The Computer is Wrong

If you think the complex microchips that drive modern computers are models of deterministic precision, think again. Their behaviour is inherently unpredictable and chaotic, a property one normally associates with the weather.

Intel's widely used Pentium 4 microprocessor has 42 million transistors and the newer Itanium 2 has no fewer than 410 million. Their performance can be highly variable and difficult to predict, says Hugues Berry of the National Research Institute for Information and Automation in Orsay, France.

Berry, Daniel Perez and Olivier Temam say that chaos theory can explain the unpredictable behaviour. The team ran a standard program repeatedly on a simulator which engineers routinely use to design and test microprocessors, and found that the time taken to complete the task varied greatly from one run to the next.

But within the irregularity, the team detected a pattern, the mathematical signature of "deterministic chaos", a property that governs other chaotic systems such as weather. Such systems are extremely sensitive - a small change at one point can lead to wide fluctuations at a later time. For complex microprocessors, this means that the precise course of a computation, including how long it takes, is sensitive to the processor's state when the computation began (www.arxiv.org/nlin.AO/0506030).

--From issue 2507 of New Scientist magazine, 11 July 2005, page 17

My boss at Environment Canada believes that people who are truly good with computer have a certain arrogance to believe that sometimes the machine is wrong. That is, they believe that they have done everything correctly and the machine is at fault. Basically, he is correct. I said that the problem is that every layer in the computer, from the transistors at the bottom to the applications at the top, is thought of as having defined discrete states with instantaneous transitions between them. However, at the bottom level, no phyiscal system has perfect states or instantaneous transitions, so the fuzz of the lower layers silently travels up into the higher layers suddenly causing bizzare, unexpected and possibly non-reproducable behaviours.