Slashdot Mirror
Does Moore's Law Help or Hinder the PC Industry?
An anonymous reader writes to mention two analysts recently examined Moore's Law and its effect on the computer industry. "One of the things both men did agree on was that Moore's Law is, and has been, an undeniable driving force in the computer industry for close to four decades now. They also agreed that it is plagued by misunderstanding. 'Moore's Law is frequently misquoted, and frequently misrepresented,' noted Gammage. While most people believe it means that you double the speed and the power of processors every 18 to 24 months, that notion is in fact wrong, Gammage said. 'Moore's Law is all about the density...the density of those transistors, and not what we choose to do with it.'"
I suppose it does both.
The drum beat of progress pushes development to it's limits, but at the same time hinders some forms of research or real world tests of computation theory, for all save the few chip makers dominating the market currently.
If only because it keeps us tied to the x86 instruction set. If we didn't have the luxury of increasing the transistor count by an order of magnitude every few years, we'd have to rely on better processor design.
No folly is more costly than the folly of intolerant idealism. - Winston Churchill
It's not a law, it's an observation.
"Moore's Law" is not a real law. In reality, it is not relevant at all. It's kind of a cute thing to mention, but when it gets down to the real world engineering, it has no significances.
If you want news from today, you have to come back tomorrow.
The "speed (capability) increase" is what matters in determining if moore's law is helpful to the computer industry.
the short and sweet.. it is.. it drives sales, provides greater resources allowing expansion of computer capability.
VLC FOR MAC IS DYING! IF YOU DEVELOP, PLEASE SAVE IT!!
If Moore's law is helping or hindering the PC industry? I don't think it could hinder it... Do you think we'd have even more powerful computers without it? Or higher transistor density, if you like?
It certainly seems to have had an effect on peoples attention to writing efficient code. Mind you, it is more expensive to write code than throw more processor at things ...
is more important to nerds than Moore's law anyway. Where's the /. article about it?
Murphy tells us that more bugs will be found on release day than any day previous. That your laptop will work fine until the very minute your presentation is scheduled to begin. And that backup generators are unnecessary unless you don't have them.
Who cares about Moore's law... it's just prophecy from some Nostradamus wannabe.
Sometimes the best solution is to stop wasting time looking for an easy solution.
"Trolls they were, but filled with the evil will of their master: a fell race..." -- J.R.R. Tolkien on Olog-hai
I always viewed this as an observation or rule of thumb, not a law.
Moore (or Mead for that matter) didn't get up one day and declare that the amount of transistors on a square centimeter of space will double every 18 to 24 months. Nor did he prove in anyway that it has always been this way and will always be this way.
He made observations and these observations happen to have held true for a relatively long time in the world of computers. Does that make them a law? Definitely not! At some point, the duality that small particles suffer will either stop us dead in our tracks or (in the case of quantum computers) propel us forward much faster than ever thought.
Why debate if a well made observation hurts or hinders the industry when it's the industry doing it to itself?!
My work here is dung.
The core of their argument is that instead of actually delivering same performance at lower prices, Moore's law delivers more performance at same prices. i.e. you can buy Cray-1 level performance for $50, but you can't buy Apple I level performance for $0.001. The second level of their argument is that this march of performance forces users to keep spending money to upgrade to the latest hardware, just to keep up with the software.
Does Cole's law help or hinder picnics?
Discuss.
- None can love freedom heartily, but good men; the rest love not freedom, but license. -- John Milton
With companies driving to increase transistor density by decreasing process size, the speed we can accurately use these methods is slowing. With each decrease in process size, a lot of issues arise with power leakage. This is where multi-core processors come in. These are the future because of the speed cap of processors. And hopefully this will spur an improvement in microprocessor architecture.
Live life to the fullest. It's not that life is short, but that you are dead for so long.
The real story is that Moore's law describes the basic goal of the semiconductor industry. Perhaps there are better goals, but they tend to get swallowed up in the quest for smaller transistors. The other real story is Gate's law: I will use up those extra transistors faster than you can create them. My hardware OEMs need a bloated OS that will drive new HW replacement cycles. I also seem to remember Moore's law was often quoted as a doubling every year, now I see some saying 18-24 months, so I think in fact the rule is slowing down. We are pushing into the area where it takes a lot of effort and innovation to get a small increase in density. Even still, Moore's law has always been a favorite of mine! Tom
Lots of sloppy, inefficient software out there. (I did not say MSFT.) It gets "rescued" by faster, larger computers. I am not advocating the old days of assembly code, but there is room for better coding.
Why do computers in general need to get any faster these days?
Ten years ago I wouldn't believe I would ever ask such a question but I have been asking it recently as my retired parents are looking to buy a computer for the web, writing letters and emails. I've told them specifically "DO NOT BUY VISTA" (why on earth would anyone want that ugly memory-hog?), so I just can't think of a single reason why they need even one of the medium-spec machines.
Personally, I like my games, so "the faster the better" will probably always be key. But for the vast majority of people what is the point of a high-spec machine?
Surely a decent anti-spyware program is a much better choice.
I've heard that companies plan, design, and release new processors based on Moore's Law. In other words, if it doesn't keep up with Moore's Law it's discarded, if it goes faster than Moore's Law its release is delayed (giving them more time to fine-tune it and get their manufacturing lines ready). If this is the case, then it could be hindering developement in new ways of processing (that have a payoff that takes more than 3 years to develop) and we might even be able to beat Moore's Law rather than follow it. Of course, Moore's Law is awesome enough as it is, I don't feel the need to complain about how it takes two whole years to double the effectiveness of my hardware.
You are reading a copy of my copyrighted post.
In fact there's alot of debate whether Moore's Law will break-down due to fundamental barriers in the physics, or whether we will first hit an economic wall: no bank will be willing (or able?) to fund the fantastically expensive construction of the new technologies.
Cue all the pedantic asshats who absolutely have to point out that Moore's Law really isn't a Law... it's an observation.
Left 404: Why the RIGHT is WRONG
- When the issue of goto was first looked at back in the 60's, they found that it was massive problems due to using gotos INTO blocks. The study found that judicious use of gotos made perfect sense but only out of blocks or within the same block. Now, we are trought that gotos are bad.
- Back in the 70's, a single report spoke that a global cooling COULD happen. Other scientists showed that it was not happening and most likely could not. But the media (Such as national enquirer which was the sensationalist equivilent of Fox news today) got it and ran with it. Now, we deal other media point to that single item and hold it up as proof that science it bad.
The problems occur when ppl who do not understand the issue or a report try to use them.I prefer the "u" in honour as it seems to be missing these days.
So much bullshit about moore's law... On an on people go, talking about how "we can't keep up with moore's law" well WHO THE FUCK CARES ABOUT THAT?? Stop talking about it like it's some holy edict passed down from the heavens that MUST be followed or there will be DIRE CONSEQUENCES! It doesn't matter! If they'd stop with their pissing contest to see "who can best keep up with moore's law" maybe we'd have some decent asynchronous chips by now.
It wasn't even supposed to be a law! It was just a guy noting the pattern in chip density! The person worst off is moore, because he has to have this bullshit repeated with his name on it all the time.
No point to this flame. I'm just sick of hearing about it.
So has chip performance doubled every 18 months?
I have tried to find out, but didn't get a clear enough answer from what is publicly available on the internet.
When his defense asked, "Which computer has Jon Johansen trespassed upon?" the answer was: "His own."
Unfortunately the performance benefits with these new technology nodes are often not enough to justify the increased power.
Add to this the fact that smaller transistors are harder to print correctly, and you see yield margins taking a huge hit.
Of course, there should be progress, but preferably without the forced "deadline" imposed by Moore's law.
Now, more than ever, it has much more to do with $$$$$$$$$ than technology. The cost of a next generation fab is between 3-5 Billion dollars. That's Billion with a capital B. A single mask set for your new design has been typically double for every new design rule node. We are currently way above 1 Million for a mask set going to 2-3 Million. Economics are driving next generation technology more than anything and will continue to do so.
"TV, a medium as it is neither rare nor well done." Ernie Kovacs
Stop writing that.
Like Murphy's Law, Moore's Law is a heuristic rule of engineering. "In general, the computing power of a commercially available CPU doubles every 18 months." Trying to define this specifically in terms of number of transistors, MIPS, processor speed, etc. is silly. The specific technological advances that drive Moore's Law are diverse, driven both by incremental improvements in existing technologies, such as shrinking die sizes in chip fabs, and occasional leaps of innovation, like multi-core CPUs. Representing them as a smoothly increasing exponential function is a massive oversimplification for the benefit of laypersons and Wall Street.
** Check out free Windows OpenGL screensavers at http://www.mounthamill.com/ **
...Like GHz or lines of code.
Take the Itanic for example, or the P4, or WindowsME/Vista.
If you use Windows, each new installation (or daemonic posession) in your computer negates whatever gains you may have had gotten through Moore's Law.
Computers only 12 months old with a _gigabye_(!) of RAM are not robust enough to run a full install of Vista with all the bells and whistles, for example.
--
BMO
Yeah, mod me troll, bitch. I've got more karma than you've got mod points.
The problem is not moores law. The problem is that as processor power increases, the tools to use that processor power get more and more bloated and useless.
Windows vista on modern hardware boots no faster then 98 did on hardware that was modern for its era. The processors are clearly way more powerful today, but all the bloated code that comes along with these processors mean applications are no faster then they were 10 years ago. After all, when you are under a deadline, it's way easier to write ineffient code and make it work by brute proccessing force available to you, then it is to write elegant code and debug it before you get fired for missing your deadlines.
Lets not forget the epedemic of bad programmers out there too. The ones who have no idea how PC's actually work, want to code EVERYTHING in java because its the only language they managed to learn part of by going to night school. This sort lack of knowledge does tend to foster the ability to jam a square peg into a round hole in the most creative of manners, but it also creates some of the worst code imaginable.
Back in the days of DOS, you could do with a 2 megabyte program and 1 mb ram, what today takes a 200mb program and 50 megs of ram. THIS IS NOT PROGRESS.
Help me out here, where is the correlation? I feel that Moore's Law effects the computer industry about as much as Rose O'Donnell leaving "The View" does.
That would be a resounding "Yes!"
Despite this, there have been complaints from the PC industry that Vista isn't enough of a resource hog to force people to buy new hardware.
Computers have become cheaper. I once paid $6000 for a high-end PC to run Softimage|3D. The machine after that was $2000. The machine after that was $600.
And yes it is severely misused by people like Ray Kurzweil KurzweilAI.net who extend it and use it to predict that by 2020 (that's only 13 years from now!) computers will be thinking for us. I'd be happy if computers run Vista smoothly by then instead.
As the island of our knowledge grows, so does the shore of our ignorance.
Moore's law isn't a law. It's an observation.
If transistor density suddenly started doubling every day or only every decade, no "law" would be broken.
Asking "does it help or hinder the industry" isn't so much the right question.
The right question is:
Does the expectation that past results will predict future performance hurt or help the industry?
Maybe it helps by driving research to "meet or beat" expectations, creating a self-fulfilling prophecy.
Maybe it hurts by sending research money elsewhere on the grounds that "it's going to happen anyways."
It might also hurt by driving money to research into "meeting expectations" about future transistor density when those expectations become unrealistic and the money would be better spent on other areas of research such as heat dissipation or other approaches to the "size problem" besides transistor-density.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
Moore's Law is a law like the Law of Gravity. Newton's formulation of the LoG didn't cause gravity to start operating. Newton simply made a mathematical expression of observed reality.
Designers aren't sitting around making sure their gate counts conform to Moore's Law. They design the fastest, highest-performance silicon they can.
I get really PO'd at references to Moore's Law being a determiner of things as opposed to an observer.
Moore's law doesn't drive anything.
Moore's law doesn't affect the computer industry, it is an effect of the computer industry.
It's just a law for chrissakes. It doesn't help, it doesn't hinder. Under certain circumstances it holds, as every sane law would. Eventually someone would need to update it according to changes in the circumstances. Anyway, just a a reminder, Moore was talking about the number of transistors, which has more or less been OK up to now. But thing is, Moore didn't just invent a law out of thin air and the industry followed that rule [I hope you feel the stupidity in that], but he observed how the technology evolves and said it probably will keep going this road [i.e. the repetitive approx. doubling in transistor count]. He was no fortune teller, and as nobody else, he also couldn't foresee technological evolution. Thus, one day, inevitably, Moore's law will be history. Until that, please dumb a bag of bricks on everyone's head who asks whether such a law would hinder or help.
I am putting myself to the fullest possible use, which is all I can think that any conscious entity can ever hope to do.
What we do with the transistors? Run software of course. Enter Wirth's Law:
"Software is decelerating faster than hardware is accelerating."
More Twoson than Cupertino
186,000 miles per second, That's a law.
Use your head, can't you, use your head,
You're on earth, there's no cure for that - S. Beckett
GP is absolutely correct; one interesting question we might ask, however, is whether Moore's "Law"/Observation has actually "driven" development that wouldn't have existed otherwise. That is, has the mere existence of Moore's Law resulted in it growing legs at any stage and actually *driving* the changes it supposedly just observed?
"Slashdot - News and Chat Sites Deviant". (Click "homepage" link above for details).
Marketers, execs, and pundits have stolen ownership of Moore's law. What none of them understand is the computer geek culture that the "law" was spawned from. Moore's law is in the same realm as Murphy's law, which is also not a law, but fun to invoke.
However, the paid talking head pundits grab it and start talking about it and dissecting it and taking it literally. It's not a topic for geeks any more, it's not funny, and it's stupid to be discussing it in an article.
I propose a real law. A legal law. A law that states if the editors post another stupid article about Moore's law, the entire slashdot community is allowed to line up and each get to spank all of the editors one by one.
Except for CowboyNeal... he might like that... his punishment is he has to watch.
"All great wisdom is contained in .signature files"
I'm just going to refer you to my comment made earlier today when discussing a "new, better" processor architecture. Because there's always someone who thinks we are somehow "hindered" by the fact that we can still run 30-year old software unmodified on new hardware.
See here.
They submit "$module is too slow" bugs. I do nothing for 18 months and then resolve it saying, "it is fast enough now".
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
One valid argument is the frustration of having to upgrade hardware to get acceptable performance on newer applications. I can empathize, but what's the alternative? You've got three choices, really:
- Upgrade hardware to run newer software
- Arbitrarily stifle software innovation so that you don't have to upgrade every 12 months
- Increase development time so that developers can wring every ounce of efficiency or shave down a footprint*
In my opinion, the the first option lesser of three evils.I agree. Still, let's extrapolate to some sort of conclusion. Do we tell hardware vendors to hold off until The Next Big Thing warrants better hardware? What if The Next Big Thing is out of the question on current hardware?
Software innovation prompts hardware innovation, and hardware innovation prompts software innovation.
Consider folding@home. Developers said, "hey we've got these GPU's that would be perfect, lets use them". Now GPU clients are some of the top performers. Supposing perfected 2D VGA videocards and said "well, that'll do", such an innovation would be impossible.
*I'm sure that someone will reply with "Well this wouldn't be a problem if people didn't write such sloppy code!". Yes, sure sure. People wrote much better code back in the day and you had to walk 5 miles in the snow just to get to school and all that. Whatever. Even if all coders were brilliant, I would still prefer to have their brilliant minds focused on new features than on fitting code into hardware constraints.
Moore's Law is a huge help for technical computing. Anytime you need to crunch numbers for something, it's a good bet that you'll have more processing power in next year's hardware. This gets us closer to really important breakthroughs in science and technology.
It's a monster hindrance for mainstream computing. Having all this processing power available to you, coupled with cheap memory, means you can be as lazy as you want when you write software. I do systems integration work for a large company, and the bloated, inefficient garbage that runs fine given enough hardware is mind-boggling. I may seem a little bitter, but it seems like apps written for internal use only survive due to pumped up memory specs. I'm not saying you need to do funky tricks to squeeze a program into 4K anymore, but at least optimize code so you're not doing crazy things like iterating through each row in a database table, etc.
Another subtle problem is that it causes inflation to be underestimated, because the Bureau of Labor Statistics figure that a computer that is five times as "powerful" is worth more, even if you personally are not doing significantly more with it than the computer you bought five years ago.
"How to Do Nothing," kids activities, back in print!
It contributes to the production and distribution of really bad code. Firefox with tens of millions of copies is a case in point. (Oh yes, they *claim* with version 3 they are going to consider performance). I'm still waiting for Firefox to run in the same memory footprint and as fast as Netscape 4.72 did. (Firefox will not start with less than ~55MB of memory under Linux.)
When excessive amounts of memory and processor speeds allow you to release software which by any stretch of the imagination is "bloatware" (could you do the same job with significantly less memory and processor utilization? I strongly suspect so...) then the hardware capabilities is facilitating really "dumb" development processes.
Its like putting an AK-47's (with 300+ rounds) in the hands of people who are hardly qualified to operate pocket knives.
As Forrest was prone to observe, "Stupid is as stupid does." Abusing the CPU or memory capacity at ones disposal is not something I would want tacked onto my resume.
Increased transistor count can be used for higher volumes, greater performance, or bigger cache.
Up until a few years ago, more performance and memory resulted in a distinct return on investment. Right now, most machines are "good enough" for present apps. I predict a shift to system on a chip designs driving small reasonably powerful systems like the OLPC.
The problem is the industry adapting to this new model.
Take it easy? I'll take it anyway I can get it . . .
It's a law just as much as the law of supply and demand and the law of universal gravity are. Only mathematical laws are deductively proven.
[sig]
Just an after the fact observation, which yes, has held approximately true until now. But its NOT A LAW and stop treating it as such !!
A million monkeys and this is the best sig they could come up with...
Moore's Law has never been specifically about the density of transistors. Moore's Law has always been about the total number of transistors per chip. Moore's Law says that the total number of transistors per chip will increase at certain rate, bit it never said that the increase will be implemented specifically through increasing density. Density is just one factor in the equation. Another major factor is the sheer area of the chip. The total number of the transistors can obviously be increased by increasing the area of the chip.
In the past, the density was the dominating factor. It was dominating do much that we could satisfy the Moore's Law and at the same time actually decrease the size of the chip. However, further increases in density are no longer possible. We have already hit the limit for the traditional technology. Density is no longer the dominating factor. <i>Area</i> is. The Moore's Law continues to be satisfied today by increases in the area of the chip. Chips become bigger. A lot bigger. An this is the direction the industry will continue to move in for a while, until the next density-related breakthrough comes over (like 3D designs, optical chips etc.)
Herb Grosch's Law - the way I have always heard it; No matter how much the hardware speeds up, the software will piss it a way.
The only reason customers buy new computers is because of the chicken and egg of Moore's law and code bloat. If computers stayed the same, there would be no code bloat and nobody would buy new computers. The profits of the industry pay for research which furthers Moore's law which allows code bloat which causes people to buy computers.
We can debate if it helps or hurts the hardware industry, but I don't think any one can deny that it helps the software industy...
And "640K ought to be enough for anybody"!
But I do understand your point. Almost all machines being sold today fit most people's needs (email, web browsing, word processing, file management) and every time I get asked the question "what do I need to look for in a new computer" I almost always say almost any machine available now is capable of doing everything they need.
However, that view is a short-sighted; while in the past people were writing up documents without graphical interfaces, people also were not able to use their computers for any other tasks. Today, some sort of multimedia capability is expected and standard; if the picture doesn't load instantly and in full 24bit color I'm pissed. Same for sounds and videos and presentations.
Now we've hit a point where we're not sure what we're going to do with all this extra processing speed other than a few niche areas like games, video processing, and number crunching and I think the industry is reacting and trying to find alternatives or other design constraints that better meet the consumers needs. Take for example batteries and efficiency: many companies in the computer industry have made it a new priority to maximize battery life instead of performance. Also look at the progress of laptops and how much closer they are getting to desktops in terms of cost and performance.
So to answer your question, while I don't think there is an immediate need for faster computers for everyone, I do believe there are other ways to apply our technology advances to meet other requirements other than performance (think convenience).
Just because the hardware gets faster and faster and CPU buses, networks and graphics cards change and improve peple think that everything has changed and that the old school rules don't apply. I'm thinking about team organizations and how one produces a product. There are lessons we learned a hundred years ago during the industrial revolution that I continually find people relearning today. About making changes as the last minute and quality assurance. Information technology allows some parts of managing an organization to react quicker and better than we could in the past but there are something about producing a model T and producing a Firefox extension that are the same. For example, understanding what your customers want and testing it for quality. The rapid pace of change in computer hardware enables people forget that not all the rules have changed.
The correct spelling is definitely.
30 year old software won't run on new operating systems, and 30 year old software won't support modern hardware. You won't be getting USB support, and parallel/serial ports are quickly disappearing. Where would you find a modem with drivers for your old OS? Where would you find a dial up ISP, let alone one that would support 1200 baud or whatever you'd be limited to.
You're going to be *far* better off running 30 year old software under emulation, where these things can be faked.
I disagree that PC's are reaching a point of stability, but even if they were, it wouldn't matter. We'd just find new sorts of computers to build. Like smartphones, in dash navigation systems, media players and such.
100% of observed technological species seem to follow Moore's law. Further, expanding it to a more general form, ie "Technological development follows an exponential curve." is also valid for 100% of observed technological species.
An exponential increase on 0 is still 0.
Here is some background reading for you.
You can get more than Apple I level performance for substantially less than $1.00 these days. I don't know what the cheapest 8-bit microprocessor is going for these days, but you can buy flash-based microcontrollers with substantially more power than a 6502 for less than $0.50 each these days.
It seems like the ultimate limiting factor is in packaging and testing - you'll be spending a certain amount for a fully-tested chip in a plastic package, no matter what the actual chip is. That price will have more to do with the number of pins on the package than the number of transistors on the chip.
In really cost-sensitive applications, they already do away with the packaging, and just glue the chip directly to the PCB. Nearly every electronic doo-dad you see on the drugstore shelf is built this way.
Excerpts from A Conversation with Gordon Moore: Moore's Law.[PDF]
I don't think Moore ever foresaw a situation where we couldn't USE the transistors to improve single processor power. The letter of Moore's Law is still true, but the spirit is broken.
So what if we can fit 20 processors on a single die, if each one of those processors has reached a functional plateau? Our languages and methods are still largely designed for a single processor.
Gee folks, it might be a good idea to finally shuffle off the mortal coil of the last remnants of the X86 architecture. Otherwise, I suspect SMP multi-cores are going to be very disappointing.
Since the X86 is the legacy he helped pioneer, his "Law" is now irrelevant. The number of transistors, while notable, no longer means the same thing.
--
Toro