Jonathan Richards
Enter our Snapshots of Summer photography competition
For decades it has been the benchmark by which advancements in computing are measured.
Now Moore's Law - the maxim which states that computers double in speed roughly every two years - has come under threat, from none other than the man who coined it.
Gordon Moore, the retired co-founder of Intel, wrote an influential paper in 1965 called 'Cramming more components onto integrated circuits', in which he theorised that the number of transistors on a computer chip would double at a constant rate.
Silicon Valley has kept up with his widely accepted maxim for more than 40 years, to the point where a new generation of chips, which Intel will begin to produce next year, will have transistors so tiny that four million of them could fit on the head of a pin.
In an interview yesterday, however, Mr Moore said by about 2020, his law would come up against a rather intractable stumbling block: the laws of physics.
"Another decade, a decade and a half, I think we'll hit something fairly fundamental," Mr Moore said at Intel's twice-annual technology conference. Then Moore's Law will be no more.
Mr Moore was speaking as Intel gave its first demonstration of a new family of processors, to be introduced in November, which contain circuitry 45 nanometres - billionths of a metre - wide.
The 'Penryn' processors, 15 of which will be introduced this year, with another 20 to follow in the first quarter of 2008, will be so advanced that a single chip will contain as many as 820 million transistors.
Computer experts said today that a failure to live up to Moore's Law would not limit the ultimate speed at which computers could run. Instead, the technology used to manufacture chips would shift.
The current method of Silicon-based manufacturing is known as "bulk CMOS", which is essentially a 'top-down' approach, where the maker starts with a piece of Silicon and 'etches out' the parts that aren't needed.
"The technology which will replace this is a bottom-up approach, where chips will be assembled using individual atoms or molecules, a type of nanotechnology," Jim Tully, chief of research for semi-conductors at Gartner, the analyst, said.
"It's not standardised yet - people are still experimenting - but you might refer to this new breed of chips as 'molecular devices'."
Anthony Finkelstein, head of computer science at University College London, said, however, that a more pressing problem in the meantime was to write programs which took full advantage of existing technologies. "It's all very well having multicore chips in desktop machines, but if the software does not take advantage of them, you gain no benefit."
"We are hitting the software barrier before we hit the physical barrier," he said.
Mr Moore, who is 78, pioneered the design of the integrated circuit, and went on to co-found Intel in 1968, where he served as chief executive between 1975 and 1987.
Asked what area he would research if he were a graduate today, he told the conference: "I'd probably look at something more in the biology mould. The interface between computers and biology is a very interesting area."
Win a luxury weekend to Newcastle and its neighbour Gateshead, find out more here
Risk, resilience and embracing new technology
Industry sectors news at a glance. Interactive heatmap, video and podcast
Discover the collective power of smart thinking. Submit a solution and be in with a chance to win a Flip MinoHD Camcorder
The inside track on current trends in the charity, not for profit and social enterprise sectors
Everything the Business Traveller needs to know to make a better trip
Make the most of the summer and enter our fabulous photographic competition, you could win a £5000 holiday
Corsica is an island of beauty and contrast, an ideal holiday destination
Enjoy further reading from Travel to Fashion, Business to Sport, discover more
Shortcuts to help you find sections and articles
The clever way to lease a new car is with Car leasing made simple™
2009
42,945
2008
71,450
Car Insurance
Not Specified
MI6
UK-based
£60,000
The Environment Agency
Bristol
Up to £90K
Boots
Midlands
OTE £85k
Credit Protection Association
Nationwide Opportunities
Completely London
Luxury Condo's in Manhattan with NYC views
The best new homes in Wimbledon?
Nationwide
Save up to £1,000 per couple with Elite Vacations at the five-star Constance Lemuria Resort
and do the British Isles this Summer.
Save up to 60% with Oxford Hotels and Inns
Try our inspiring luxury holidays to the Indian Subcontinent and South East Asia.
Great offers available
8 fabulous Canadian cities ...you won’t find cheaper
Contact our advertising team for advertising and sponsorship in Times Online, The Times and The Sunday Times, or place your advertisement.
Times Online Services: Dating | Jobs | Property Search | Used Cars | Holidays | Births, Marriages, Deaths | Subscriptions | E-paper
News International associated websites: Globrix Property Search | Property Finder | Milkround
Copyright 2009 Times Newspapers Ltd.
This service is provided on Times Newspapers' standard Terms and Conditions. Please read our Privacy Policy.To inquire about a licence to reproduce material from Times Online, The Times or The Sunday Times, click here.This website is published by a member of the News International Group. News International Limited, 1 Virginia St, London E98 1XY, is the holding company for the News International group and is registered in England No 81701. VAT number GB 243 8054 69.
I take a more balanced view of these. I am completely new school, a Cincinnati based J2EE contractor and have never written in a more primitive language than C, so i am a far cry from that of old. I concur with Name Witheld in that there are a lot of crap algorithms written, and in my experiences in engineering large scale projects there are just hoards of minor inefficiencies that collectively add up to a much slower program than you could have. I think what he means is the speed of the CPU obscures this.
I prefer to look at it like this. We do need the advanced CPUs and layers of encapsulation to make it easier. That coupled with the actual writing of efficient algorithms and proper language semantics would allow us to accomplish more work with the same resources than we do now. I don't see any advantage in taking away the hardware in an effort to push software optimization, this will happen when market pressure mandates it due to a lack of funding for an extra box.
Joe, Independence, Ky
We also use Linux to develop our applications whilst using Java to deliver a front end which will run on MS, Apple or Linux desktops.
We find it allows us to use less resource without the need to take longer than neccessary to produce the final version of the code.
Regardless of how much processing power is available it makes sense to use the resources as economically as possible.
Economics is the driving force for us as we can only continue as long as customers are prepared to pay for what we do.
Using slower processors, less memory, less bandwidth and particularly a rapid development cycle are key when competing for business.
Bill Wilson, DULVERTON, Somerset UK
"I'd probably look at something more in the biology mould. The interface between computers and biology is a very interesting area."
Is that not just Cybernetics? I could be wrong...
John Westlake, Portsmouth, UK
I disagree with that statement. There is a need for faster CPU speeds. As systems become ever more complex, the programming platforms move to ever higher levels of abstraction so that the human mind can cope with the complexity. If all things were equal and we were writing the same applications then it would have been a valid point. But we are writing vastly more complicated applications these days. And we are writing them faster.
Matt Smith, London,
Too right, Name Withheld, perhaps it should be called sloppyware rather than software.
Kidd Garrett, Bristol, UK
When the limit of CPU power is reached we need to look into generating a multi-cpu computer to speed things up. Having multiple CPU's dedicated inside the computer to take the load off the main CPU and speed things up. I have already done this with my computer as I have the main CPU, with a Radeon GFX CPU so that it doesn't have to do the GFX work, and MPG Encoder/Decoder CPU dedicated to video work, and a SB X-Fi Sound Card which has it's own dedicated sound CPU to generate the Dolby Digital 5.1 sound taking the sound work off the CPU.
Darren Forster, Warrington, UK
This debate essentially boils down to economics: developer time vs cost of hardware. Right now, hardware is still cheaper, and this encourages sloppy programming. It might not be a bad thing if the cost of developing faster processors increases and redresses the balance a little between the two. It is a scandal - an absolute scandal - that, even with ever faster processors, software is getting progressively slower, because resource guzzlers such as .NET keep upping the cycles needed to do even the most basic operations. Each generation of coders is told: "no need to learn that old complex stuff; this new platform may be slower, but it's easier to code in and CPUs are getting faster all the time." Check out Visual Studio 2005. Its performance is an outrage. There's even an environmental aspect to all this - all this inefficient coding requiring faster processors means more energy consumption. The massive energy requirements of today's server farms are the result.
David Pritchard, Madrid, Spain
I actually agree with name withheld, my current pc is athlon 4600 x2, 2gb ram, 1 tb hd , 7900gt gfx , and yet at times it seems that it really isnt much faster than my childhood amiga 7mhz cpu, amiga o/s was more stable than xp too
the faster and more complex hardware gets, the less the software catches up, people really are writing rubbish code using visual studio using apis and objects of which no clue is given of the underlying code , resulting in vastly bloated and unstable executables .. bring back simplicity , bring back assembly language and rid users of c++/delphi/VB bloatware
Paul, Bournemouth , UK
Dear Name Witheld,
Sorry but it is a bit more complicated than you suggest.
Twenty years ago I wrote a realtime sonar display program running in 32kwords (64 KBytes) on a PDP11. Very nice code but I spent a lot of time optimising the code to run within the available resources. I had to be very aware of the computational resources I was using, the code was tied closely to the hardware.
These days I can write code that addresses far more complicated issues, that will run on Linux and on Windows. With users able to see the results by connecting over the web from anywhere in the world.
I, personnally, get more kick out of solving a difficult problem, than running on an antique processor. In the same way that I'm not looking to buy a vintage car; I just want reliable one that starts, stops and steers when I want it too.
Let's not over-romanticise antiques; they were the miracle of their time, but then the world moved on. I like mobile phones, digital TV and the internet
Paul, Exeter, Devon
There's no need for the speed of cpu's we have today even. The problem is that programmers have gotten sloppy. Look at how fast the C64 and Apple 2c were. They were lightening quick compared to today's applications. The reason was they only ran a few apps at a time and they were coded very tight. Today we attempt to mimic that by creating protected rooms of memory for the application but they don't work half as well as the computers of old and the programming is filled with junk created by visual programming. That is why Linux for example can do so much more with far less resources.
Name Withheld, somewhere, NE