Why Intel Insists Rumors Of The Demise Of Moore's Law Are Greatly Exaggerated (fastcompany.com) 106
From an article on FastCompany: Intel hasn't lost its zeal for big leaps in computing, even as it changes the way it introduces new chips, and branches beyond the PC processor into other areas like computer vision and the internet of things. "Number one, too many people have been writing about the end of Moore's law, and we have to correct that misimpression," Mark Bohr, Intel's technology and manufacturing group senior fellow and director of process architecture and integration, says in an interview. "And number two, Intel has developed some pretty compelling technologies ... that not only prove that Moore's law is still alive, but that it's going to continue to provide the best benefits of density, cost performance, and power." But while Moore's law soldiers on, it's no longer associated with the types of performance gains Intel was making 10 to 20 years ago. The practical benefits of Moore's law are not what they used to be. [...] For each new generation of microprocessor, Intel used to adhere to a two-step cycle, called the "tick-tock." The "tick" is where Moore's law takes effect, using a new manufacturing process to shrink the size of each transistor and pack more of them onto a chip. The subsequent "tock" introduces a new microarchitecture, which yields further performance improvements by optimizing how the chip carries out instructions. Intel would typically go through this cycle once every two years. But in recent years, shrinking the size of transistors has become more challenging, and in 2016, Intel made a major change. The latest 14 nm process added a third "optimization" step after the architectural change, with modest performance improvements and new features such as 4K HDR video support. And in January, Intel said it would add a fourth optimization step, stretching the cycle out even further. The move to a 10 nm process won't happen until the second half of 2017, three years after the last "tick," and Intel expects the new four-step process to repeat itself. This "hyper scaling" allows computing power to continue to increase while needing fewer changes in the manufacturing process. If you divide the number of transistors in Intel's current tick by the surface area of two common logic cells, the rate of improvement still equals out to more than double every two years, keeping Moore's law on track. "Yes, they've taken longer, but we've taken bigger steps," Bohr said during his three-hour presentation.
Slash's Law, my eyes! (Score:3, Funny)
Seems the size of the paragraph doubles every 2 stories.
Re: (Score:2)
Seems the size of the paragraph doubles every 2 stories.
While processor speed has been stuck at 4GHz for several years. Adding cores improves performance only while you have one process you can assign in parallel to each core.
Moore's, Ohm's, and other empirical laws (Score:3)
Moore's law is an empirical law [wikipedia.org] in the sense of Ohm's law, which most familiar materials obey but some do not [wikipedia.org].
Re: (Score:2)
no, it is invalid, Moore's law can't be used to predict, while empirical laws are useful for that purpose.
Re: (Score:2)
no it isn't, empirical laws can be made to predict future behavior, for example Ohm's law of the many materials for which it is useful, while Moore's law cannot at all.
1975 law predicted behavior for 4 decades (Score:2)
Moore's law has predicted future exponential growth of integrated circuit density fairly well since it was proposed over four decades ago.
Re: (Score:2)
yes, great for looking backwards
it does not guarantee any such future growth.
ohms law let's me predict current after doubling the area of a copper conductor in the next five minutes or 50 years from now. moore's law might be hogwash in 5 years.
Re: (Score:2)
How well does Ohm's law let you predict whether newly discovered materials will be ohmic?
Re: (Score:2)
but the real Ohm's law of 1827 is for wires of metal (and even included temperature term!). For newly discovered metal alloys it's pretty damn good. Are there metals that don't obey Ohm's law? I'm looking that one up.
Re: (Score:2)
false, LHC has to borrow/rent compute cycles from other labs for processing the volume of data it gets from experiments, and that processing goes on for months after experiments end and that constantly varies, it's dynamic. they don't buy what they need based on moores law
Re: (Score:2)
Moore's Law utterly ignores the fact that even two or three nanometer transistors are practically impossible given bond length combined with atomic diameter for a three-atom transistor made from today's materials, or basically any feasible combination on the periodic table, unless we figure out some way of making transistors out of quarks. It is essentially marketing and typical soft science prediction (much like economics.) And, Moore's Law was never stated as performance doubling, but transistor/component
Re: (Score:2)
3D.
Re: (Score:2)
3D Does absolutely nothing, actually. See, Moore's Law was explicitly implied as to be in the 2D planar space, so you can fuck right off with this bullshit metric of '3D' right now, you ill-educated shit.
Re: (Score:2)
Somebody must have pissed in your wheaties.
Re:Not a law (Score:4, Insightful)
Re: (Score:2)
Where are my mod points when I need them.
Mod parent up.
There needs to be a Slashdot moderator option called "Pedantry Slashing"
Re: (Score:2)
No, you're looking for the count of transistors.
The transistors are placed in an area.
The transistors in a new process take up space space roughly equivalent to [(new feature size)/(old feature size)]^2 * (old space).
Even if you assume the marketing number (10nm vs 14nm vs 16nm vs whatever) are both accurate and representative (they're not, they refer to the minimum feature size), you're still missing the real issue. Intel's main motivation for moving to 10nm isn't to give you more transistors or save you
Re: (Score:3)
Interesting. But let's not forget that Moore's law is about exponential growth of transistor density, not NOT exponential improvement in performance. The transistors are (I believe) being fruitful and multiplying as they have been for four decades. What, if anything, useful is done with them is another issue.
Re: (Score:2)
Interesting. But let's not forget that Moore's law is about exponential growth of transistor density, not NOT exponential improvement in performance. The transistors are (I believe) being fruitful and multiplying as they have been for four decades. What, if anything, useful is done with them is another issue.
Moore's law is actually about the price per transistor. Cost for a die is proportional to area so denser transistors means cheaper transistors but if your transistor density is limited by thermal density, then power per transistor becomes more important than density neatly explaining the recent push for lower power over higher density.
If the cost per transistor does not decrease allowing increased integration, then the fab has nothing to sell.
It's too hard is an excuse (Score:2)
Re: (Score:2)
intel needs to up the number of pci-e lanes (Score:2)
intel needs to up the number of pci-e lanes.
AMD has more.
Hm.. (Score:5, Insightful)
"This "hyper scaling" allows computing power to continue to increase while needing fewer changes in the manufacturing process."
This "hyper scaling" allows Intel to continue to milk customers who expect more than modest gains with every generation.
FTFY
Re: (Score:1)
Go peddle your FUD on infowars, not a reputable site... like... heh, never mind.
Re: (Score:1)
Re: (Score:2)
Re: (Score:1)
So Intels processors are being slowed down because of the spy chip? Citations needed by reputable sources like Wikipedia and Youtube.
Really? You need a citation to explain that two things competing for and obtaining a finite resource will make that resource attainable at less than peak rates? You don't belong here.
Re:Doesn't Keep Up With ME (Score:4, Insightful)
Re: Doesn't Keep Up With ME (Score:2)
Re: (Score:2)
Re: (Score:1)
Re: (Score:2)
So Intels processors are being slowed down because of the spy chip? Citations needed by reputable sources like Wikipedia and Youtube.
There is only so much real estate on any die, what management engine takes, can't be used for things like cache which are well known performance enhancers.
This is something so trivial even non technical people should understand it intuitively, so are you just trolling or is it something else ?
Re: (Score:2)
Re: (Score:2)
Moore or Less Law (Score:5, Insightful)
Number of Transistors may still be increasing in density but computers aren't seeing the revolutionary jumps in power and performance- it's not scaling to us end users. I have a 5 year old PC at home I built, and it rivals most of the mainstream PCs being put out today. Even if Moore's law is still holding true, it's not really relevant anymore.
Computers aren't getting much faster any more. Processors may be getting smaller as transistors density gets higher, but your average home PC isn't getting much better.
Re: (Score:2)
Yes, it has never been a better time to buy a used computer. Scoring a used system with Windows 7 (or earlier) is a bonus.
Re: (Score:2)
I realized recently this situation is happening with laptops: reasonable quality quad core 1920x1080 systems from 4 generations ago are showing up for under $500 US, and at that price Big Box is still wants to sell you a 1366x768 dual core system (in light of how stagnant it's been it almost seems like an accomplishment that they've even gotten past wanting to only put 4GB of RAM in them).
Re: (Score:1)
I think what's happening is they do get twice the performance per area, but it's also twice as expensive per area. Who cares about Moore's law if "performance per dollar" stays the same? Heck, the best value per dollar [cpubenchmark.net] for a processor scoring at least 10000 on PassMark is a processor from 2012.
Re: (Score:1)
Re: (Score:2)
Typically the payback for replacing based upon power savings alone is too long to be worth it. If she leaves her computer on 24/7 then maybe, but you'd realize a much better savings by having it go to sleep when it's not being used. You're usually better off just to keep using what you've got until it either is truly obsolete or breaks down.
Re: (Score:2)
From the 90's, to about 2010 or so, I would upgrade my parents' machine every 2 to 3 years.
Now that they are on an SSD and 16 gig of RAM there is absolutely no reason to.
Re: (Score:2)
Same situation here. I have a machine I built in mid-2011 around the best AMD CPU available (Phenom II x6 1090T, released in 2010 -- couldn't find the 1100T at that moment) and it still runs in the middle of the i5 pack today. Even Ryzen is only a tempter. I don't need more CPU, if the cost is a new motherboard and RAM as well.
Pathetic (Score:3)
Is it just me or does this whole diatribe just ooze "pathetic marketing maneuver"?
It's one thing to admit that things are getting more challenging cause the low-hanging fruit is gone and Intel's having to put more time and effort into their manufacturing, but for the love of Pete, redefining Moore's Law is just lame.
I really wish Apple had a tightly held patent on their reality distortion field cause now everyone else is trying to use it and it's just... cringeworthy.
Re: (Score:2)
For every action there's an equal and opposite reaction. Now that half the population has levelled-up on aggressive ignorance, the standard for those who still pretend to know better only becomes that much higher.
Intel is very much one of those pretenders.
Paging Gordon Moore.
Gordon—dressed exactly like Bob Pinciotti—steps into the Tardis that just materialized outside his office, and grabs the ringing phone—
Re: (Score:2)
You mean despite the poor thermals, weak components, and locked down OS, what you have to complain about most is the lack of more RAM?
Investor relations insist (Score:1)
They claim that emperor still has clothes...
Re: (Score:2)
Seems like a really pathetic response to Ryzen's debut. More cores with nearly equal performance per core for less $$$ is hard to argue with, so they spew this marketing blather.
The 10-20% per year performance increase Intel has been offering is just sad and pathetic. More cores should have been the next step, but they have been slapping huge markups on anything with >4 cores for years. At least now there is some actual competition and they might wake up and start trying again.
Remake of old movie (Score:2)
This is a remake of a 1999 movie. Plot summary: AMD had a chronically weak offering, Intel was in the habit of dribbling out the performance gains. AMD suddenly came on very strong with Athlon, a completely new chip which was arguably faster than Intel and definitely cheaper. Almost overnight, Intel suddenly figured out how to make much faster chips, and so did AMD. Performance doubled, tripled, with AMD being the first to crack the 1GHz barrier the next year. That spiral continued for a few years and the u
Re: (Score:2)
This is a remake of a 1999 movie. Plot summary: AMD had a chronically weak offering, Intel was in the habit of dribbling out the performance gains. AMD suddenly came on very strong with Athlon, a completely new chip which was arguably faster than Intel and definitely cheaper. Almost overnight, Intel suddenly figured out how to make much faster chips, and so did AMD. Performance doubled, tripled, with AMD being the first to crack the 1GHz barrier the next year. That spiral continued for a few years and the users were happy, but AMD ultimately fell behind and Intel went back to their old tick-tock.
Obtaining higher and higher chip performance seems analogous to natural resource extraction. If it is harder and harder to keep getting the same gains, let those gains sit in the ground until they are actually needed.
Re: (Score:2)
More cores should have been the next step, but they have been slapping huge markups on anything with >4 cores for years. At least now there is some actual competition and they might wake up and start trying again.
Until you try running something that isn't optimized for multiple-core support. Then no matter how many cores you have it doesn't help a bit. I'm not arguing that we shouldn't strive for more cores- I'm merely saying that's only one part of the puzzle.
Parallel Processing (Score:2)
Demi Moore (Score:2)
I originally read the title as "Why The Rumors Of Demi Moore's Demise Are Greatly Exaggerated".
Murphy's Law + Moore's Law = Moorephy's Law (Score:2)
Diminishing returns or lack of invesment into R&am (Score:3)
If this is diminishing return on hardware - then next area is software optimization. So far, most of our progress was carried by hardware. This is not going to be popular view among programmers, but default mode of operation is "how much resources do I have, lets use it all" in software engineering. There is no thought given to making it leaner and more efficient, because it used to be that hardware gains over time would make such effort moot. Well, there might not be any more notable gains. We will hit the next nm fab level, get 3D layout process in place and not have a good way to move forward other than occasional specialized optimizations (e.g. AES) acceleration). This might take a form of optional co-processors.
Re:Diminishing returns or lack of invesment into R (Score:2)
Most of this is because of the proliferation of Virtualization and "Cloud computing"
If most of what you do is in a web app or a virtual machine or compatibility layer.
Your never going to see the hardware gains.
Intel processors may be 10X faster than they were 10 years ago but your ISP damn sure isn't.
When basically all programs have to phone home before they do anything, hardware improvements become irrelevant.
My take on Moore's Law (Score:2)
I generally feel that Moore's Law is still well in affect. While Moore's law is just about number of transistor's will double ever 18 months, I take a broader look at this. I see it as processing power will double ever 18 months.
How do I define 'processing power'? That's up to a lot aspects. Obviously, A CPU's is raw CPU clock speed, bogoMIPs, etc. But there are other ways we can game power/efficiency too. Multiple cores and SMT. But also power efficiency should be considered, if we can do more processing p
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
It's not embarrassing. It's a fact. And I said "modern usage". The example they gave (of Edgar Allan Poe) is over 150 years old. Even the erasing of the difference between "shall" and "will" is more modern than the death of Edgar Allan Poe. The difference between "affect" and "effect" (as verbs) is arguably artificial even when both are used correctly. It's meant to emphasize the traditional English-language distinction between subjects and objects of a sentence ("itch" vs "scratch", "lend" vs "borrow
Re: (Score:3)
Moore's Law isn't just dead. It's dead, buried, flowers have been laid, songs have been sung, and the mourners have left.
Re: (Score:2)
Re: (Score:2)
Just because some processing power gains don't apply to you doesn't mean that don't happen or exist.
And you lucky you to live in place that has reliable and cheap power. Not everyone has that luxury. Places in the world that need to have reliable computing with unreliable power have to make sure that have properly scaled generators and power storage. This limits what they can run. If some lower power chip comes out with
the same Ghz, it extends their ability to process.
Possibly, Intel (Score:3)
Re: (Score:2)
Re: (Score:2)
If it's 10 years old, it's almost certainly a DDR2-era machine. If it's an Intel machine, you've still got the memory controller sitting in the Northbridge rather than integrated into the CPU. The Core i3/5/7 machines were a pretty big step up from the Core 2 machines, especially once you get into the 2nd gen Sandy Bridge processors (which performance-wise are still very competitive with Intel's current offerings). Of course, Sandy Bridge is more like 2011, not 2007.
With that said, I'm actually typing th
Re: (Score:2)
I agree. I have an 8 year old desktop Linux PC that plays Netflix, streams youtube, twitch, etc. just fine, and can play 1080p video both x264 and x265 just fine. The x265 stresses the dual core CPU a bit, but no frames lost. It's not my only PC, and it's more of a Linux toy and a backup machine for when my other PCs are busy... but, there's really not much it can't do other than play serious games (though I could upgrade the vid card on the cheap). I could put Win10 on it and it'd likely get more
Entering aircraft industry regime (Score:2)
buh (Score:2)
Why Intel Insists Rumors Of The Demise Of Moore's (Score:2)
Why Intel Insists Rumors Of The Demise Of Moore's Law Are Greatly Exaggerated
Because Intel wants you to buy their CPUs which haven't seen worthwhile improvements in over 5 years.
Unless you need bleeding edge performance, just pay half the money and get a Ryzen CPU.
Re: (Score:2)
No, If you want bleeding edge performance find a Ryzen cpu/mobo/memory combo that lets you boost the memory to 3600 and CPU to 4.1Ghz and remain stable. At which point you will be equal or better in performance to the 7700k @ 5Ghz(Interestingly, memory speed increases beyond 2400 have very little effect on Kaby Lake performance) in games and better than everything else in the other metrics.
Re: (Score:2)
The 7700k isn't the performance king. Some "Extreme" series (6870x or whatever) is. The 7700k is the popular gaming CPU.
Most Ryzen chips can't sustain 5.1 GHz on air. The highspeed memory situation is getting better, but it's still a crapshoot in many cases.
Another issue is that as you crank up the memory speed, increasing the speed of the "infinity fabric" connecting the CCX units, you have more power draw and more heat to deal with, which can actually hurt performance in certain workloads.
For anything
Listening to Intel (Score:1)