Slashdot Mirror

← Back to Stories (view on slashdot.org)

Intel's 14nm Broadwell Delayed Because of Low Yield

Posted by Unknown on from the physics-hates-materials-engineers dept.

judgecorp writes "Intel has put back the delivery of its 14nm Broadwell desktop chip by a quarter because of a manufacturing issue that leaves it with too high a density of defects. The problem has been fixed, says CEO Brian Krzanich, who says, 'This happens sometimes in development phases.'" The good news is that it is just a defect density issue. A first round of tweaks failed to increase yield, but Intel seems to think a few more improvements to the 14nm process will result in acceptable yield.

96 comments

  1. 14 nanometers should be enough for anyone. by Anonymous Coward · · Score: 5, Funny

    14 nanometers should be enough for anyone.

    1. Re:14 nanometers should be enough for anyone. by Anonymous Coward · · Score: 0, Redundant

      Your exgirlfriend didn't buy that bullshit line of yours either, and now she's sucking on my 304,800,000 nanometer.

    2. Re:14 nanometers should be enough for anyone. by Bacon+Bits · · Score: 2

      Nah, we're not quite at that point yet. I've seen estimates that we could see 1nm processes by 2030, but many people say anything below 5nm (expected circa 2020) isn't feasible. Either way, we're at about the manufacturing limit of the Newton and Thomson/Bohr/Rutherford universe. Atoms are between 0.3 to 3 Angstroms in size. That's 0.03nm to 0.3nm. If we want to go smaller than that, we have to construct our devices out of something other than atoms, and it's assuming that subatomic/quantum forces don't negate the practicality of devices at that scale already.

      --
      The road to tyranny has always been paved with claims of necessity.
    3. Re:14 nanometers should be enough for anyone. by Charliemopps · · Score: 1

      or stop using electrons

    4. Re:14 nanometers should be enough for anyone. by K.+S.+Kyosuke · · Score: 2

      14 nanometers should be enough for anyone.

      It's not a problem of 14 being enough, it's a problem of 14 being too much.

      --
      Ezekiel 23:20
    5. Re:14 nanometers should be enough for anyone. by K.+S.+Kyosuke · · Score: 1

      Must be pretty rigid if you can measure it reliably to four decimal places. Are you sure it isn't a fake?

      --
      Ezekiel 23:20
    6. Re:14 nanometers should be enough for anyone. by K.+S.+Kyosuke · · Score: 2

      If we want to go smaller than that, we have to construct our devices out of something other than atoms

      You could shrink the atoms using muons, but overclockers will suffer a nasty surprise.

      --
      Ezekiel 23:20
    7. Re:14 nanometers should be enough for anyone. by peragrin · · Score: 2

      Two minor points.

      Electrons are Easy to use and last a long time.

      Subatomic particles on the other hand are much hard to deal with. Also not many of them are actually smaller.

      --
      i thought once I was found, but it was only a dream.
    8. Re:14 nanometers should be enough for anyone. by cheater512 · · Score: 1

      Photos are easy to use (kinda) and last a long time. Oh and they are definitely smaller.

    9. Re:14 nanometers should be enough for anyone. by slack_justyb · · Score: 1

      The thing to remember is that roughly about 1nm is the end of the line for the current process for chip making. It doesn't mark the end of circuits, it just means we need a different method. That could be with Silicene, nano tubes, or heck even quantum computers. This isn't the end of the progress of processors, it's just the end of Moore's Law. There might be even a 0.5nm or 0.1nm era, but there will be some serious diminishing returns for that (unless someone is really clever)

    10. Re:14 nanometers should be enough for anyone. by hairyfeet · · Score: 4, Informative

      Actually if the rumors some of the other sites are saying is true the 14nm delay isn't because of low yield...its because nobody is buying. Oh sure the yields aren't great but like AMD it looks like Intel has realized there really ain't a point in constantly putting out faster and better chips when they can't move the ones they got.

      What both need to realize, and what will be biting ARM right in the ass in less than 18 months by my calculations is thus....The software just hasn't kept up with the hardware and X86 by switching from MHz wars to core wars went from "good enough" to "insanely overpowered" and when you can't even stress the one you have, what is the point of buying a new one? Intel and AMD are finding this carries over to other areas as well, take laptops for example. Used to you could set your watch by my customers replacing their laptops, every 2 years for the business guys and every 3 max for the home users, because the combo of heat cycling and software requirements would make them break or painfully slow, now? Well most of the time the laptop is twiddling its thumbs so its not getting hot enough to kill it and even a 5+ year laptop these days is a C2D or Turion X2 with 3+ GB of RAM and 300GB+ HDDs, more than Joe and Sally Average need frankly.

      Oh and for the guys praising ARM and thinking that train is gonna keep on rolling? You got 18-24 months by my calculations and then? Hope you enjoy the same boat Intel and AMD are in now. The reason why is simple...ARM doesn't scale well and there is only so many cores and so much MHz you can push in a thin and light before you end up with battery life measured in minutes so just like how Intel and AMD hit the heat wall? So too will ARM hit the battery wall. When you combine this with the incredible race to the bottom going on right now, we are talking about dual core tablets in the $70 range at Chinamart and quads starting at $100? it won't be long before everybody and their dog has a phone and tablet that is faster than they know what to do with and then like X86 they won't replace until the unit dies.

      so I wouldn't be surprised if intel just sits on 14nm until they get it down so well they can sell it as cheap or cheaper than current chips, after all AMD has already said it'll be a year before they release a new chip and why should they? Thanks to having a mature process they can sell hexacores for $100 and octocores for $130 and their yields on the APUs is so good the OEMs are selling quad laptops for $399, why spend all that money for a new chip when sales are already depressed? The same goes for Intel, they have chips at just about every price point, mature process means high yields and more profits per wafer, and with the global economy a crawl and PCs becoming like appliances why come out with a new chip? Stick with what you've got, they are several orders of magnitude faster than Joe and Sally know what to do with anyway.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    11. Re:14 nanometers should be enough for anyone. by Anonymous Coward · · Score: 0

      Not what she said.

    12. Re:14 nanometers should be enough for anyone. by Anonymous Coward · · Score: 0

      There are no decimal places in the number 304,800,000, genius.

    13. Re:14 nanometers should be enough for anyone. by Arancaytar · · Score: 1

      There are in the number .3048, which is in the standard SI unit.

    14. Re:14 nanometers should be enough for anyone. by synapse7 · · Score: 2

      Yeah, because nobody that runs devices on battery power wants faster more efficient devices. Until my phone can last a week while running crysis you have no point.

    15. Re:14 nanometers should be enough for anyone. by Anonymous Coward · · Score: 0

      "Actually if the rumors some of the other sites are saying is true the 14nm delay isn't because of low yield...its because nobody is buying."

      If this were true and the yield is ok Intel would be using it right now. Making chips smaller means more chips per wafer and less waste.

    16. Re:14 nanometers should be enough for anyone. by Bengie · · Score: 1

      Smaller than a 0 dimensional point particle?

    17. Re:14 nanometers should be enough for anyone. by Bengie · · Score: 1

      The retooling is crazy expensive, in the order of tens of billions. If you can get more money from your old machinery while refining your next gen, then to keep from competing with yourself.

    18. Re:14 nanometers should be enough for anyone. by Anonymous Coward · · Score: 0

      There are no decimal points, but there certainly are decimal places in that number, unless you are implying it is not base 10.

    19. Re:14 nanometers should be enough for anyone. by Anonymous Coward · · Score: 0

      Both act like a 0 dimensional point particle in some abstract way that doesn't directly matter in actual use and measurement. Both will have the same effective size from their wavelength for a given energy. But would it be easier to wrangle 1 keV photons or 1 keV electrons?

    20. Re:14 nanometers should be enough for anyone. by hairyfeet · · Score: 1

      Exactly and if a run of the new chips cost more to make, counting lower yields and the retooling, than they can get in a downturn? Then it simply makes no sense to go 14nm right now.

      This is why when AMD said they were gonna wait a year to release new chips while the pundits screamed "they are giving up!" I pointed out it was a DAMN smart move, the current process is mature and allows them to sell nearly 100% of the yields. With the current process they can get octo-hexa-quad-triple-dual, all by simply turning off bad cores and they can crank them out cheaply enough you can get a hexa for $100 and an octo for $130. This lets them offer more bang for the buck while covering the market and minimizing waste...where is the downside? Its not like folks are lining up at midnight to get the latest chips as many have more power now than they can use, so by sticking with a proven reliable mature process they can make the most profit with the least expense...it just makes good business sense.

      --
      ACs don't waste your time replying, your posts are never seen by me.
  2. 640nm ought to be enough for anyone.... by Anonymous Coward · · Score: 0, Funny

    Really - why do we keep shrinking these things. 640nm chips worked just fine and probably had fewer defects.

    1. Re:640nm ought to be enough for anyone.... by Anonymous Coward · · Score: 0

      Serious or weak troll.... Weak either way frankly.

    2. Re:640nm ought to be enough for anyone.... by Anonymous Coward · · Score: 0

      Someone doesn't know his epic moments of computing history.

    3. Re:640nm ought to be enough for anyone.... by unixisc · · Score: 2

      Upto 50nm, it was fine, but is now in the region of diminishing returns. The cost savings that were always synonymous w/ shrinks are no longer there, since the process costs easily outweigh the cost savings per die, even assuming a 100% yield.

    4. Re:640nm ought to be enough for anyone.... by Ken_g6 · · Score: 2

      Because a wafer of pure silicon has a high, fixed cost. The smaller you can make a chip made from that wafer, the more chips you can make and the lower the cost for each chip. Intel's recent 22nm chips have recently been around 200mm^2. If they were made on 640nm, they would be about 200*640/22=5800mm^2, or a square of about 7.6cm (3 inches) on a side.

      Also, smaller wires use less electricity and generate less heat. You wouldn't want 640nm Haswell in a laptop.

      --
      (T>t && O(n)--) == sqrt(666)
    5. Re:640nm ought to be enough for anyone.... by Antipater · · Score: 1, Informative

      For others like me who had to look this up, link.

      --
      Everything is better with chainsaws.
    6. Re:640nm ought to be enough for anyone.... by ImdatS · · Score: 1

      I guess it is less for cost savings reasons (during production) and more about power savings (during operation) and probably also for performance reasons.

      Caveat: I'm not a hardware specialist, not to mention a chip specialist, but I assume (from physics perspective) that shorter distances will reduce power consumption for same amount of transistors. Also, if the total die-size required for e.g. 1b transistors goes down from x to x-y, you could keep the die size but increase the transistor numbers.

      Smaller chips as well as lower power consumption is a requirement especially in handheld devices, wearable computing, and also for "Internet of Things" (or whatever it is called today).

    7. Re:640nm ought to be enough for anyone.... by Anonymous Coward · · Score: 0

      Not only that, but unless there's a lot of redundancy built in, your yield for these chips will be poor relative to the 640nm chips that were shipping when that node was state of the art. The yield of a chip is inversely related to the size of the die, as there's more area for possible errors to hit and screw your chip up.

    8. Re:640nm ought to be enough for anyone.... by Anonymous Coward · · Score: 0

      Another way smaller processes decrease power usage is that more transistors means that you can have more transistors that are rarely used to do perform specialized tasks with minimal power usage. This turns out to be a good idea as once chips get small enough, the number of transistors that can be powered (with a cooling system of a given quality) increasing much slower than the number of transistors that can actually fit on the chip in that space. This is called the "dark silicon" problem.

    9. Re:640nm ought to be enough for anyone.... by Dachannien · · Score: 2

      For others like me who had to look this up, link.

      You had to look this up? Now I feel old. Also, get off my lawn!

    10. Re:640nm ought to be enough for anyone.... by unixisc · · Score: 3, Insightful

      The main reason die shrinks happen is usually at the behest of the manufacturer, and rarely at the behest of the customer (unless we are talking die sales, which we usually don't for Intel, and which is a whole different ball game). There are always downwards pricing pressure on the manufacturers (a tad less on Intel, I'd think, given their elite position amongst fabs) and in order to preserve their margins, they work in these shrink transitions w/ their customers. But make no mistake - for customers, those shrinks imply requalificaiton and a whole new product development cycle before they can go to market w/ those. They'd rather get their price cuts on the same die, except that the manufacturers typically won't give them that.

      In the past, the reason to go for shrinks was improved clock speed, and more recently, it's power consumption. But power consumption alone doesn't drive such market trends, particularly given the expenses incurred - what really drives it is cost. But again, as I said, we're really past the point where shrinks would result in any significant cost savings.

    11. Re:640nm ought to be enough for anyone.... by peragrin · · Score: 1

      No but I would love someone to build one just to compare the performance.

      --
      i thought once I was found, but it was only a dream.
    12. Re: 640nm ought to be enough for anyone.... by Anonymous Coward · · Score: 0

      Did you type this from your iPhone?

    13. Re:640nm ought to be enough for anyone.... by slew · · Score: 1

      Upto 50nm, it was fine, but is now in the region of diminishing returns. The cost savings that were always synonymous w/ shrinks are no longer there, since the process costs easily outweigh the cost savings per die, even assuming a 100% yield.

      I wasn't aware of a 50nm node or shrink, but 65nm was the last "cheap" node (which corresponded to a 55nm shrink). The next popular node was 45nm didn't really take off before the shrink to 40nm. TSMC (one of the major foundaries) actually blew-off the 32nm and 22nm nodes completely and only productized the 28nm shrink and 20nm shrink.

      Usually, the "shrink" is purely a cost/die = (cost/wafer)/(yield*die/wafer) issue. Since in a shrink, the cost/wafer is mostly constant but the die/wafer goes up, the cost/die goes down (usually the yield goes up too if it's defect density limited, but not so much if it is parametric limited).

      Moving to a new node, however, tends to give other benefits (e.g., faster transitors, lower voltage swings), but at higher wafer costs and more design restrictions and complexity. Recently this has been diminishing returns from a cost point of view. For example, the Intel's 22nm node with fin-fet/tri-gate gave the benefit of a much lower static power, but the average transitor size didn't go down very much and I'm sure the wafer cost was much higher. Still, the benefit of lower static power gates for mobile devices (or other thermally limited devices) is hard to overlook even if the cost per die is higher (meaning the new features are worth the additional cost per die).

    14. Re:640nm ought to be enough for anyone.... by ImdatS · · Score: 1

      Thanks, that was quite helpful in understanding the manufacturer's perspective as well.

    15. Re:640nm ought to be enough for anyone.... by Anonymous Coward · · Score: 0

      I'm glad you're not in charge, or if you were, you would've gotten thrashed in the industry. We're here at 14-20nm today because of competition; some of which you correctly indicate was not an immediate financial benefit to the chip producers. But that's the way it should be. Why can't we just improve for the sake of improvement?

    16. Re:640nm ought to be enough for anyone.... by Technician · · Score: 1

      The comment is only half the storry. The other half is related to yield. Take for example a 12 inch wafer and make chips 3 inches square. You get at most 4 chips. Smaller chips = more die.

      Now the defect density part of the formula. Say you have 100 defects randomly spread on the wafer. Your yield of 3 inch chips would be zero for most of your manufacturing. Dice the wafer into 500 chips by having smaller die size and now your yield is 4 of 5 die or better is your yield as sometimes randomly scattered defects will be on one die so on that wafer you have fewer dead die.

      Defect denisty is measured in defects per square cm. Lowering total defect count is important in smaller die as smaller defects result in dead die, so a product shrink = higher defect density. Reducing defect density = higher yield. Reducing die size = higher yield, lower power, lower cost, and faster. Lower cost, faster chips that run long times on small batteries is a requirement to make chips for tablets and phones.

      --
      The truth shall set you free!
    17. Re:640nm ought to be enough for anyone.... by bill_mcgonigle · · Score: 1

      Why can't we just improve for the sake of improvement?

      Don't worry, Intel doesn't invest in the next generation of chip technology and say to themselves, "this is cool, but we're never make money on it."

      --
      My God, it's Full of Source!
      OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
  3. Geat! Time to cover my short position in Intel... by bobbied · · Score: 1

    Short term setback for Intel. They will get yield up eventually. I just hope it's before they run out of cash to run operations...

    --
    "File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
  4. Hey... by Anonymous Coward · · Score: 0

    I have always been told size doesn't matter!

    1. Re:Hey... by Anonymous Coward · · Score: 0

      By who? Your mother?

    2. Re:Hey... by Anonymous Coward · · Score: 0

      Looking for a date from women who like small equipment?

    3. Re:Hey... by Anonymous Coward · · Score: 0

      No, by *your* mother.

      BAZINGA!

    4. Re:Hey... by Anonymous Coward · · Score: 0

      Yeah but she does anal even with dirty hobos. Enjoy getting the clap and gonorrhea.

    5. Re:Hey... by Anonymous Coward · · Score: 0

      Steve Stifler, is that you?

      How is your mother?

      Finch

  5. No real long-term impact by ericloewe · · Score: 1

    Since, in practice, they'd want to get rid of old stock before selling their shiny new product, this isn't really that much of a problem.

    It's not like AMD is going to magically beat Haswell before Broadwell is released. It would be nice if they did, though...

  6. Re:Geat! Time to cover my short position in Intel. by RightSaidFred99 · · Score: 1

    I just hope it's before they run out of cash to run operations...

    Lolwut? Yeah, umm, that's not even remotely a concern.

  7. Next steps: 7nm, 1nm (1000pm), 800pm, ... by ImdatS · · Score: 1

    ... where does it end? I had to actually check what the atomic size of Silicon is (111pm), so there are only a few years left (maybe 10-20) to reach the atomic level. Then what? I'm really curios as I'm quite impressed how this development came - actually how quick...

    1. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by stms · · Score: 4, Interesting

      Potentially it can keep going until the size of a transistor is just a few electrons across but as we get closer to that point quantum teleportation becomes more of an issue. This is cool video that explain some basic stuff about transistors and the end of Moore's Law.
      https://www.youtube.com/watch?v=rtI5wRyHpTg

    2. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by Anonymous Coward · · Score: 0

      Graphene!

    3. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by TechyImmigrant · · Score: 2, Funny

      Maybe we could build computers out of Planck planks. They're really small.

      --
      I should use this sig to advertise my book ISBN-13 : 978-1501515132.
    4. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by Anonymous Coward · · Score: 0

      Potentially it can keep going until the size of a transistor is just a few electrons across

      Well, electrons don't have size, and are kind of difficult to hold together without some protons in there....

    5. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by Anonymous Coward · · Score: 0

      There will come a time when some there is some drastic breakthrough and we move completely away from silicon-based transistors and onto something better. It would have to provide high-density, high-reliability, lower power consumption, faster switching times, etc, etc.

      I imagine it being some form of "spintronics" or "photonics"... or even some sort of non-binary sub-atomic thing.

      (Captcha: wafers)

    6. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by Anonymous Coward · · Score: 1

      Potentially it can keep going until the size of a transistor is just a few electrons across but as we get closer to that point quantum teleportation becomes more of an issue.

      More often referred to as electron tunneling or quantum tunneling.

    7. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by Anonymous Coward · · Score: 0

      Pashaw... That's just PHYSICS man.. Subatomic Physics... Don't confuse the joke OK?

    8. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by TechyImmigrant · · Score: 1

      If you made a law preventing any transistors below 14nm, the architectural work would continue making things faster.
      Architectural changes have contributed more to speedup than transistor size over time. They are not independent, since smaller transistors allows more integration and co-location, but from where I sit, there's plenty to be done in computer architecture to make them faster and plenty to do in software to stop blowing away so much performance on fripperies, bad drivers, bad memory management and nightmare call back trees in GUIs.
       

      --
      I should use this sig to advertise my book ISBN-13 : 978-1501515132.
    9. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by slashmydots · · Score: 1

      I prefer magic quantum unicorn horn dust. It can solve the meaning of life in 10 seconds!

    10. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by TechyImmigrant · · Score: 1

      Too bad that all the unicorn horns have been hunted and sold to China for superstitious erectile dysfunction remedies.

      --
      I should use this sig to advertise my book ISBN-13 : 978-1501515132.
    11. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by marcosdumay · · Score: 1

      I can't see how photonics could be made fast. It can lead to a huge bandwidth, but not much improvement on single-core speed from electronics.

      That is, unless your photons have a very big wavelenght, and are poralized, because then you enter spintronics domain, and those have either a huge potential, or some very big problem that nobody discovered yet. The breakthroughs that will make spintronics real are mostly at the "how to assembly superconductors with extreme precision" area, and I think we can apply a lot of what we learned from electronics into this domain, but we'll need to learn how to make the superconducting crystal behave better, because nowadays their behaviour is quite random. It's improving, but not on the headlines.

      Anyway, we'll probably see carbon based transistors before any of those. It's a smaller gain from silicon electronics, but it's an improvement, and compared to those techs, they seem to be quite straightforward.

      --
      Rethinking email
    12. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by Anonymous Coward · · Score: 1

      a few electrons across?

      An electron is a point particle, it has no diameter. I think you mean a few *atoms* across. You cannot make a transistor out of a single atom.

      Quantum teleportation?

      I think you mean quantum tunneling.

      sigh...what has happened to the Slashdot of Old, where the comments were insightful and informative. and written by people who actually know the subject they are talking about....

    13. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by Anonymous Coward · · Score: 0

      and dont call me Pashaw

    14. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by justthinkit · · Score: 1

      Is the next microchip frontier something else entirely? Specifically I'm wondering if RAM will become more important as CPUs stop shrinking. It seems like RAM is 15 to 20 times slower than the CPU (for DDR3) at this time. Will this ever change? Will cache RAM grow by a greater factor than CPU transistor size will shrink? If, hypothetically, RAM became as fast as the CPU, we would have vastly increased performance. But how likely is this?

      --
      I come here for the love
    15. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by Anonymous Coward · · Score: 0

      Well, there was that discovery photon molecules a few weeks ago. My guess is the mechanism is linked to how quarks formed from gamma radiation right after the big bang. If I'm right, that means you can custom make your own quarks from photons, which leads to the possibility of creating subatomic particles which are turning complete.

    16. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by Anonymous Coward · · Score: 0

      My guess is the...

      You might want to guess again... or better yet, don't guess, and spend some time learning actual science to get an idea that might vaguely resembles reality.

    17. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by Anonymous Coward · · Score: 0

      So much pseudoscience in such a short post that even Deepak Chopra would be jelous.

    18. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by Bengie · · Score: 1

      RAM vs cache are two different things. Cache scales something like O(n^2), but is a small number, while RAM scales O(1), but is a large number. The latency of a huge sram cache would be horrendous. Very very generalized.

    19. Re:Next steps: 7nm, 1nm (1000pm), 800pm, ... by Bengie · · Score: 1

      Heck yeah. At least 10x more efficient than silicon per clock cycle, supports speeds into the 100ghz ranges, and can handle temperatures well over 200c.

  8. Re:Geat! Time to cover my short position in Intel. by bobbied · · Score: 1

    Sorry, tongue was firmly in cheek on that one..

    --
    "File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
  9. Re:Geat! Time to cover my short position in Intel. by bob_super · · Score: 1

    On the other hand, the guys at Altera have bet the bank on Intel, so they're likely praying that Xilinx's 16nm TSMC process gets delayed.
    While Intel has utter dominance on their market, Altera is in catch-up mode...

  10. Wouldn't put too much in this by Kjella · · Score: 1

    They released Haswell in June, they've barely had time to sell that so Q4 2013 to Q1 2014 is still ahead of their yearly tick-tock. They're not announcing any delay to Airmont which is their mobile 14nm chip and we all know one quarter to or from won't change much in the desktop/server market. In related news AMD posted their Q3 earnings today and their CPU sales are still down, their gross margin is down but on the bright side the console sales are finally coming in so overall they're making a profit this quarter. Inventory is way up but I hope that is due to build up before the PS4/XBone launch, what disturbs me is that their R&D is still going down. That's a death spiral in the CPU/GPU/APU business.

    --
    Live today, because you never know what tomorrow brings
    1. Re:Wouldn't put too much in this by marcosdumay · · Score: 1

      their CPU sales are still down

      So are Intel's. This is not a good time to sell x86 CPUs. AMD has a chance to reverse their trends, since they are just a small player, but they'll have to steal that market from Intel.

      --
      Rethinking email
  11. Not the real issue by Anonymous Coward · · Score: 4, Interesting

    Intel has produced two new generations of processor that were WORSE than Sandybridge. Higher power use (under load) and far less over-clockability. The newer part were ONLY better (in desktop systems) if you intended to use their new instructions (vanishingly unlikely) or the integrated graphics (which would be pointless- people buy expensive Intel CPUs to partner them with expensive GPUs from AMD or Nvidia).

    Intel, of course, were in the same position with the waves of Core2 parts, each of which essentially overlapped each other in performance generation on generation (although power consumption was much improved over the first generation of Core2 i7 parts).

    Intel currently doesn't know exactly where to go in the near future, and is attempting to hedge its bets by trying various things. It is currently undercutting its own HYPER-expensive ULV mobile high-end parts with the new 4-core 'atom' Bay-trail chips that seek to go head-to-head with ARM. Because current high-end ARM is so good, Intel is forced to sell a very dangerously good chip (dangerous to Intel's profits, that is) into low and mid-end tablets, running Android or Windows8.1

    However, even Intel's first decent 'Atom' part ever (after 5+ attempts) is beaten by Nvidia's somewhat lame Tegra 4, and Qualcomm's Snapdragon 800. It is exterminated by Apple's new ARM chip, soon to be seen in Apple's new iPad refresh.

    Intel's 22nm process, and use of FinFETs, has been a total disaster so far. A process advantage, and custom designed chips, doesn't allow Intel to beat ARM parts coming from commodity foundries at TSMC and Samsung in 28nm. Sure, Intel can make its own chips smaller than those on the previous process, and theoretically get more parts per wafer, but the per wafer costs rocket, the yields drop (initially), and insanely expensive new plants have to be built to service the new process.

    What does Intel get from spending all this new money on R+D? At this moment in computer history, almost nothing. The x86 is dying, and everyone BUT Intel builds ARM solutions. Every major player has a GPU (graphics) solution as good as Intel, and Intel isn't within a million miles of matching the AAA-gaming GPU designs from AMD and Nvidia (despite the fact that Intel has spent more money than every graphics company combined, across their combined periods of existence, to create its own GPU solutions).

    Intel simply has no current use for its expensive 14nm process. It has built the factories, so it is engaged in a waiting game- waiting for mobile parts to roll off the 14nm production lines that have clear market advantages over its current mobile chips. It just isn't worth Intel's time launching another round of non-improved parts. The market has changed forever, and on-one wants to buy "this season's Intel" for the brand loyalty reasons previously apparent.

    Intel fanboys want 6-core and 8-core parts, but Intel is extremely loathe to risk introducing better value into the desktop market. If Intel properly sold 6-core solutions, they would have to sell 6-core i5 parts, and these would beat-up their EXTREMELY profitable 4-core i7 parts. Intel is too in love with the status quo.

    If Intel's bay-trail 4-core parts prove good enough for tablets and non-gaming laptops, and they will do having greater performance than the more than adequate mobile 2-core core2 parts used in the first decent cheap laptops years ago, where does most of Intel's mobile biz go from here? Bay-trail parts (unlike those years old mobile 2-core core1/core2 laptop chips) also do all the video decoding in hardware, allowing flawless playback of all current video content (and bay-trail is strong enough to do CPU enhanced decode of 4K video recorded in h264).

    Bay-trail is the part Intel moved Heaven and Earth NOT to produce. Bay-trail is the final step on the race-to-the-bottom for x86 based computers that most non-AAA gamers will need. If the only real money Intel makes ends up from chips lie Bay-trail, Intel is done.

    Think about this. In a few weeks, you

    1. Re:Not the real issue by 0123456 · · Score: 1

      In a few weeks, you will be able to buy quite decent Android tablets for $150 using 4-core bay-trail. A little hacking, and you've got yourself a $150 dollar Windows8.1 tablet. A $150 tablet running PROPER unrestricted Windows.

      Yeah, 'cause Windows is, you know, free and stuff.

      At retail, you'd be paying about $100 for Windows alone.

    2. Re:Not the real issue by alvinrod · · Score: 3, Informative

      The reason that there's less overclocking headroom has little to do with the architecture design and a lot to do with the crappy thermal solution Intel started using on their IB and Haswell chips. Basically they started using some really crappy thermal paste instead of soldering the IHS like they did with Sandy Bridge. People who delid their chips and use better thermal interface material get far better results. Some people can get upwards of an extra GHz in speed when over-clocking and see some fairly substantial temperature drops as well.

    3. Re:Not the real issue by baka_toroi · · Score: 1
      Re-read this part:

      A little hacking, and you've got yourself a $150 dollar Windows8.1 tablet.

      Do you really think we won't be pirating the shit outta Windows when we do that?

  12. Pull an AMD by slashmydots · · Score: 2

    So make a "triple core" edition called an i4 where really 1 core just didn't pass quality control so they turn it off. AMD did it and it sold so well they had to purposely cripple working quad cores to meet the demand for triple core chips.

    I have a theory that those T-edition chips Intel made that are just underclocked, hyper-efficient, ultra-low wattage editions of their recent chips are actually just ones that wouldn't run properly at the normal stock clock. I never heard a solid claim that they actually had different voltage regulation circuits or something like that. They just underclocked them and made them have a higher tendency to not click to a full multiplier level as often or for as long.

    1. Re:Pull an AMD by 0123456 · · Score: 4, Interesting

      Except the cores are probably small compared to the L3 cache, so most failures will be in the cache, not the cores.

      Back when I worked in the chip business, we designed them so some components could be disabled if they failed the manufacturing tests, but there were very few that we could actually sell that way. Either the fault would be in the components that couldn't easily be disabled, or there'd be multiple faults in too many places to make it viable.

      I have wondered myself whether the low-power CPUs are just the bin that wouldn't work at normal power levels.

    2. Re:Pull an AMD by Anonymous Coward · · Score: 0

      It's actually the other way around.

      The low power cpus are the cream of the crop top of the bin and demand premium prices.

      The better yield cpus are able to stay stable at low power. On the opposite end of the spectrum you have overclocking, where you add power (bump the voltage) to achieve stability at higher than rated clock speeds.

    3. Re:Pull an AMD by Anonymous Coward · · Score: 0

      Hold on there. Traditionally you try a chip at X Mhz and if it fails you try at X minus Y% until you get it to work. If 60% of your chips can run at 3.0ghz, perhaps only 5% can run at 3.5ghz, which is why there is such a premium on price for those chips.

      As with what 0123456 said, it seems plausible that a T/U/Y chip running at 1.8ghz is just a K/S chip that couldn't hit 3ghz.

    4. Re:Pull an AMD by nateman1352 · · Score: 1

      Actually check out Haswell's die configuration, the integrated graphics takes up about 2 times more area than the L3 cache. Also, look at how dense the transistors are in the GPU area, looks as dense or maybe even more dense than the cache. It wouldn't surprise me if graphics are a source of manufacturing problems in addition to L3 at this point.

    5. Re:Pull an AMD by PPalmgren · · Score: 1

      Problem with this logic is that power consumption is a factor. While the chips are underclocked, they are also undervolted by a proportional amount. Undervolting with underclocking was a rare pasttime by overclockers, some done as hobby and others done in a quest for performance/watt crown. The chips binned for the highest clocks and the chips that overclock the best run at higher clocks on the same voltage, or more efficient. The same chips can be undervolted and perform the highest clock at their lower respective voltage. Since the low power chips are sold on battery life, you can't just take an inefficient processor and undervolt it and call it a day.

    6. Re:Pull an AMD by Bengie · · Score: 1

      Power consumption scales linearly with frequency and with the square of the voltage. "high end" low-power CPUs are not only low frequency, but low voltage. They are more rare than your more general CPUs that are between high-end low-power and high-end higher-performance. Lowering the voltage not only reduces power lost by oscillation of a conductor with capacitance(the square scaling), but also electrical leakage.

  13. Re:AMD by Anonymous Coward · · Score: 4, Funny

    Might explain why I went with an AMD processor, and might get a video card from them soon.

    'I like my chips like my women: hot and slow.'

  14. Good news? by marcosdumay · · Score: 1

    The good news is that it is just a defect density issue.

    And what kind of problem you can have on a fab that is not a "defect density issue"?

    In a related question, can I declare Moore's law dead already, or is there some current fab upgrade that isn't delayed by at least 18 months?

    --
    Rethinking email
    1. Re:Good news? by radarskiy · · Score: 1

      'And what kind of problem you can have on a fab that is not a "defect density issue"?'

      Systematic flaws, like all horizontal wires are printing 10% narrower than intended or the effective dielectric constant at a particular layer trails off evenly from the center of a wafer to the edges.

  15. The biggest savings are in the capacitors by marcosdumay · · Score: 1

    If you reduce all 3 dimensions of a wire by the same proportion, you'll get a wire with highter resistance, not lower. The energy savings from reducing the feature sizes come from reducing transistors and inter-wire capacitance. With a smaller capacitance, you need less charge to turn the transistors on or off, using less power and letting them switch faster.

    That is, untill you let too much current leak through them. Make them too small, and you'll consume more power again.

    --
    Rethinking email
  16. Re:AMD by AHuxley · · Score: 1

    You want a good "cpu" not using too much power, not creating too much heat and fair price.
    At this point in time Intel on the desktop covers more of the first three aspects.
    If Intel goes for a more mobile offering for a few generations, AMD will be back in consideration.

    --
    Domestic spying is now "Benign Information Gathering"
  17. Getting in touch with my inner Grammar Nazi by garyebickford · · Score: 1

    ... because of a manufacturing issue that leaves it with too high a density of defects.

    Sorry, after a long time, I have to put myGrammar Nazi hat on for (I think) the first time. It's not just you, this is just the example that tipped me over the edge - much like the "leaves it with too high [of] a" phrasing leaves the reader tipping off into.... what?

    This type of construction has become endemic in conversation in the last few years, and I'm sorry, but it's cumbersome, ungainly, unsightly, and painful to hear or see. Perhaps, just perhaps, if I say something, this bad practice will lose some momentum and die out.

    I would like to suggest to everyone who uses this "too high of X" construction, to consider using something more like this: "... leaves it with a defect density that is too high." See, that flows, it keeps the primary objective phrase (leaves it with a defect density) nicely collected, with the modifying phrase (that is too high) also nicely collected. IANA GN to the point of recognizing if the problem is a split infinitive or other formal error, but I know the original is bad.

    --
    It's easier to be a result of the past, but more fun to be a cause of the future! http://www.spacefinancegroup.com/
    1. Re:Getting in touch with my inner Grammar Nazi by Daikiki · · Score: 1

      It's too short a season to grapple with so harsh a critique of this minor a transgression.

      Sorry. I guess that was a bit like scratching a chalkboard, but I personally rather like this particular grammatical construct. It's efficient and it front loads the subjective point the author is trying to make, making comprehension easier. Compare: "The season is too short to grapple with a critique that's so harsh of a transgression that's this minor."

      --
      I want the fire back.
    2. Re:Getting in touch with my inner Grammar Nazi by garyebickford · · Score: 1

      It's too short a season to grapple with so harsh a critique of this minor a transgression.
      Wow, you managed to get three of them into a single sentence! :)
      How about "The season is too short to spend it grappling with such a harsh critique of a transgression this minor".

      Our brains read predicively, constructing the most probable usage as we go. (There was an article on slashdot about this recently). In this case, I would say that "of this minor a transgression" is first interpreted by our brains first as referring to a minor - a person of age less than 18. Then we run into "a transgression", which breaks our predictive model. So at first we will try to add a comma, making it "of this minor, a transgression", which seems to imply that this poor child was possibly the result of some midnight tryst. But that seems unlikely, so we backtrack, and try "of this transgression, which was minor", finally arriving at the most probable meaning. That's a lot of work for our poor brains to do, and it illustrates why (contrary to your assertion), it is neither efficient, not correctly loading the point. The point is not the minor, it is the transgression. (The same analysis applies to the other two instances in your sentence - "It is too short a season" is the natural reading. So you are forcing the reader to backtrack and reconstruct for each instance. And "so harsh a critique" is just a poor substitute for "such a harsh critique", but I'll assume you added that one just for completeness and grins! :)

      --
      It's easier to be a result of the past, but more fun to be a cause of the future! http://www.spacefinancegroup.com/
  18. Re:Geat! Time to cover my short position in Intel. by Anonymous Coward · · Score: 0

    16nm has the same wire pitch as 20nm so it's not much smaller than Intel's 22nm.

    Lower yields are also not an enormous problem for FPGAs, you can ship with some gates disabled.

  19. Re:Geat! Time to cover my short position in Intel. by bob_super · · Score: 2

    TSMC's metal 1 pitch is 64nm in 20nm, and Intel's 22nm is 90nm.
    14/16 is indeed expected to have ~64nm pitch, so it's not better than TSMC's 20, but it's a great leap for Intel.

    "disabled": Not quite. When I'm running 491MHz internal, you can't just disable arbitrary logic on me. The slower parts may get away with disabling columns, but you can't change my timing without breaking my design.
    Also, all the hard IP is not redundant, and there's more and more of it.

  20. Re:Geat! Time to cover my short position in Intel. by Anonymous Coward · · Score: 0

    "disabled": Not quite. When I'm running 491MHz internal, you can't just disable arbitrary logic on me. The slower parts may get away with disabling columns, but you can't change my timing without breaking my design.

    This, this, this. And I tend to doubt anybody is even trying to improve yields in slow grades that way. FPGAs need every possible path to be predictable, from data files distributed with the tools. It's not just about setup time -- variability in hold time can be really nasty for timing closure, and hold time violations can't be resolved by reducing clock speed.

    On the Xilinx side, the only yield game I know of is their pseudo-ASIC program offered to volume buyers of large-die parts. You prototype using normal fully functional chips. Once you're ready for production, you supply them with your final bitfile and they handcraft a test program which tests only the fabric resources you're using. This can dramatically improve yield for a large part, allowing them to offer substantially lower prices. You get one mulligan if you need an ECO (and presumably no guarantees that the parts they've already shipped to you will pass the new test vector).

    For small-die parts they don't even bother with that stuff.