6 ms·

From how Apple and AMD are doing with their own processors though, it seems like Intel is just fundamentally doing worse even as things become more difficult wi

From how Apple and AMD are doing with their own processors though, it seems like Intel is just fundamentally doing worse even as things become more difficult with smaller transistor sizes. Apple is going to replace Intel with their own processors because Intel has failed to meet requirements. AMD, with a shoestring budget basically on the verge of bankruptcy the entire time they were doing their R&D, managed to build out a new architecture that has provided amazing results while Intel has basically had nothing to show in the same time. But perhaps there's something I'm missing here. Is there a misconception or lack of information here on my end that needs to be clarified? I can only make my analysis largely as an outsider looking in when talking about semiconductors.

dgacmu 7y ago
Intel was ahead, and hit the wall first. Apple & AMD are not ahead, they're just catching up. I don't want to understate how big a problem that could be for Intel, of course. But they're also doing it on low margin parts, and Intel continues to make bank with their data center parts. I don't think any of this represents a short-term problem for Intel, other than the general downturn in processor sales because fewer people will need to upgrade. But I think it represents a very serious long-term threat. They have some really cool technical advances, like 3D xpoint. But I'm concerned that they do so badly on embedded and custom integration from a long-term perspective.
- vbezhenar 7y ago
  Apple sold millions of iPhones with 7nm chips while Intel struggles to build comparable 10nm chips and keeps releasing 14+++ nm. AMD will release 7nm chips very soon. It does not seem like they are catching up. Quite the opposite.
  - akvadrako 7y ago
    You can’t compare nm between vendors - it’s just marketing numbers.
    - throwaway2048 7y ago
      Not directly no, but the actual feature size and density of TSMC 7nm and Intel 10nm are comparable.
      - pkaye 7y ago
        What about die size?
        throwaway2048 7y ago
        Die size is independent of process size, customers can order most any die size they want.
        wtallis 7y ago
        Then you have to ensure you're comparing chips designed for the same market segment. Die size comparisons work well if you're talking about a Cortex-A53 on 16nm vs 12nm. It doesn't work as well when you're talking about a full SoC, or even a desktop CPU+GPU combo where core counts for both sides of the chip can vary greatly.
  - Wowfunhappy 7y ago
    And iPhones still don't come close to competing with desktop-class processors in terms of performance. iPhones also use much less electricity, of course, but the point remains. I don't know enough about this, but the GP's argument of "Intel hit the wall first because they were the first to reach that level of performance" makes logical sense to me.
    - robocat 7y ago
      "Apple’s iPhone Xs is faster than an iMac Pro on the Speedometer 2.0 JavaScript benchmark" Yes, Safari, but iPhones do compete with your average desktop processor (not the top end). https://macdailynews.com/2018/09/23/apples-iphone-xs-is-faster-than-an-imac-pro-on-the-speedometer-2-0-javascript-benchmark/ https://macdailynews.com/2018/09/23/apples-iphone-xs-is-fast...
      - Wowfunhappy 7y ago
        Apple's newest CPUs have hardware explicitly for accelerating Javascript, so it's not surprising they'd pull ahead there. And as you said, you're comparing the top-of-the-line iPhone to an "average" CPU.
        saagarjha 7y ago
        > Apple's newest CPUs have hardware explicitly for accelerating Javascript They really don't. A12 added a couple of instructions for floating point conversions, but contrary to claims making rounds on Twitter at the time, they were not even generated by WebKit when the benchmarks were run.
      - spronkey 7y ago
        They can compete in certain workloads. As a computational tool however, desktop Intel CPUs can be optimized far, far beyond the capabilities of any A-series CPU. Don't forget that Intel CPUs have things that A CPUs are missing like QuickSync, AVX2, massive PCIe interconnectivity. Whether the A-series CPU could be modified into something competitive on that front is yet to be seen. Whether this actually matters considering the state of our compilers and software development is yet another question.
  - tntn 7y ago
    Right, but Intel isn't falling behind AMD or apple in this comparison - it is falling behind TSMC.
  - rbanffy 7y ago
    My simplest laptop in current use has 4 times more memory than my current phone and I probably would need to make huge compromises to live with half as much. A lot of the chips in phones don't even have external memory buses. A top-of-the-line iPad Pro sports an 8-core asymmetrical core design, with 4 fast cores and 4 slow ones and, overall, is slower than a 2-core Core M-based MacBook (although it feels great because iOS does a lot less than macOS). Also, Apple doesn't make its own A-series processors - it uses TSMC for that.
- atq2119 7y ago
  If the speculation about AMD's Rome / EPYC 2 performance is true, they have now surpassed Intel.
  - Filligree 7y ago
    More accurately, AMD's 7nm processors should be ahead of Intel's 14nm processors. That's great, but it's not a huge surprise. We've yet to see how competitive they'll be once Intel leapfrogs that 10nm node. That's assuming that they can, of course...
- rbanffy 7y ago
  Intel made one single bad bet - their 10 nm process didn't work as well as they expected - and TSMC, who made the right bet, leapfrogged them. In terms of architecture and vulnerabilities, it's not prudent to bet Intel chips are more vulnerable to exploits than others - it's just that we know more about those vulnerabilities. If you want to find vulnerabilities with high impact in cloud and enterprise data centers, Intel Xeon CPUs will be your primary research target.
  - spronkey 7y ago
    We're not really sure yet whether TSMC have leapfrogged Intel in the longer term though. Intel's 10nm issues seem to have delayed their smaller process nodes in the medium term, but by how much is yet to be seen. It seems, for example, that Intel 7nm isn't in quite as much trouble as one might expect. It's also naive to dismiss the possibility for Intel to have learnt a lot from some of the failures in 10nm that will prove useful in accelerating node development in the future.
creato 7y ago
Are they doing worse? Or are they still ahead, just not as far ahead as they used to be? I wouldn't expect any massive leads in any industry to last for long. This might just be regression to the mean.
- vbezhenar 7y ago
  They are behind. I think they have a chance to catch up with their 7nm which is supposed to be better than TSMC 7nm. But it won’t be soon.
  - jjeaff 7y ago
    But what does that really mean? They still seem to be producing the fastest processors available for desktop and server. It doesnt matter if someone else is making even a 3nm chip if the chip still can't outperform the current offerings.
    - dodobirdlord 7y ago
      The sizing numbers are also just nonsense marketing. They stopped meaning anything in particular a long time ago. Intel's '10nm' and TSMC's '7nm' are about the same size.
      - Filligree 7y ago
        The size numbers do measure something, but what they're measuring differs between manufacturers. They aren't comparable between AMD and Intel. They absolutely are comparable between Intel and Intel.
    - shaklee3 7y ago
      I wouldn't say that. Why are Epycs worse than Skylake xeons?
- robdachshund 7y ago
  The reason they are having problems is that they just continued doing die shrinks and speculation hacks to increase performance. They've essentially had the same core since sandy bridge. They didn't see zen coming, didn't have to compete with bulldozer, and thought they could just keep shrinking rather than building a new core design. Once they hit 10nm, they failed, and their old core got some healthy competition from zen. Now AMD is looking to take a serious lead with zen 2, aka ryzen 3000 series. I don't think Moore's law is dead, Intel just gave up on real r&d because it was cheaper. I can't wait for arm and risc v to enter the playing field.
ChuckMcM 7y ago
Oddly enough, the challenge of estimating who is "ahead" is kind of like traffic. Intel arrived at the scaling traffic jam way before anyone else, and has been slowly slogging through it. New entrants are catching up to the traffic jam and will have to make their way through it as well. If there is no breakthrough, then everyone will find themselves more tightly bunched in feature/performance curves than they have been in the past. The spoiler though is that different architectures have different scaling properties and limitations. IBM's Power architecture has already scaled past where Intel is, not because of the semiconductor process, but because the architecture is more streamlined. ARM is somewhere in the middle, it started off pretty streamlined but it has been adding warts (special instructions) to more directly compete with Intel and that creates impediments to scaling.
- IgorPartola 7y ago
  Bad analogy. You can prove that getting in line earliest will get you out of it earliest. If you postulate that it’s more complex than that, it might hold up. You could say that Intel is driving a semi, while others are mini coopers and motorcycles, splitting lanes and better at speeding up/slowing down. At which point no analogy is necessary: startups and smaller companies are more nimble than larger companies, at the trade off of resources.
  - dfrage 7y ago
    It's a good analogy as long as Intel was the first to experience the end of Dennard scaling (https://en.wikipedia.org/wiki/Dennard_scaling https://en.wikipedia.org/wiki/Dennard_scaling) because their fab lines were ahead of the rest of the industry's. And fabs are all "semis", due to the massive amounts of capital and talent needed to move to the next node. So much so that we're now we're down to two companies in the whole world who are successfully executing the smallest CPU nodes, unless Intel manages to make their "10nm" work, or pulls off their "7nm". While we're hearing the very roughly equivalent TSMC "5nm" node is starting risk production (https://wccftech.com/tsmc-5nm-production-euv/ https://wccftech.com/tsmc-5nm-production-euv/ beta testing, you might say, someone outside of TSMC has to be the first, second, etc. to try to get real world dies that work on a new node). Intel isn't saying anything, but Semiaccurate has reported at least two fab lines that were slated to move to their 10nm are installing lots of EUV equipment consistent with using them for their 7nm node (and at least one fab moving back to 14nm).
tntn 7y ago
Apple and AMD just have to ask TSMC to do their magic to make 7nm chips - they haven't had to do anything spectacular, just use TSMCs design libraries. Intel is struggling because of their struggles with 10nm. Apple and AMD are not because TSMC has pulled off 7nm. Architecture matters, but process node matters a lot too.
- lawrenceyan 7y ago
  What would you say are the primary differences between the two companies? Is it more just a matter of luck that has allowed for TSMC to have been able to succeed where Intel hasn't? Or is there actually a meaningfully different process design and/or problem solving approach that is enabling this?
  - dfrage 7y ago
    Not tntn, but I've been following this and it seems to be both Intel's now decades long history of very bad high level engineering and personal management catching up with their crown jewel, and being more aggressive than TSMC's initial 7nm node. Perhaps Intel depending on a particular lithography? technique that TSMC isn't, or isn't yet heavily, but we don't really know, no one authoritative is talking, and Intel is still claiming 10nm is going to make it.
  - wtallis 7y ago
    The full story on how Intel managed to fuck up 10nm so badly may not see the light of day for years if ever. But generally, it seems that Intel tried to make too many changes in one generation. They probably wanted their 10nm to be the most advanced process that didn't require EUV lithography. Some features of their 10nm process ended up not working (evidence points to the cobalt interconnects as one of the hang-ups). In the meantime, it looks like EUV is coming along nicely. They compounded their problems by essentially stopping microarchitecture development on 14nm, which is why eg. their laptop processors still don't support LPDDR4, and they're still shipping basically the same CPU core they released in 2015. Coupling microarchitecture and fabrication development has at times been an advantage for Intel, but for the past few years it's been a huge mistake, and they've promised changes to their design processes so that they don't get stuck like this again in the future if fab advances aren't ready when new microarchitectures are. TSMC naturally doesn't have this problem, because they're a pure play foundry. Their customers have to each make their own bets on when new fab processes will be truly ready, and how well they will perform in practice.

dgacmu 7y ago

Intel was ahead, and hit the wall first. Apple & AMD are not ahead, they're just catching up. I don't want to understate how big a problem that could be for Intel, of course. But they're also doing it on low margin parts, and Intel continues to make bank with their data center parts. I don't think any of this represents a short-term problem for Intel, other than the general downturn in processor sales because fewer people will need to upgrade. But I think it represents a very serious long-term threat. They have some really cool technical advances, like 3D xpoint. But I'm concerned that they do so badly on embedded and custom integration from a long-term perspective.

vbezhenar 7y ago
Apple sold millions of iPhones with 7nm chips while Intel struggles to build comparable 10nm chips and keeps releasing 14+++ nm. AMD will release 7nm chips very soon. It does not seem like they are catching up. Quite the opposite.
- akvadrako 7y ago
  You can’t compare nm between vendors - it’s just marketing numbers.
  - throwaway2048 7y ago
    Not directly no, but the actual feature size and density of TSMC 7nm and Intel 10nm are comparable.
    - pkaye 7y ago
      What about die size?
      - throwaway2048 7y ago
        Die size is independent of process size, customers can order most any die size they want.
      - wtallis 7y ago
        Then you have to ensure you're comparing chips designed for the same market segment. Die size comparisons work well if you're talking about a Cortex-A53 on 16nm vs 12nm. It doesn't work as well when you're talking about a full SoC, or even a desktop CPU+GPU combo where core counts for both sides of the chip can vary greatly.
- Wowfunhappy 7y ago
  And iPhones still don't come close to competing with desktop-class processors in terms of performance. iPhones also use much less electricity, of course, but the point remains. I don't know enough about this, but the GP's argument of "Intel hit the wall first because they were the first to reach that level of performance" makes logical sense to me.
  - robocat 7y ago
    "Apple’s iPhone Xs is faster than an iMac Pro on the Speedometer 2.0 JavaScript benchmark" Yes, Safari, but iPhones do compete with your average desktop processor (not the top end). https://macdailynews.com/2018/09/23/apples-iphone-xs-is-faster-than-an-imac-pro-on-the-speedometer-2-0-javascript-benchmark/ https://macdailynews.com/2018/09/23/apples-iphone-xs-is-fast...
    - Wowfunhappy 7y ago
      Apple's newest CPUs have hardware explicitly for accelerating Javascript, so it's not surprising they'd pull ahead there. And as you said, you're comparing the top-of-the-line iPhone to an "average" CPU.
      - saagarjha 7y ago
        > Apple's newest CPUs have hardware explicitly for accelerating Javascript They really don't. A12 added a couple of instructions for floating point conversions, but contrary to claims making rounds on Twitter at the time, they were not even generated by WebKit when the benchmarks were run.
    - spronkey 7y ago
      They can compete in certain workloads. As a computational tool however, desktop Intel CPUs can be optimized far, far beyond the capabilities of any A-series CPU. Don't forget that Intel CPUs have things that A CPUs are missing like QuickSync, AVX2, massive PCIe interconnectivity. Whether the A-series CPU could be modified into something competitive on that front is yet to be seen. Whether this actually matters considering the state of our compilers and software development is yet another question.
- tntn 7y ago
  Right, but Intel isn't falling behind AMD or apple in this comparison - it is falling behind TSMC.
- rbanffy 7y ago
  My simplest laptop in current use has 4 times more memory than my current phone and I probably would need to make huge compromises to live with half as much. A lot of the chips in phones don't even have external memory buses. A top-of-the-line iPad Pro sports an 8-core asymmetrical core design, with 4 fast cores and 4 slow ones and, overall, is slower than a 2-core Core M-based MacBook (although it feels great because iOS does a lot less than macOS). Also, Apple doesn't make its own A-series processors - it uses TSMC for that.
atq2119 7y ago
If the speculation about AMD's Rome / EPYC 2 performance is true, they have now surpassed Intel.
- Filligree 7y ago
  More accurately, AMD's 7nm processors should be ahead of Intel's 14nm processors. That's great, but it's not a huge surprise. We've yet to see how competitive they'll be once Intel leapfrogs that 10nm node. That's assuming that they can, of course...
rbanffy 7y ago
Intel made one single bad bet - their 10 nm process didn't work as well as they expected - and TSMC, who made the right bet, leapfrogged them. In terms of architecture and vulnerabilities, it's not prudent to bet Intel chips are more vulnerable to exploits than others - it's just that we know more about those vulnerabilities. If you want to find vulnerabilities with high impact in cloud and enterprise data centers, Intel Xeon CPUs will be your primary research target.
- spronkey 7y ago
  We're not really sure yet whether TSMC have leapfrogged Intel in the longer term though. Intel's 10nm issues seem to have delayed their smaller process nodes in the medium term, but by how much is yet to be seen. It seems, for example, that Intel 7nm isn't in quite as much trouble as one might expect. It's also naive to dismiss the possibility for Intel to have learnt a lot from some of the failures in 10nm that will prove useful in accelerating node development in the future.

creato 7y ago

Are they doing worse? Or are they still ahead, just not as far ahead as they used to be? I wouldn't expect any massive leads in any industry to last for long. This might just be regression to the mean.

vbezhenar 7y ago
They are behind. I think they have a chance to catch up with their 7nm which is supposed to be better than TSMC 7nm. But it won’t be soon.
- jjeaff 7y ago
  But what does that really mean? They still seem to be producing the fastest processors available for desktop and server. It doesnt matter if someone else is making even a 3nm chip if the chip still can't outperform the current offerings.
  - dodobirdlord 7y ago
    The sizing numbers are also just nonsense marketing. They stopped meaning anything in particular a long time ago. Intel's '10nm' and TSMC's '7nm' are about the same size.
    - Filligree 7y ago
      The size numbers do measure something, but what they're measuring differs between manufacturers. They aren't comparable between AMD and Intel. They absolutely are comparable between Intel and Intel.
  - shaklee3 7y ago
    I wouldn't say that. Why are Epycs worse than Skylake xeons?
robdachshund 7y ago
The reason they are having problems is that they just continued doing die shrinks and speculation hacks to increase performance. They've essentially had the same core since sandy bridge. They didn't see zen coming, didn't have to compete with bulldozer, and thought they could just keep shrinking rather than building a new core design. Once they hit 10nm, they failed, and their old core got some healthy competition from zen. Now AMD is looking to take a serious lead with zen 2, aka ryzen 3000 series. I don't think Moore's law is dead, Intel just gave up on real r&d because it was cheaper. I can't wait for arm and risc v to enter the playing field.

ChuckMcM 7y ago

Oddly enough, the challenge of estimating who is "ahead" is kind of like traffic. Intel arrived at the scaling traffic jam way before anyone else, and has been slowly slogging through it. New entrants are catching up to the traffic jam and will have to make their way through it as well. If there is no breakthrough, then everyone will find themselves more tightly bunched in feature/performance curves than they have been in the past. The spoiler though is that different architectures have different scaling properties and limitations. IBM's Power architecture has already scaled past where Intel is, not because of the semiconductor process, but because the architecture is more streamlined. ARM is somewhere in the middle, it started off pretty streamlined but it has been adding warts (special instructions) to more directly compete with Intel and that creates impediments to scaling.

IgorPartola 7y ago
Bad analogy. You can prove that getting in line earliest will get you out of it earliest. If you postulate that it’s more complex than that, it might hold up. You could say that Intel is driving a semi, while others are mini coopers and motorcycles, splitting lanes and better at speeding up/slowing down. At which point no analogy is necessary: startups and smaller companies are more nimble than larger companies, at the trade off of resources.
- dfrage 7y ago
  It's a good analogy as long as Intel was the first to experience the end of Dennard scaling (https://en.wikipedia.org/wiki/Dennard_scaling https://en.wikipedia.org/wiki/Dennard_scaling) because their fab lines were ahead of the rest of the industry's. And fabs are all "semis", due to the massive amounts of capital and talent needed to move to the next node. So much so that we're now we're down to two companies in the whole world who are successfully executing the smallest CPU nodes, unless Intel manages to make their "10nm" work, or pulls off their "7nm". While we're hearing the very roughly equivalent TSMC "5nm" node is starting risk production (https://wccftech.com/tsmc-5nm-production-euv/ https://wccftech.com/tsmc-5nm-production-euv/ beta testing, you might say, someone outside of TSMC has to be the first, second, etc. to try to get real world dies that work on a new node). Intel isn't saying anything, but Semiaccurate has reported at least two fab lines that were slated to move to their 10nm are installing lots of EUV equipment consistent with using them for their 7nm node (and at least one fab moving back to 14nm).

tntn 7y ago

Apple and AMD just have to ask TSMC to do their magic to make 7nm chips - they haven't had to do anything spectacular, just use TSMCs design libraries. Intel is struggling because of their struggles with 10nm. Apple and AMD are not because TSMC has pulled off 7nm. Architecture matters, but process node matters a lot too.

lawrenceyan 7y ago
What would you say are the primary differences between the two companies? Is it more just a matter of luck that has allowed for TSMC to have been able to succeed where Intel hasn't? Or is there actually a meaningfully different process design and/or problem solving approach that is enabling this?
- dfrage 7y ago
  Not tntn, but I've been following this and it seems to be both Intel's now decades long history of very bad high level engineering and personal management catching up with their crown jewel, and being more aggressive than TSMC's initial 7nm node. Perhaps Intel depending on a particular lithography? technique that TSMC isn't, or isn't yet heavily, but we don't really know, no one authoritative is talking, and Intel is still claiming 10nm is going to make it.
- wtallis 7y ago
  The full story on how Intel managed to fuck up 10nm so badly may not see the light of day for years if ever. But generally, it seems that Intel tried to make too many changes in one generation. They probably wanted their 10nm to be the most advanced process that didn't require EUV lithography. Some features of their 10nm process ended up not working (evidence points to the cobalt interconnects as one of the hang-ups). In the meantime, it looks like EUV is coming along nicely. They compounded their problems by essentially stopping microarchitecture development on 14nm, which is why eg. their laptop processors still don't support LPDDR4, and they're still shipping basically the same CPU core they released in 2015. Coupling microarchitecture and fabrication development has at times been an advantage for Intel, but for the past few years it's been a huge mistake, and they've promised changes to their design processes so that they don't get stuck like this again in the future if fab advances aren't ready when new microarchitectures are. TSMC naturally doesn't have this problem, because they're a pure play foundry. Their customers have to each make their own bets on when new fab processes will be truly ready, and how well they will perform in practice.