Expert: Daniel Mazur - 9/10/2009

Question
Respected Sir,
I am a post graduate in the physics (Basically I am from INDIA)and i would like to know the information that how the PHYSICS is useful in the computer programing or is their are any computer applications are related with our physics.
I know about the quantum computers but a normal engineering students can't understood the concept.
I will be very thankful to you if you could have help me in this issue with any small applications.

Thanking You Sir,

Thanks and Regards
Aniketh Kulkarni
Asst. Professor,
BVRIT engg College.

"Don't go with the life where it takes "U" ,Take the life where "U" want"

Answer
Dear Akineth,

I understand there are two questions in what you ask: 1) what use is physics to programmers and 2) what use are programmers to physics.

1) Programmers or computer scientists do not need physics to do their work. But knowledge of physics helps them better to understand the purpose of their products - if they work on a development of software for science. It is one thing to develop a fast FLOP operation algorithm, another thing to implement it - the latter usually needs some close contact with electronic engineers to design a dedicated device. It is still a different matter, when you think of your algorithm by itself or when you think about it in context of what it might be used for.

A childish example might be thinking about the standardized numeric formats. Each of them is useful to us, physicists, in a different situation. That's because some problems' solutions need a large number of significant digits, others need a large range in exponent, yet other problems need both. Another example would be the decision making on which algorithm (from many) to use in implementing numerical integrations, because more precise ones take more time. Knowing some physics will help programmers evaluate, which realization they should use, when on mathematical level there is a negligible difference or none.

2) The Physics basic research nowadays relies on three pillars: theory, experiment and simulations. It is in the simulations (computer modeling of physical systems), where people's programming skills are tested most. A lot has been done by the hardware advances in the past three decades, but there is a constant need of fast algorithms and their implementations for specific tasks, so that predictions of modern theories (from solid state to particle physics) can be calculated in a reasonable time. For example the solution to the Schroedinger equation (basic quantum mechanical eq.) can be found analytically for only a few select model systems. A complete unit cell of a crystalline material already has to be calculated by computers and it can take between hours and months (in fact up to infinity, but sensible people don't set up such a calculation), depending on complexity of the material's unit cell and the computational power of the computers used. A desktop PC is good enough for some tasks, while supercomputers are needed for others.

At some point the computational physics meets experimental one, if the algorithm in robust (convergent) enough. Then we can actually automate the theory computation into a data fitting procedure. There is also quite a lot of programming work involved in data acquisition - setting up all the sensors with their calibrated control units and then connecting those to the DAQ computers, securing communication channels, writing scripts for the measurement sequences, implementing feedback (fast or slow), eliminating cross-talk between data channels.... Programmers are usually involved in developing the more universal applications for instrument control, data acquisition, analysis, modeling... It is then up to us, physicists (chemists, biologists,...) to buy and make use of them in our particular fields. Examples: Mathematica, Matlab, Igor, Origin, LabVIEW, Gwyddion, SPIP, ... and many other applications.

From contemporary physics view the quantum computing is in fact just a fancy outlook of computational physics. Its importance is currently quite negligible compared to the vast amount of physics work done by and on computers. From the computer science point of view the quantum computing is big, because it allows many work on something new, which possibly might be very useful in future. It gives people many topics for publications and all the effort is justified (in my eyes) in the fact that it might one day become practical.

Hopefully this helps you a bit,
Daniel