumer level. I don't see how else Intel could afford to keep developing new architectures.
    • I don't see how else Intel could afford to keep developing new architectures.

      Maybe Intel can, but sadly they aren't. AMD is responsible for the vast majority of x86 innovation over the last 4 years: high IPC cores, larger L1 caches, on-die memory controllers, high performance serial chip-to-chip interconnect, 64 bit extensions with more general purpose registers, no-exec page protection, etc.

      Thank AMD for fixing x86.
  • This cost does not include money spent on marketing or development

    Yea... way to leave out the most expensive part. You think designing a microprocessor is cheap?
  • "This cost does not include money spent on marketing or development"

    And guess what development is the biggest expense in making chips. That statistic is pretty meaningless if used to determine whether intel charges fair prices for their chips.
  • by fdawg ( 22521 )
    It costs MS fractions of a penny per cd (or dvd, i dont even know, im an FOSS slut). Does that mean this is where all they're profits are from?

    R&D for ANY company are astronomical. My lab designed radios that cost around a grand to manufacture. The R&D costs were being measured in hundreds of thousands, and thats still cheap.
  • Calculating the chip cost this way but leaving out "marketing and development" is like saying that your car's cost is simply the cost of the fuel to run the engine, about 15 cents per mile. Neglecting the fact that you had to spend many thousands of dollars to buy the car in the first place, and continue to spend hundreds of dollars to maintain it, et cetera. Do we have to endure trolling on the front page posts now?
  • by leehwtsohg ( 618675 ) on Wednesday September 14, 2005 @10:55PM (#13563257)
    Dropping R&D and marketing, you'll get for microsoft:

    price of CD: ~1$
    price of office/windows XP: 340$/170$

    profit: lotsa %!

  • by Doc Ruby ( 173196 )
    What kind of gibberish is that? The average currency in my pocket is worth $15.37. The value of the currency in my pocket is as high as $100. How come my currency is undervalued? Who writes this crap, anyway?
  • by ScentCone ( 795499 ) on Wednesday September 14, 2005 @11:03PM (#13563309)
    That's pretty much a ridiculous way to describe things. Saying that it costs X to produce something, but ignoring the actual overhead is completely sophomoric, and an obvious attempt to pander to the corporations=bad and profit=bad crowds (never mind that only a large, profitable entity could possibly produce things like Xeon or AMD-64 chips and keep coming up with and delivering more, better, faster). It's like the people who think that they only cost their employer what they see on their pay stub. There's a little more to it, folks!
  • by G4from128k ( 686170 ) on Wednesday September 14, 2005 @11:03PM (#13563312)
    Intel's recurring costs are irrelevant to its business model. The bulk of the cost is in R&D and the fabrication plants. R&D at Intel is about $5 billion per year and the company has almost $16 billion in plant and equipment. Worse, Intel's fabs aren't really a long-term assets in the traditional sense. Unlike most manufacturing companies, Intel's plant and equipment goes obsolete on a time scale not that different from the chips. An old 130 nm fab or one using the old 8" wafers is increasingly obsolete. Even today, Intel is looking to replace its 90 nm fabs wiht 65 nm fabs in 2006 and 45 nm fabs in 2007. And at $1 to $3 billion for each new new fab, the money comes from chips.

    The only way to pay for all this expensive equipment and R&D that is obsolete with a few years is to maximize revenue on every fab line. In that regard, Intel is in the same boat as the pharmaceutical and airline companies -- low recurring costs but huge upfront investments.

    I'm not saying that Intel isn't hugely profitable only that the "cost" of a chip is much much higher than $40.
  • Misleading (Score:3, Insightful)

    by uimedic ( 615858 ) on Wednesday September 14, 2005 @11:07PM (#13563327)
    It's a bit like saying that it only costs $5 to make 30 pills of a prescription medication that they sell for $30. Sure it may cost that much to make the pills from raw materials. However, it takes a lot of money to invent the medication, test it, and then get it approved by the FDA. Those costs aren't represented in the amount it costs to mass-produce the medication itself.
  • Wow, no wonder some companies hate revolutions and advancements. You would think that tech. companies would always aim for the next step but now that you look at the numbers it's different. Once you invested all that money to R&D, you would want to just sit and make a ton of those high margin products. This may explain why Microsoft or other tech companies aren't always so keen on exploring new ideas.
  • Yeah, that is like ripping on computer game companies for the fact that the cost to produce a game CD is only $5 when they are charging $40-50 for the title. It completely ignores all development and production costs.

    I agree that Intel severely overcharges for their flagship processors, but I don't think these findings are going to support our argument at all.
  • This is little different from most other industries. A huge markup provides allowance for further research and design (and I'm not talking about computers here, either) for the manufacturing company. But mostly, its just easy profit.

    A typical iPod costs $200-400. How much do you think it costs to manufacture? Probably $20 or less. A Chanel handag? $60 to manufacture, but costs $3,500 to purchase at the retail level. The markup is absolutely ridiculous, but Intel is no different than Chanel or Apple and mo

  • Typically stupid (Score:4, Insightful)

    by n54 ( 807502 ) on Wednesday September 14, 2005 @11:12PM (#13563359) Homepage Journal
    "This cost does not include money spent on marketing or development"

    Which translates into: "Nothing to see here except a fine example of bullshit reporting which actually doesn't contain any useful information. Made up to get people riled up about something that isn't actually relevant while keeping any reader ignorant or paying homage to aleady delusional ideas on how things work out there in the real world".

  • by fermion ( 181285 ) on Wednesday September 14, 2005 @11:13PM (#13563373) Homepage Journal
    I am not defending intel, because clearly the make a handsome profit, which they must to compensate all the unproductive middle managers that are present in all large organizations. However, without RFTA, i can say the speculate the person who wrote the summary, and perhaps even the person who wrote the article, are probably without a clue.

    First, average cost does not really tell us what the cost of a particular chip is. Does one chip set cost $200, one cost $100, another cost $50, and all the legacy costs $10? I mean a low end computer can be had for a few hundred dollars, and the chip itself can be had for mere dollars in quantity, so the cost to produce has to be a few dollars. This would mean the top end chip might costs a few hundred, or more, to produce.

    Second, comparing an average to a maximum is about the most devious thing a person can do. Again, the top product might cost a few hundred dollars. The average offer taken price of a chip might be under a hundred dollars, again noticing that a computer can be had for a few hundred dollars.

    Finally is this number fixed, variable, or simple material cost? Does it take into account the higher rate of defects on new products, and higher risk of returns? Is this a number with any credibility whatsoever?

    This is what we do know. For the fourth quarter of last year Intel earned about 2 billion on sales of about 9 billion. That is about 20% profit. Because these are intel numbers we can assume the sales are inflated and the profit fudged. However, if even 10% of this revenue went to chip production, at $40 per chip we are looking at 90 million chips, give or take. Did they ship this many? Perhaps. And they did sell them for $80, would that leave any money to pay the fancy salaries and benifits that the average worker, quite greedily, expects.

  • by Have Blue ( 616 ) on Wednesday September 14, 2005 @11:18PM (#13563393) Homepage
    "If you ignore all the other things Intel spends money on, manufacturing a CPU only costs this much." This is the same fallacious argument as claiming that album CDs only cost a record label as much as a blank CDR does in a store- the final manufacturing cost is only a tiny portion of what has been spent to make the final manufacturing possible in the first place. The acquisition of the knowledge of where exactly to put the copper dust on the silicon wafer is what makes the difference between a cutting-edge microprocessor and a worthless sliver of rock; neglecting it is simply stupid.
  • "The second pill costs six cents; the first pill costs five hundred million dollars." -- Aaron Sorkin, The West Wing.

    I'm sure Intel will be happy to sell you the second processor that rolls off any given production line for $40, as long as you also buy the first one, which costs -- what would it be? -- about $200,000,000 or so?

  • From the freakin' post: "This cost does not include money spent on marketing or development, but it does explain how Intel can continue its profits even in this era of quickly dropping prices in computer hardware"

    Let me guess... The poster and article author both stayed at a Holiday Inn Express last night?? I mean, talk about a moronic statement.

  • Before you get all...

    Int3l is teh LamErz!
    Ripoff!!1one!! ... please understand that the little shacks they knock up to research and make the chips cost around a BILLION dollars each.

  • by osrevad ( 796763 ) on Wednesday September 14, 2005 @11:30PM (#13563460)
    Intel's AVERAGE cost per chip is about $40. These same chips, such as the Pentium 4s, can cost consumers UP TO $637.
    • These are not the same costs. The $40 average is the manufacturing cost. The "UP TO $637" represents the retail price to the consumer. An analysis of any manufacturing industry will yield similar markups (profit margin), although the percentage varies for every industry.
  • With margins like that, and considering Intel's involvment with Linux, I wonder when the company will attempt to start a coup against MS. ... Alas, probably when Windows upgrades don't involve upgrading CPUs as well.
  • The 1 billion fab they built here in Phoenix (Chandler actually) and the next one should be paid for by... erm... the government? The Red Cross? The Boy Scouts?

    This is exactly the same argument used with commercial software. OMFG THEY SELL IT FOR $300 A MILLION TIMES!!1! OMFG!!

    Besides, the chips cost whatever the market will bear. No more, no less. Get used to it - it's called "capitalism" and it seems to have worked until now.

    This type of 'article' with interesting 'statistics' only gets play here wi

  • by scdeimos ( 632778 ) on Thursday September 15, 2005 @12:03AM (#13563621)
    Intel doesn't just make Pentium 4 processors. The $40 dollar average is based on the annual manufacturing costs of their entire portolio.
  • Intel's Costs (Score:5, Informative)

    by Bender0x7D1 ( 536254 ) on Thursday September 15, 2005 @12:10AM (#13563656)
    Some numbers from their financial report []...

    For 2004, Intel had a net income of US$7.5 billion on revenue of US$34.2 billion.

    Overall tax rate expected for 2005: 31%
    (With 2004 earnings as a guide, taxes will be US$10.6 billion)

    Their expected R&D budget for 2005 is: US$5.2 billion

    Capital spending for 2005: US$4.9-5.3 billion

    Overall, Intel pays 31% of their revenue in taxes. 30% in Capital spending and R&D, which leaves 39%, or US$13.4 billion, to pay salaries, benefits, cost of fabrication (not including the facility itself), cover the cost of their bad chips/wafers, and sending some cash to their stockholders.
  • by rpdillon ( 715137 ) * on Thursday September 15, 2005 @12:15AM (#13563675) Homepage
    I have a friend who works in IBM's fabs for nVidia chips and he was explaining to me that when you buy a top end chip, you have to pay for all the failed chips produced in order to get a good one. In the case of the 6800, he mentioned numbers along the lines of 20% when the 6800 was new. Obviously, as the 90nm (or 120, I forget) matured, this number goes up, but even so, they have to offset 4 other failed chips for every chip they ship.

    This is probably not as bad for x86 chips, as they can just underclock less well fabed chips, but the point remains: at $40 a pop failure can get expensive fast. The article mentions that the $40 figure doesn't take this into is a fairly big omission, IMHO.

    Coupled with them ignoring other huge expenses like the entire cost of the design of the chip, $40 seems kind of high. I wonder if it takes into account the creation, operation and maintenance of the fab facilities. I get the feeling they are simply pricing the cost of raw materials here, and the article is skimpy on details about what exactly IS included.

    Take with a healthy dose of salt, I'd say.
  • Reall Cheap (Score:3, Insightful)

    by tsotha ( 720379 ) on Thursday September 15, 2005 @12:41AM (#13563786)
    So, Intel chips are cheap if you don't count most of their costs. Wow, that's some analysis. WalMart's costs are cheap too, if you don't count what they paid to the manufacturer. Drug companies make boatloads of money if you don't count development and marketing costs. This business stuff is pretty easy...
  • by gnetwerker ( 526997 ) on Thursday September 15, 2005 @03:46AM (#13564420) Journal
    This reminds me of a comment I heard early in my career at Intel, when the 387 (the original match co-processor, anyone remember those?) went on sale:

    "We make higher margins on those than the US Treasury does making dollar bills".

    The margin on today's chips is nowhere near that high.

    Seriously, though the comments about R&D and marketing costs are on track, but leave out an important one: for each new generation of chip, one or more entire fabs (manufacturing lines) need to be built. Lately this costs $2bn (yes, billion) or more. When the next chip process comes along, the whole plant is essentially thrown away (yes, in reality it gets used for down-rev chips, but the lifetime isn't long). The difference between the actual capital deprecitation of these and the real cost/lifetime is another "hidden" component of chip cost. This applies pretty much equally to anyone making cutting-edge chips, including AMD.

    One of the reasons AMD stayed so far behind for so long was that its chips, generally a generation behind Intel (in the 1990s) didn't generate enough profit to build these truly leading-edge fabs. The "treadmill" as it was known at Intel, ran too fast for them to catch up. When the market hiccuped in 2000, things changed. Before that was a truly fine time to own lots of Intel stock options.

    -- gnet

I've finally learned what "upward compatible" means. It means we get to keep all our old mistakes. -- Dennie van Tassel