Physics, movies and simulating reality

Posted by Rezwan on Jan 15, 2011 at 08:50 AM
Tools: Print | | (0) Comments

Physics not completely realistic? Shocking! But consider a simple classroom example: measuring the depth of a well by dropping in a pebble and timing how long it takes to hit the water. For the calculation we would typically write a simple equation that says the Earth is flat and has no atmosphere.  (in other words, the gravity field is constant and there is no air resistance.) Furthermore, we’d assume that the speed of the sound produced by the pebble hitting water is infinitely high and ignore all sorts of effects like the gravitational attraction force of Jupiter.

Why would we digress so far from reality? If all possible influences were somehow miraculously accounted for, we’d be unable to measure the difference. Hence, the physics model used for measuring depth, while not absolutely realistic, is operationally realistic. It accounts for the major effect and yields predictions well within our measurement accuracy. 

There’s a big difference between operationally correct movie physics and the ridiculously spectacular marketing-driven special effects designed to boost box-office take.

Operationally Realistic Models in Fusion

Meanwhile back in the realm of fusion physics, Gary Weisel’s article on “Properties and Phenomena” suggests that much of plasma physics research is dominated by patron-driven fusion research,  focused on complex machines - the simulation of which is mind boggling and requires massive computational power.  This is at the expense of simpler studies that might shed more light on the forces involved.

Rynn characterized research with Q machines by saying, “if you want to study water waves, drop a pebble into a quiet puddle - not into the surf at the seashore.” Or as Francis Chen put it:  “usually fusion reactors have such complicated geometries that it would be almost impossible to make careful analysis of what is going on.  This is the rationale for having small, simple experiments.”

Note on Computational Power: the computational needs of fusion are greater than those for fission.  As one physicist noted,

How many CPU’s are there in the world, 1 per person? It would take 3 hours of the whole world’s computational capability to model 1 second of a proposed reactor’s operational scenarios. Thousands of scenarios have to be studied.

Don’t panic.  It’s not all about the computer.  You don’t have to model everything for advances to be made:

computers have limited creativity and insight and ... human ingenuity can, at present, do certain things much better than computers. If we waited for computers to understand internal combustion engines, my horse buggy whip business would still be thriving.