Forgot your password?
typodupeerror

Intel's Sales Down, Current Gen of Products Weak 249

Posted by Zonk
from the ouch-indeed dept.
DoctorBit writes "According to an article in EETimes, Intel's processor sales dropped 52 percent this April as compared with April one year ago. Unit sales dropped 21 percent and prices dropped 40 percent. The article concludes with an industry analyst's assertion that 'Intel has obviously given up on making any money on their current generation of processors and has started a price war with AMD.' The San Jose Mercury News is reporting that Intel has just put several of its money-losing communications businesses up for sale and notes that 'it remains to be seen what Intel will do with its other money-losing businesses, Itanium microprocessors and flash memory chips.' The article quotes an industry analyst saying 'If you look at Intel today, it's hard to find a trace of the technology or the people that they spent more than $10 billion on.' Ouch."
This discussion has been archived. No new comments can be posted.

Intel's Sales Down, Current Gen of Products Weak

Comments Filter:
  • by Anonymous Coward on Tuesday June 06, 2006 @01:31PM (#15481238)
    When the 2 Core 2 Double Duo comes out.
    • SSIA.
    • by 0racle (667029)
      Why would it? A P4 or even a P3 is still more then enough for the majority of people. Unlike going from a 386->486->Pentium, when you upgrade from a P4 to whatever is current you get very little tangible results. That is the current problem, several generations ago they had chips that did what is needed well enough for some time. They have to enter a price war with AMD because AMD is the current favorite among people who are silly enough to upgrade every time there is a 5% increase in performance. Int
      • True, but the general CPU market cool-off doesn't account for Intel's erosion in market share. AMD now has 22% [informationweek.com] of the overall x86 market. That's still a strong lead for Intel, but on top of everything else, it sure doesn't help. If nothing else, significant AMD market share foiled Intel's plans to keep the party going by making everybody upgrade to Itanium.
  • by 13bPower (869223) on Tuesday June 06, 2006 @01:32PM (#15481245) Homepage Journal
    1: buy stock now 2: wait for conroe to come out 3: ????? 4: buy quad core amd
  • by cnettel (836611) on Tuesday June 06, 2006 @01:34PM (#15481275)
    It's not just that their current gen is weaking, they are actively hyping a product that's coming in a few months. That marketing tactic might cause some buyers to delay in their AMD purchases, but it will surely hurt Intel sales more.

    OTOH, if Conroe really performs well, we might actually see the first big step upwards in performance for any mainstream desktop CPU in the last year or so.

  • Marketshare (Score:5, Insightful)

    by 3p1ph4ny (835701) on Tuesday June 06, 2006 @01:34PM (#15481277) Homepage
    Although it's good that Intel has lost some of their marketshare, we'd better hope this Conroe thing turns out to be as good as promised. I hate to see AMD and Intel just switching places... competition is good for everyone.
    • That's like hoping Vista turns out as good as promised. Sure competition is good, but someone always has to be on the losing side... and it might as well be the side you don't like :-)
    • Re:Marketshare (Score:4, Insightful)

      by Borland (123542) on Tuesday June 06, 2006 @02:05PM (#15481564)
      I doubt that they will simply "switch places". Even if they do switch marketplace positions then AMD has a long way to go in production capacity. Heh, I am a fan and a stockholder (very minor) of AMD, but it will take years to pull ahead of Intel in absolute terms.

      Besides, useful competition simply means having someone nipping at your heels. One does not have to be neck and neck with a competitor to spur innovation.
  • by DAldredge (2353) <SlashdotEmail@GMail.Com> on Tuesday June 06, 2006 @01:35PM (#15481285) Journal
    Sells in April fell thru the floor because Intel stuffed the channel in Feb and March.

    Question - when did Intel hire SpongeBob and Patrick to start naming their processors?
  • Naming (Score:5, Insightful)

    by qw(name) (718245) on Tuesday June 06, 2006 @01:35PM (#15481287) Journal

    Their naming convention needs to be more user friendly. The average consumer has no idea what a Pentium D processor is but they could understand that a Pentium 4 was better than a Pentium III. It's all about marketing to the masses in this over-saturated market.

    • I say they go with an Apple-like naming scheme. The next processor will be the the Pentium 10 (marketed as Pentium X). After that will be the more popular Pentium XX and of course, the AMD-killer Pentium XXX.
    • I think they've already realized this and acted appropriately. Intel's new naming scheme seems to be Core Solo/Duo (or Centrino Duo) for the current Yonah chips and Core 2 Solo/Duo for the new Conroe chips. Pentium M is obsolete and Pentium D will be soon enough.
    • by PCM2 (4486) on Tuesday June 06, 2006 @02:09PM (#15481599) Homepage
      Their naming convention needs to be more user friendly. The average consumer has no idea what a Pentium D processor is but they could understand that a Pentium 4 was better than a Pentium III. It's all about marketing to the masses in this over-saturated market.

      No, the real problem is that Intel is accustomed to a market where you sell specific processor brands direct to consumers, which is crazy. In a world where a $500 PC is going to be able to do 99.9 percent of anything that the average home user wants to do, that home user doesn't give a rat's ass what kind of processor is in there. Pentium M, Celeron M, Core 38 Double-D ... who cares? If their friend tells them AMD is a little better, then fine, they're going to buy AMD. Either way, all the same. All the shiny Intel Inside stickers in the world aren't going to make a bit of difference.

      People see an Intel commercial on TV and they tune it out. A guy in a weird space suit? OK, whatever.

      Gamers still care what kind of chip is in there, but gamers also have the option of consoles. Plus, the CPU matters a lot but the video card is the really sexy component for them. On the other hand, people who run servers might care about CPUs, but those kind of people are going to want to see real-world benchmarks.

      Intel needs to get over it ... and it is getting over it. Notice how Apple Macs all have Intel chips in them now. Dumb luck? I doubt it. Intel made the kind of deal it needs to keep making to stay on target, deals that are based on a simple old-fashioned idea: You're a component manufacturer. Sell components to manufacturers of consumer goods and let the consumer-goods manufacturers do the selling to consumers. If your product delivers the performance the manufacturer needs with a good integrated suite of products around it (e.g. chipsets, drivers, compilers) at a price point that the manufacturer can afford, then the manufacturer will buy your components.

      In a way, the last thing Intel needs is semi-informed consumers starting flamewars over this component vs. that component, Brand X versus Brand Y. A lot of the engineering decisions that get made in the CPU world aren't things that can be easily explained to consumers, so what you end up with is a bunch of FUD and name-calling. Intel's better off receding into the background and letting its engineering do the talking (if it's still got it).

      • ask any car salesman. in america, horsepower sells. sure a 75HP Geo would get Mr. Guy to work just fine, but you see a lot more 340HP Chrysler 300C's on the road than your theory would account for.
        • ask any car salesman. in america, horsepower sells.
          So why have all the chip manufacturers quit citing performance specs and begun renaming their product lines around meaningless numbers, rather than processor speed? Not just Intel is doing this.
          • When they were trying to sell chips with a lower clock speed as equivalent to Intel chips with higher clock speeds. So, for instance, AMD claimed the 2800 was as fast as a Pentium running at 2.8 GHz. (Whether the claims were true or not is a different issue). Now Intel has a problem, because the clock speeds are less important on a dual-core chip. Or rather than less important, say not equivalent to a single-core chip.

            If there were a way to claim duos had processor speed x 2 as their clock speed, Intel woul
      • In a world where a $500 PC is going to be able to do 99.9 percent of anything that the average home user wants to do, that home user doesn't give a rat's ass what kind of processor is in there.

        On that note, I want to know when the promised System On A Chip computers will be hitting Walmart, etc.

        • On that note, I want to know when the promised System On A Chip computers will be hitting Walmart, etc.

          Oh, man, where have you been all of this time? The system-on-a-chip has come and gone already, in at least the form of the Cyrix MediaGX series of processors. It integrated graphics, audio, memory and PCI controllers into the CPU die. Compaq sold a few machines with them, IIRC, but they sucked badly since users could not upgrade the soldered-on-the-motherboard processors. I've not seen any other devices

      • In a world where a $500 PC is going to be able to do 99.9 percent of anything that the average home user wants to do

        I've heard this argument before, but in the world you're talking about, average home users aren't given nearly enough cool software. I know a lot of average Mac users will gobble the extra Intel horsepower up using the free included software, doing multi-person iChats with high compression, iMovie+iDVDs, Quartz Compositions, etc. The world where the average home user does no video codecs or
        • "The world where the average home user does no video codecs or high-horsepower entertainment uses is ending, and the home computer is for a lot more than the kid's school paper these days."

          I wonder. I just got back from a client's shop where they're still on Windows 2000. At lunch, my clients girlfriend called him, because her computer running Windows 98 is acting funny. LOTS of people in the USA still don't watch TV or Movies on their computer.

          It is probably more correct to say "The world where they trie
          • I concur I find myself in the last week firing up my 100 dollar PIII Dapper Drake machine more than my 2000 dollar dual G5 with 2.5 gigs of ram. The only time I really need the G5 is editing video which I do once a month. And I'm a geek, how many even do that much on their computer? For 90+% of people a PIII 1 ghz machine with 512 megs ram is more than enough than what they do, I know that my be shocking to slashdotters who at least IMAGINE themselves compiling gentoo while watch a DVD, and downloading w
            • "...most people just need web, e-mail, chat, word processing, and BASIC DVD/Web video capacities, all which you can get on a 100 dollar used p.c."

              As an aside to this, most companies I've work for would get a better return on their investment if they spent (more) money on training people in basic computer skills, than on buying shiny new computers. I think over the years Joe Sixpack has acquired a little more computer savvy, but not in proportion to the power of today's hardware. Alas, hardware is sexy an
  • Itanium (Score:5, Insightful)

    by BJZQ8 (644168) on Tuesday June 06, 2006 @01:35PM (#15481288) Homepage Journal
    Itanium exemplifies the "modern" Intel; a processor that looks great on paper, but fails to perform in the real world. Marketing pushes it for all it's worth, but of course it's a cart with no wheels. Perhaps they just underestimated how rapidly multi-core processors would overtake Itanium in processing power? I realize that for certain scientific applications it's still the processor to beat, but there's no chance of it pushing down to the consumer market like the Pentium Pro's did back in their day...and that's what would have made the difference. There never was a critical mass that made people switch, and pushed operating systems to keep up.
    • Strangely enought "modern intel" seems to own most of the laptop market - and not because of marketing, but because Centrino and Core Duo looked good, both on paper and on real-world performance a power consumption tests. Conroe also looks good on paper, and it pretty much seems to beat current AMD offerings in "real"-world (soon people will be able to benchmark Conroe everywhere). And without using (clever) tricks like integrating the memory in the CPU.

      Just because Itanium was a failure and presscot was cr
      • Not so modern. (Score:5, Informative)

        by LWATCDR (28044) on Tuesday June 06, 2006 @02:53PM (#15481988) Homepage Journal
        First the Centrino is not a CPU. It is a chip set. You are thinking of the PentiumM. Don't feel too bad Intel's marketing monster caused more than one person to think that the Centrino was a CPU.
        Second the PentiumM and the CoreDuo are a step back to the PentiumIII.
        Yea the Core Duo may be a new winner for Intel but it all seems to point to one thing. Intel is a one trick pony. Every time they try and replace the X86 line it turns out to be a total failure. The 432 was supposed to be the next big thing way back when. Then came the i860 CPU which Intel pushed as a Cray on a chip. Now we have the Itanium. At some point the X86 will just run out of steam. At that point Intel will be in major trouble.
        • Re:Not so modern. (Score:2, Informative)

          by Anonymous Coward
          Grandparent may have been confused, but parent is far more so. Centrino is not a processor. Centrino is not a chipset. Centrino is a processor + chipset + wifi card. All 3 of them together is what makes a centrino sticket.
        • Intel's marketing monster caused more than one person to think that the Centrino was a CPU.

          I like to call it Centrino. It's a better name.

          Second the PentiumM and the CoreDuo are a step back to the PentiumIII

          P4 was like it was for a reason. They clearly _wanted_ to be able to clock the P4 that high from the start. I'm sick of reading that P4 is crap - P4 "crappiness" is what has allowed Intel to continue selling CPUs. Indeed, P4 is crap compared with AMD but it'd be _much_ worse if intel wouldn't have design
          • Re:Not so modern. (Score:3, Informative)

            by LWATCDR (28044)

            There is every reason to think that Intel wanted to replace the X86 with the Itanium. They said so when the Itanitum was first announced. The Itanium was going to be a family of chips and would be in everything from desktops to supercomputers. If you can find any press releases from that time you will find your proof.
            I don't think that Intel is stupid. Just too that they have been too successful. The PC is their bread and butter. If a chip doesn't run the current software stack faster it is dead on the PC.
        • The point will happen once someone brings out a processor which can emulate the x86 faster than a native x86 can run the code. (Well native is not the right word, even the x86 processors do not run the code natively anymore, but their main exposed instruction set still is x86) Until now this has not happened. Even the Power5 (not the G5 the real Power5) pretty much the fastest mass processor you can get cannot do it, although in raw computing it beats everything by miles the x86 world has to offer. If tha
    • no (Score:3, Interesting)

      It's not a weakness in Itanium, it's the strength of Athlons/Opterons and now Conroe/Merom/Woodcrest. Intel wanted Pentium 4s to be good media processors (and they are), but they didn't want to compete against their own Itaniums, and that's what allowed AMD to steal their lunch. They have no choice but to respond with their own strong x86 processor.

      It's not the weakness of Itanium that's the problem, it's apparently their ability to make any architechture fast and not just the one they want.
    • Itanium exemplifies the "modern" Intel; a processor that looks great on paper, but fails to perform in the real world.

      Some of us remember your "modern Intel" from the days of the iAPX-432 [wikipedia.org]. Has it really been 30 years? Tempus fidget.

  • by tizzyD (577098) <.moc.liamg. .ta. .dyzzit.> on Tuesday June 06, 2006 @01:38PM (#15481308) Homepage
    They got fat, dumb, and happy on their profits and the ubiquity of the WinTel platform. OTOH, AMD has stepped up, innovated, and in places leapfrogged Intel in design, performance, and price.

    Sadly, this effect happens to all large companies. They stop innovating. They instead focus on shoring up their stock and capital to keep investors happy, lowering risk, and start focusing on acquisitions to bring in new ideas and new blood. Intel is no different than any other company.

    There is a solution: skunkworks new ideas. Throw time, effort, money, and brainpower at innovation, with NO guarantee of returns. Then, when new products are created, don't squash them under a controlling corporate culture. Let them grow internally, with support. Intrapreneurship does not look at org charts or worry about the bottom line as much as the next 5 years.

    Welcome to commoditization.
    • Itanium was a 'new' idea at Intel - the problem was that if failed and cost them 5 - 10 billion dollars.
      • Errr, no, it was not a "new idea at Intel." It was an idea that had been festering at HP for awhile under the moniker PA-WW [clemson.edu], and had acquired a thoroughly pervasive case of Kitchen Sink Syndrome. It ended up at Intel through a combination of factors, probably not the least of which was Intel's growing realization that x86 was getting hard to scale. (Intel may have been emboldened by the Mac 68K to PPC switch as well, given that had happened just before Intel joined forces w/ HP.) Thing is, when they to

    • Its their failed skunkworks projects they are selling right now! Intel does this, but have been losing alot lately. As DAldredge points out, Itanic is one of these failed projects.

      Refocus on their core business, that is what they need to do.
    • by Ancil (622971) on Tuesday June 06, 2006 @01:54PM (#15481464)
      What world are you talking about, precisely?

      Intel had profits last year of almost $8 billion versus a market cap of $106 billion.
      AMD had profits of about $370 million on a market cap of about $15 billion.

      That means Intel is giving about three times the return on investment.

      Oh, and their newest chip pretty much squashes everything on the market, including their own current "Extreme Edition" offerings.

      AMD has no viable laptop chips, while the Core Duo has been out for months. Did I mention that laptops account for the majority of new computers purchased? And that they're far more profitable that desktops?
      • Note to the financially illiterate: Profits/Market Cap != return on investment.

        I could buy 100 shares of AMD at $30 for $3000 today. If the stock goes up to $40/share in a year and I sell, I've made $1000 for an investment of $3000. In the meantime, AMD might have actually lost money that year (yes, companies can lose money and still gain market cap). So in this scenario, even though AMD lost money, I as an owner of that stock, had a return on investment of 33% in one year.

        The numbers I'd be more intere
      • So Intel's 95% of the profits and 88% of the market cap only nets them about ~20% performance bonus over AMD's tech? Something must be wrong with Intel, I think they're hiring either bad engineers or bad management when compared with AMD.

        Of course their focus has not been on the x86 desktop market for some time..
      • Intel had profits last year of almost $8 billion versus a market cap of $106 billion.
        AMD had profits of about $370 million on a market cap of about $15 billion.


        Note which direction AMD's profits are going [yahoo.com]. Their stockholders' equity [yahoo.com] is growing very nicely too. Current assets are up, long-term debt is way down.

        Now look at Intel. Profits plummeted last quarter [yahoo.com]. Stockholders' equity [yahoo.com] is down, thanks to their cash balance plummeting from $11B to $5B over the past year.

        You can make Net Income say whatever you
    • There is no reason for including their "trusted computing" crap on their processors/platform.

      Objects telling their OWNERS they refuse (this is more than just software suggesting that an action may be a violation of law) to do something because they think it might be illegal and secretly send private info to a mothership. What kind of BS is that? Clearly only fools would buy such a device unless they hadn't other options. And intel misbetted that customers had other options.

      It's basically untrusted computing
    • " They stop innovating."

      Actually I think the problem at Intel was they innovated too much and in the wrong direction. They poured billions in to Itanium and its horrible for any volume market, its really only good in the tiny supercomputing niche. They really took their eye of the commodity market, where their profits come from, fiddling with that monster.

      They also innovated too much on Pentium 4 and it ended up with an excessively complex, hot expensive processor that didn't perform well, and its not go
  • It is bad for consumers when this happens. We need a healthy competition between AMD and Intel to keep prices down and performance up. When either side becomes dominant, the consumers are the only ones to suffer. Let's hope the Core Duo redresses the balance.
    • It is bad for consumers when this happens. We need a healthy competition between AMD and Intel to keep prices down and performance up.

      Though there wasn't much to the article, and it repeated itself. However, it gave indications that we have very healthy competition: "The average price of a PC processor in April was less than half what it was in March[...]". Intel, last I heard, still has a commanding majority of the processor market. Even if they falter a bit for a product generation or two, they have bi [yahoo.com]
  • by Sparagmei (877929) * on Tuesday June 06, 2006 @01:41PM (#15481342)
    I've been thinking recently: what if Intel, realizing it had made serious errors in architecture and pricing from which it could not readily recover, decided to effectively feign death and allow its competitors to get hazardously cocky?

    AMD overextending itself in an attempt to grab lots of market-share from Intel could prove very damaging when Intel 'gets it right again', such as with the Conroe exploding all expectations. An Intel offering that relies on sheer quality, rather than extortionate market dealing, could wreck AMD's edge and turn all their forward-thinking investment into a Sisyphean debt load.
    • You're right!

      So, is AMD overextending itself and getting cocky?
    • I've been thinking recently: what if Intel, realizing it had made serious errors in architecture and pricing from which it could not readily recover, decided to effectively feign death and allow its competitors to get hazardously cocky?

      AMD overextending itself in an attempt to grab lots of market-share from Intel could prove very damaging when Intel 'gets it right again', such as with the Conroe exploding all expectations. An Intel offering that relies on sheer quality, rather than extortionate market de
  • by frankie (91710) on Tuesday June 06, 2006 @01:43PM (#15481361) Journal
    ...and I mean literally TODAY, you'll notice that they are presenting their brand new Core2 CPUs at Computex. I suppose it's true that you "won't find a trace" of them in the current retail channel, becuase Core2 starts shipping in 3 weeks.

    That's where Intel is focusing, and that's where Intel wants their customers to focus. I bet 80486 sales dried up pretty badly right before the Pentium launched too.
    • Very good point. Anyone who has seen the Core2 benchmarks wouldn't think of buying a new computer in the last 3-6 months. Core2 will quite simply blow everything away. Besides an overclocked Pentium D at 4GHZ (see tomshardware), the Core2 will provide the biggest jump in performance released by anyone since the original Pentium.

      AMD currently has a lot of momentum though, when the Core2 is a huge success, it will take some time for people to realize it, and perhaps give enough time for AMD to come back wi

    • Why would a user buy a Core2 machine when you can get a Pentium D for far less money?

      They're both dual core, and the D is older slower and puts out far more heat. But for most people, paying several hundred dollars isn't worth getting the Core 2.

      I'm amused that Intel is now competing with their old kludges.

      Did you read the article? Do you see how much sales are down in the US? It's huge.

      The whole market is hurting, it mentions that AMD won't see much of a change in their income.
    • The 486DX (and the SX at the very low end) continued to sell well for a couple years after the Pentium was introduced. It occupied the "Celeron" area of the market, and same for the 386 when the 486 appeared, and for the 286, and so on.
  • Intel had it coming (Score:2, Informative)

    by alfs boner (963844)
    They've been screwing over their customers for 15 years. With stuff like the spying serial number, tpa, etc, they've had an attitude of buy what we tell you or get lost. Not to mention price manipulation. They hold back each new iteration until prices slack off on the current product. AMD beat them to the 1 GHz punch because intel was holding back their own 1GHz chip to squeeze more profit. After AMD beat them, they released theirs 2 days later.

    Now that it's coming back to bite them on the ass, I think it

  • The company I work for has Intel as its biggest customer. BTW, you guys should use Intel's new logo - it's been out since January.
    • Slashdot has been using the old Gnome logo for so long its becoming a running joke. They changed their logo in 2002. See the response from the Gnome foundation here. [slashdot.org]

    • BTW, you guys should use Intel's new logo - it's been out since January.

      Gnome updated their logo in 2003, and Slashdot still does not use it. Slashdot does not cater to corporate or marketing whim when choosing its logos - Slashdot liked the old one better, so that's the one it's gonna use!

      After all, it's been years since Microsoft used the "Bill Gates is the Borg" logo for the company, since Star Trek is so out of vogue. And yet, Slashdot flat-out refuses to move to the new "Steve Ballmer is Darth Vader"
  • Imagine (Score:3, Funny)

    by bblboy54 (926265) on Tuesday June 06, 2006 @01:51PM (#15481436) Homepage
    All those who maintain the power grid can now relax a little.
  • Well Duh? (Score:4, Insightful)

    by TheSkepticalOptimist (898384) on Tuesday June 06, 2006 @02:02PM (#15481542)
    Actually, I am surprised it has taken this long for people to realize that the P4 generation of products suck. I mean, for the last 3 years it has been reported that Intel hit a brick wall with the Netburst architecture and they needed to revamp their architecture before they could become competitive again. Each new revision of the P4 architecture simply prove that point, where even the P4EE CPU's with their $1000+ price tags fail to outperform $300 CPU's from AMD. New chips based on the Pentium 3/M architecture proved only to be able to offer decent performance per watt, but still failed to outperform AMD in most benchmarks.

    I am also surprised that it has taken Intel so long to realize this. Even today, they are still flogging the P4 architecture. With the Duo Core CPU's out, you can't even buy this as a desktop system yet, and they are set to release the Duo Core 2 CPU's later this year. Intel should have scrubbed ALL products with the P4 architecture and simply moved forward to their Core architecture.

    Anyways, I will once again be an interesting time in the CPU market as Intel releases their next generation products. Initial reviews seem positive that Intel has something that can compete against AMD, and this will only motivate AMD to produce new technology (AMD has been stuck in a rut as of late). A price war is necessary as CPU prices are staying far to high these days as neither company has really been in competitive form. AMD has locked the gaming market and Intel has the business market locked, these are two non-competing markets, and both companies have pretty much set their price lists accordingly.

    I don't care who makes the next best CPU, I am neither an Intel nor AMD fanboy, I want a system that performs well for the money. Its been AMD for the last 5 years and if Intel finally puts their money where their mouth is and actually delivers a product that offers good price/performance/power features, I will switch back to the Intel platform. Just, its about time Intel started focusing on RELEASING their next generation architecture to the masses and stop talking about it.
    • "With the Duo Core CPU's out, you can't even buy this as a desktop system yet,"

      Hmmm. The Intel iMac sitting on my desk seems to prove you wrong.
  • by a_nonamiss (743253) on Tuesday June 06, 2006 @02:02PM (#15481543)
    For the last 5-10 years, (Really since the Athlon was released) AMD was indisputably the price/performance king. Hands down, no competition. During most of that time, they had the fastest chip on the market, and they were consistently cheaper than Intel at every price point. Intel only kept its market share high using artificial means, by exclusivity contracts (*cough* Dell *cough*) and by spreading inaccurate FUD about AMD's potential compatibility issues. I personally NEVER had a SINGLE compatibility during that time, and I used AMD's on dozens, if not hundreds, of computers. Finally, after 10 years, Intel's complacency has finally caught up with them. AMD has always had a strong connection with the geek community, but after all this time, the word has finally gotten out to the mainstream, and Intel is losing market share FAST. If the largest PC manufacturer didn't have an exclusive contract with them, they would be a distant second (or third) in the chip race right now.

    Ironically, if their roadmap is to be believed, Intel may have just begun a genuine turnaround. They already have 65mn fabs running at higher capacity than AMD, and they are near to bringing 45nm fabs online, which AMD has not done. Also, for the first time in a number of years, Intel actually has a production chip that looks to be genuinely faster than AMD's best offering. (Although it costs an arm and a leg.)

    The question really is, is it too little, too late? One thing's for sure, the competition is great for the consumers. :)
    • They already have 65mn fabs running at higher capacity than AMD, and they are near to bringing 45nm fabs online, which AMD has not done.

      That could have something to do with Intel having 11 chips fabs across the world compared to AMD's 2 (including Fab 36 in Dresden which isn't even fully online yet).

      Also, for the first time in a number of years, Intel actually has a production chip that looks to be genuinely faster than AMD's best offering. (Although it costs an arm and a leg.)

      Actually, largely because AMD

  • Intel designed the "ultimate" microprocessor - on paper. They might have been successfully in forcing every software company on the x86 instruction set to recompile but AMD saw the weakness in their strategy. Yes, x86 is weird, arcane, backwards, and messy. The instruction set is rooted in late 1960s and early 1970s computing. But, it would cost millions if not billions of dollars for the companies dependent on x86 designs to completely ditch the x86 architecture. AMD exploited this lack of foresight b
  • If sales were unexpectedly slack in April, Intel would have built up excess inventory. With Core2 impending, they would need to bleed excessive previous generation parts off in hurry. Excluding coercive tactics, the one of the best ways to get sales quick is price cuts.

    The price cuts probably have little to do with competitve reasons and are more likely operational.

    Oops, just relized the pun. Funny anyway.
  • by Channard (693317) on Tuesday June 06, 2006 @02:36PM (#15481848) Journal
    Working in a computer shop, I rarely get people asking for a specific processor. If anyone actually comes with a definite spec they'll say how much memory they want, how much hard disk and so forth. The actual CPU rarely comes up unless they're seriously into gaming. And everyone I've had come in looking for a Pentium doesn't know the difference between Pentium and Pentium D. They just know that Pentium is apparently a good thing to have, but don't know more than the word Pentium. If we had any Pentium 2s still kicking about I could probably sell them that and they'd nod 'Ooh, Pentium, yep, that's good' without being aware of the distinctions between the chips.
  • Flash will never go (Score:4, Informative)

    by FuzzyDaddy (584528) on Tuesday June 06, 2006 @02:39PM (#15481868) Journal
    Intel will always make flash memory. Flash provides an ideal platform for verifying a fab process and/or facility. The reason is that it consists of huge arrays of identical transistors in a grid, with a relatively small amount of additional circuitry. This makes it very easy to find and diagnose bad ones. It may not be a big business for them, and they may always lose money on it, but it remains an important part of keeping a handle on the process.

    A bad transistor (or contact or whatever) on a microprocessor can be very difficult to track down. Pass/fail testing is pretty good on them, but actually identifying the source of the failure can be really tricky and time consuming.

    • Flash memory is 4 extra masks over a straight CMOS process, which will add complexity which you will need to debug. It makes much more sense to test a fab with regular test patterns and SRAMS (Which will be needed in the final CPU). Do they even run flash in the same fabs as CPUs?
      • I don't know if they run in the same fab. I do remember being told (when I was in the IC business) that flash is what you run first when you are bringing a new fab on line. Another reason it makes a good test product is that it provides an excellent test of the quality of the gate oxide - any defects can be detected by looking for charge leaking off the floating gate.
        Perhaps someone a little closer to the business would weigh in on this one?
  • Why Buy Now (Score:4, Insightful)

    by Nom du Keyboard (633989) on Tuesday June 06, 2006 @02:39PM (#15481876)
    Intel's processor sales dropped 52 percent this April as compared with April one year ago.

    And seriously folks, why would you buy a 32-bit only, non-VT enabled, hot running, substantially slower Intel chip today if you can possibly delay until Conroe comes out?

    Heck, even if you don't want a Conroe Core 2 Duo 64-bit VT chip, the prices on the old stuff are going to drop through the floor. Golly, they're already through the floor. How about drop to the center of the Earth.

    An Intel PC -- including all Macs -- is just not a good buy today with the next generation so close at hand.

  • by suv4x4 (956391) on Tuesday June 06, 2006 @02:43PM (#15481910)
    It's not a current gen, it's a filler between last gen and next gen.

    And their marketing and branding sucks. I still have no clue which Core Duo chip is supposed to be fastest or whatever. So I just don't buy.

    Way to go Intel.
  • Core Duo = Duo Price (Score:2, Interesting)

    by Beefslaya (832030)
    When my server supplier switched to the Duo format from HT, the cost of my servers went up 250 dollars (Double the CPU cost). Consequently, I moved manufacturers.

    I still stick with Intel procs in my servers (AMD X2 at home) for stability reasons (Troll me if you must) but NFORCE (IMO the only really stable AMD system)support for a 1U rack mount server using Linux has been a bitch for me to find.

    I'm out buying up P4HT procs as fast as I can get them.

    As Intel keeps jacking prices, I keep looking at AMD to be
  • Offtopic (Score:3, Informative)

    by tetabiate (55848) on Tuesday June 06, 2006 @03:06PM (#15482098)
    Reading Slashdot news has become a pain in the *ss after somebody decided to use the ugliest and smallest fonts available.
    • Re:Offtopic (Score:3, Informative)

      Reading Slashdot news has become a pain in the *ss after somebody decided to use the ugliest and smallest fonts available.

      It's a markup language, so fix it already. Specify your own CSS for the site, or just to always use +1 fonts sizes on this site. Another issue is that it looks like it uses a special style for IE to make up for the fact that IE breaks the font size conventions, so if you're using a decent browser, but identifying it as IE, you are probably getting smaller text than everyone else. That

    • Re:Offtopic (Score:2, Informative)

      by anaesthetica (596507)
      Tahoma is an abomination of a font to begin with. The move away from a serif font for articles and comments was a huge mistake. Sans-serif is fine for headers, titles, and sidebar options. But for paragraphs of text meant to be read, 14px serif fonts designed for the screen are best (read: Georgia [microsoft.com] (also here [wikipedia.org] and here [will-harris.com]).

      Tahoma was made specifically for small-font-size menus and titles, not for large blocks of text (see here [microsoft.com]).

      Keep the left and right sidebars in small-size Tahoma, but please please pleas

  • What with processors like the Pentium D 805 costing a mere $130. It might not be the kind of competition I want, but it makes CPUs damn cheap.
  • Moore's Law? (Score:3, Interesting)

    by PingXao (153057) on Tuesday June 06, 2006 @04:58PM (#15482972)
    I have a P4 running at 3.2 GHz right here right now. It will be 2 years old in October. For the first time I can ever remember, a PC is going to reach its birthday and NOT be obsolete. Something's going on, I tell ya. Normally there would be at least a 6.4 GHz machine out there somewhere begging me to buy it.

    Multipl cores are nice but really need software to evolve more before they will hit their stride. Multithreaded programming separates the men from the boys and an entire segment just doesn't "get it". IMO that's because a well-designed multithreaded program can't be developed by sitting down and banging on the keys. It is no longer enough to say "the documentation is in the code" because there are architectural aspects of the code as it relates to multiprocessing and multithreading that don't lend themselves to being documented that way.

    Outside of server apps can anyone point to a software package that really needs more than one processor? Even my 2-year-old P4 has 2 logical processors inside it but I recall there is more than one software package out there that won't run properly unless that feature is turned off. Nice job.
    • Simple - desktop responsiveness. I started going dual-proc at home when I picked up a 2-way PII-300 system that was perceptibly faster than a higher-clocked box for a long time. Sure, long running CPU-hogging processes were slower for the most part, but I didn't run many of them. What I did do was have multiple apps up at the same time, all doing little things for me, while wanting decent performance from my file manager et al. That is where the modern dual-core trend really shines.

      Interestingly, dimini
  • Ah intel (Score:4, Insightful)

    by MemoryDragon (544441) on Tuesday June 06, 2006 @05:05PM (#15483035)
    Processor strategies straight out of the applied idiocy. Intel has excellent processor designers, but up until recently their roadmap sucked. They overlooked the point where pushing the GHz boundary did not make sense anymore. AMD basically gave them the first kick in their butt by extending the x86 architecture top 64 bit. The second kick was when AMD dropped out of the real GHz game, by limiting their clock speed but still bringing out faster processors. Intel could have coped with both. First dont play the deny everything game in case of AMD64 and then bring out the same under your own name because we fucked up big time. Second, their israeli design team gave them the biggest kick ass processor of its generation, the pentium-m, yet intel needed another 2 years to recognize what gem they had at their hands, low power consumtion, beating their fastest ghz monsters left and right and only due to excellent design. They pushed those processors into the notebooks while they could have been AMD killers on the desktop and servers left and right. It took them another two years to recognize that while everyone was telling them openly, to give up the P4 line and go for the -M line in the desktop as well. Well their roadmap now shows sanity with going for extended -M processors with multiple cores, but since around 2000 until recently Intels road maps and plans were as idiotic as Sonys handling of the PS3 currently is. It is a wonder that Intel did not get kicked more than it already has.
  • I'm sure it does not account for the full drop but I am amongst what appears to be a large group waiting for Conroe to be released.

    I had been mulling an upgrade for a few months but now that I know Conroe will be out in July and have seen some of the benchmarks people are making public I'm going to wait until I can get a Conroe.

    I know Superpi is a pretty artifical benchmark, however, my current P4 2.8 does 1M in around 50-52s. IIRC a Conroe at 2.67 Ghz does it in something like 16-18s. Considering how good

Today's scientific question is: What in the world is electricity? And where does it go after it leaves the toaster? -- Dave Barry, "What is Electricity?"

Working...