Slashdot Mirror
Understanding Moore's Law
S. Blocher writes "Ars Technica has a great article up, 'Understanding Moore's Law', that I think most geeks should read. The misrepresentation of Moore's Law in the media has always been a real pet peeve of mine, and this article does a great job of looking at the flipside of the 'bigger and faster' thesis to show how the Law isn't really just about doubling computer power."
It seems like a good start would be to stop calling it a "law," a term that has some kind of meaning, in a scientific sense. Exactly where that line gets drawn may be a little fuzzy, but I think it's fairly obvious that Moore's observations don't make the grade.
It Is the Nature of Information to Transgress Artificial Boundaries
I'll comment w/o reading the article.
Roger Moore's Law is: you get more chicks when you are James Bond than when you're Roger Moore. That's it right?
Microsoft's law: what you get when you put Moore's law and Murphy's law together.
...just my 2 gil.
For every news artical about Moore's Law, there will be a two fold reduction in the amount of Moore's Law that is explained. Which at some point it becomes impossible to comprehend the difference between Moore's Law and Artical Fodder about why you need a new computer.
Neck_of_the_Woods
#/usr/local/surf/glassy/overhead
The rate at which the Moore's law is quoted in lectures and articles doubles every 18 months
it is a theory, or a hypothesis, or an observation. A law in the scientific jargon must be something which is:
6 a : a statement of an order or relation of phenomena that so far as is known is invariable under the given conditions b : a general relation proved or assumed to hold between mathematical or logical expressions
Moore's Law only holds true to a point. There comes a time when only so much can be fit on that piece of silicon. The term conjecture might also be applicable.
In SOVIET RUSSIA... erm...NSA AMERICA, the Internet logs onto YOU!
Haven't we been beating this dead horse long enough?
Admonitions against misinterpreting Moore's law are about 3 minutes less old than Moore's law itself, and will probably be the part of Moore's law that outlives the law by 20 years.
Moore's Law is so perennially protean because its putative formulator never quite gave it a precise formulation.
I tried to read further, but started twitching uncontrollably. How about Mace's Law: "The skill of tech writers halves, and their pride doubles, every 12 months."
I always gave about the same credence to both Moore's Law and Murphy's Law.
Ok, I finished the article. I learned some history, saw some graphs, and care not one bit more about Moore and his infernal Law.
...
>Wouldn't profit be it's own incentive?
Of course profit is it's own incentive; doesn't take a genius to realize that. What you miss in my question is this "law" based on the fact that instead of continually saturating the market with current product Moore seemed to set a goal for engineers to help that ol' profit incentive.
Life moves pretty fast; if you don't stop and look around once in a while, you could miss it. -FB
6 a : a statement of an order or relation of phenomena that so far as is known is invariable under the given conditions b : a general relation proved or assumed to hold between mathematical or logical expressions
I agree,and I would actually give Moore's Law less credit than that. I would infer from your definition that, generally, a "law" is something fundamental and immutable that arises from the underlying nature of certain phenomena. Boyle's gas law is a good example: increase the temperature of a gas at constant pressure, and it expands. This is fundamental and can be explained as such.
Moore's "law" is just a relationship built from Intel's marketing engine and economics. Let's say I was rich and decided to start marketing cheap, low-defect silicon. Moore's "law" suddenly changes.
Basically, Moore's law could change at any time (and has) if Intel decides to accelerate their R&D facilities. Or if they decide to invest more in silicon fab facilities. Or if they decided to raise their prices, allowing them to get a lower yield of smaller-featured chips.
When you get right down to it, Moore's law only holds as long as Intel wants it too. Or, if they get more competition that forces them to accelerate their chip release schedule (like the last 5 years thanks to AMD).
-Looking for a job as a materials chemist or multivariat
Is to give you an excuse to avoid work. See http://arxiv.org/abs/astro-ph/9912202 for a paper (in PDF) describing this
Amdahl's law is more of a "law" than Moore's law is. It's a way of figuring out how much of a speedup is really possible from parallelization. The formula is:
S=N/((B*N)+(1-B))
where S is the speedup, N is the number of processors, and B is the percentage of the program that must be serially. The upshot is that as long as B is greater than 0, you'll never get the N-times speedup you'd hope for from throwing N processors at the problem.
It seems a pretty trivial result, but it's basically Gene Amdahl's way of throwing the "multiprocessing will save us" theory back in the faces of its proponents. Multiprocessing is the obvious way to speed things up, but if it won't work on every problem, we'll have to look elsewhere.
If you're referring to that recent Red Herring article, my article was indeed "inspired" by it in the sense that I thought it was sensationlistic crap and I just couldn't take it anymore. For more info, see the news blurb that announces the article:
h tm l
http://arstechnica.com/archive/news/1045747027.
Senior CPU Editor | Ars Technica | http://arstechnica.com/
This is due to management overhead, ratio of parallelizable to non-parallelizable portions of the task, etc.
Very important stuff to consider when doing multi-threaded/process/tasking and clustered design and development.
No sig, sorry.
He didn't say just how long his "law" would last, however. He made the huge assumption that intel would keep on keepin on with their technology breakthroughs, and if they perfect that terahertz transistor technology, THEN Moore's law will stay in effect for a little longer
But, something that most people don't take into account (and moore probably didn't either) is the fact that we don't NEED faster computers. We want them. You don't need more than a P4 to do just about anything, but becuase of games that require horsepower, you are required to buy them. Of course theres the server side of things, which is a totally different ballgame. YES, they DO need faster computers as workloads get heavier.
Basically, Amdahl's law says that you can only optimize to a point before it becomes pointless. Also, it is often used to demonstrate that it is useless to optimize a single aspect of a system when the other aspects are still unoptimized.
For example, let's say you have a computer that executes something within 10 seconds. 5 of those seconds are spent reading from disk, 5 are spent doing calculations on the CPU.
Now if you upgrade the disk to a disk twice as good, you're gonna get an execution time of 7.5 seconds(5 for CPU, 2.5 for disk). So you gained 2.5 seconds.
Let's say you still think it's too slow, and upgrade the disk again to a disk twice as good. You're now getting an execution time of 6.25 seconds(5 for CPU, 1.25 for disk). You thus gained 1.25 seconds.
You should get the trend here, if you continue upgrading only the disk, it will come to a point where, even by increasing the speed of the disk tenfold, you will only gain small fractions of second on your execution time, and small fractions of second over more than 5 seconds is definitely not a good improvement.
The same reasonning can be applied(this is the original intent of Amdahl's law) to multi-processor machines. Assuming only 50% of a program is runnable in parrallel processing, it comes to a point where adding processors brings very little improvement, even were you to increase the number of processors tenfold. (The explanation for this is left as an exercise to the reader... Hint: it's the same explanation as the disk/cpu above.)
My new law/observation/theory/detriment/prophesy is:
Every 18 [eighteen] months a new faster and more powerful processor is released to the masses that makes my 1Ghz seem obsolete.
Not a sermon, just a thought.
[n8.r0n] http://petesweb.spymac.net/
Well then they better change the ads. On every page, there's a big animated gif ad for 'American Singles' asking ARE YOU SINGLE? Yes! Of course I'm single! I'm reading this fscking article! Do they think I could get a date? hrmmph.
Damn Marketers.
HURD - Hurd's Under Research & Development
Quite simply, companies expect Moore's law to remain true. Software companies plan product lines in anticipation of processor capability doubling every n years. Processors are going to get improve at this rate, therefore we know how quickly bus performance and peripheral performance need to improve.
Semiconductor manufacturers know this. They plan product lines in a Moore's-law-consistent manner (not necessarily explicity, but surely as a matter of economics). If they're a little behind the curve, more money gets put into keeping up with it... or somebody else steps up and keeps it true. If they're at or ahead, they hold the course.
If someone were to introduce a processor that was 10x the density/speed of current processors, don't you think more resources would then go into peripheral design/heat management/software development to utilize the improvments, rather than continuing to focus on improving the processor?
The reason the law has held so long and seems to be so consistent is that it sets everyone's expectations, and people plan towards those expectations. Not less, not more.
You've all been Jedi mind tricked.
http://firstmonday.org/issues/issue7_11/tuomi/inde x.html
This is much better article IMO.
It doesn't say Moore's Law isn't a powerful force for technical and social engineering, or that it doesn't drive the PC and high-tech industry, but the simple truth is that Moore's Law doesn't exist as a law at all - and the only place it does exist is in the minds of journalists!
How many escape pods are there? "NONE,SIR!" You counted them? "TWICE, SIR!"
This has been achieved by getting a more and more detailed understanding of the processes and eliminating the fundamental sources of the problems. The costs of doing this are immense, but it works. It's striking to look at micrographs of chips today - everything looks so good. No ragged edges anywhere. Think for a moment about what that means. In some of those pictures, you can see atoms, and they're in the right places. Atoms.
It's not like the bad old days of the "purple plague", ceramics with traces of radioactive minerals, or the HP fab with the 4% yield.
There comes a time when only so much can be fit on that piece of silicon.
e lemen t/C
.525 the size of a Si atom.b /elemen t/C
Couldn't resist replying since i'm on your foes list for god knows what.
Moores law doesn't just apply to silicone, it applies to integrated circuit design in general. If moores law was based purely on the manufacturing techniques at the time, of course it wouldn't hold true in 20 to 30 years.
What we've seen though is a change in manufacturing processes, and with the research going on in quantum computing, light computing, and biological computing, I would speculate that we will see "moores trend" (calling it that because it's not really a law) will continue on course as these advancements are made.
Take for example our use of Si (silicon) for every IC designed today. We use it because of it's strength, and low electrical conductivity. It's a rather large atom though, and we will run out of space eventually using it.
http://www.cs.ubc.ca/cgi-bin/nph-pertab/tab/
But wait! Right above Si on the periodic table of elements is C (carbon) which is roughly
http://www.cs.ubc.ca/cgi-bin/nph-pertab/ta
Now why don't we use carbon based IC's? Well for one thing, carbon does not play well with itself unless you use a lot of heat and pressure (Carbon melting point is 3727c vs Si's 1410c). Secondly we haven't really mass produce the carbon buckyballs/nanotube enough where we could actually use them in place of Si.
Now where was I going with this... Oh yeah..
Basically Moores law will continue to hold true as long as material manufacting keeps up. Silicon has nada to do with it.
experience curve. To quote an article from the Harvard Business Review
"Building stategy on the experience curve" (HBR March-April 1985, pg 143):
The author then presents a graph showing that Dynamic RAM costs fell at 30%
per every doubling of cumulated output from 1976 to 1984. Besides semiconductors,
the author gives examples from the chemical industry and also the Model T Ford.
No electrons were harmed creating this post, though some may have been subjected to electrical and/or magnetic fields.
"The misrepresentation of Moore's Law in the media has always been a real pet peeve..."
./ articles have mentioned the increasing specialization of society as a source of many problems and changes. Is the inability of the media to accurately report on subjects yet another product of this specialization? Both the reduced ability of the reporter to understand technical information and the reduced ability of Americans to spot (and hold media responsible for) incorrect/bad/misleading reporting have, perhaps, allowed the problem to bloom enormously.
Is it possible that Americans have come to mistrust the media, because the media reporters no longer understand 99% of the subjects on which they report?
Earlier
Perhaps Moore's Law is inversely applicable to media reliability. Every two years the reliability and accuracy of the media is halved. Every two years the ability of the average American to spot the innacuracies is reduced by half.
The most interesting thing about Moore's Law, to me, is that it implies that technology advances at an exponential rate. If that's true, then there are some obvious, serious consequences ahead.
IBM announced recently that they will be producing a computer with roughly the same computing power as the human brain, possibly by 2005. That's a pretty significant milestone, if you think about it. Following through with Moore's Law, we should have a computer that is 1000 times more powerful than the human brain as soon as 2020... and a computer that is a million times more powerful than the human brain by 2035.
A million times more powerful than the human brain! What will we do with a computer that powerful? Or, maybe a better question is: What will a computer that powerful do with us?
Kurzweil has a lot of great articles on this sort of thing.
There were a link on pages linked off of that story. Found it at http://www.intel.com/pressroom/kits/quickrefyr.htm and also links to History of the Microprocessor.
These show the history, compare die sizes, etc. and add some context to this article. Had to dig a bit, but I love this kinda stuff. Bet Im not the only one.
Tequila: It's not just for breakfast anymore!
Mean that buying any PC immediately causes Best Buy to sell it for $100 less than you paid.
The first computer I owned was a Compaq Portable (about as portable as a 20" color TV with a handle on it). I paid $3,000 for it - with a 10mb hard drive and a 1200bps Hayes modem.
Today, that same $3,000 - not even adjusting for inflation, and buying only retail equipment, would get me a 2.5+ghz machine with a 20" LCD flat panel. And if I noodge the sales droid enough, toss in a disposable HP printer (as if they make another kind...)
The only concern I have about personal computers these days is... how the heck do I keep track of the exploding volume of information on them?
How do I keep 2,500 family pictures? And find the one of a friend's birthday party?
How do I organize 10 years of letters and emails? And not lose track of the ones from my dad?