Ever heard of Moore’s Law? Well basically it is a theory by Intel founder Gordon Moore that states that roughly every two years computing power will be doubled from what was available in the previous two years. For essentially the entire history of computers this law has proven true but now it seems that things have changed.
My theory is that Moore’s Law is no longer relevant hinges on two issues:
The first issue is the economic crisis we’re currently enjoying. With people being unable to spend disposable income on computers, there is less money for chip makers such as Intel. Less money means less development and less development means a slowing down of Moore’s Law.
Secondly is the issue of Netbook’s. People have realised that if a “puny” 1.6 Ghz Atom chip (in terms of power it would have been considered cutting edge about four years ago) can handle Windows XP and all the other software you could require then why bother paying thousands for a more powerful processor. Intel’s most powerful processor, the Core 2 Duo and Core 2 Quad are generally overkill for the average office worker. Sure, Windows Vista is a major power hog but the initial reports on Windows 7 show that it is more than capable with lower power hardware. My theory is that this slow down in power will actually require programmers to have to optimise their code rather than relying on a faster computer handling the information.
The numbers are simple, my two year old Macbook has a 2.0 Ghz Core 2 Duo and the brand new Macbook’s have a 2.0 Ghz Core 2 Duo. Now they’re not technically the same chip and the newer one has extra memory but the thing is, I cannot open Word up any faster than instantly.
Money though is an irrelevant issue as I have a theory: do we even need such faster processors? Do we even need to be able to open Windows point six seconds faster by buying a faster processor? Ultimately the processor is not the biggest bottleneck in the system and until every other part speeds up there is no point in processors improving exponentially for a while now.
Enjoy this post? Subscribe to my RSS feed.
In his original paper, Moore predicted that the number of transistors per
integrated circuit would double every 18 months. He corrected this to 24
months in 1975.
The correct way to think of this law then, is rather in terms of minituarization far more than in terms of performance.
The 1.6GHz Intel Atom chip is probably a fraction of the size and uses far less power than the first generation 1.6GHz chips would have four years ago.
For what it’s worth (probably not much), Moore’s Law is still alive and kicking strong.
When you’ve got a hundred browser tabs open, a movie playing on a separate display and you’re busy writing a DVD while downloading a stack of mp3s, Word does tend to open a helluva lot slower than instantly. This is where faster processors can be a significant advantage, especially if they’re multi-core.