Osmos follows the laws of physics - almost

A good place for general questions and comments...
shoyer
Posts: 1
Joined: Thu Dec 23, 2010 11:51 pm

Osmos follows the laws of physics - almost

Postby shoyer » Fri Dec 24, 2010 1:51 am

Osmos is a very cleverly designed game, with a wonderful variety of physics inspired mechanics. As a physicist, I was impressed and entranced after downloading it to my iPhone earlier today.

However, as the developers probably realize, Osmos does not quite follow the laws of physics.

The trouble is that it violates the principle of conservation of energy. Here's an experiment to prove it (I recommend trying it on one of the "Impasse" levels):
1. Grow your mote until it absorbs everything else.
2. Note the speed of your mote (it's probably moving pretty slowly).
3. Now, accelerate your mote to a higher speed by repeatedly ejecting mass in the direction opposing its motion.
4. After waiting some time, your mote will reabsorb all the emitted mass. However, now it is moving at a faster speed -- your mote has somehow gained kinetic energy!

Where did this energy come from? Mote appear to have infinite stores of internal energy, some of which released each time they shoots off mass. They're perpetual motion machines!

This may appear to be a bit of nitpicking, since the violation is practically necessary for the game to be played. But there is a fairly elegant solution. Each mote would have an additional quantity of internal energy -- an amount of "gas" or "food" illustrated by internal shape, color or motion. This internal energy could be transfered between motes as is done with mass. The important aspect is that to assure conservation of energy, it is required for propulsion: every time mass is ejected, it costs internal energy equal to (or greater than) the amount of added kinetic energy.

It's a tricky question whether adding a finite supply of energy would make the game more fun or not. On this matter, I'm not sure. But it would certainly fix the only physics problem I could find without looking at the source code!

Anyways, well done! I had a blast -- especially when I was trying to come up with experiments to test your physics :).

Ayjay
Posts: 1
Joined: Mon Apr 04, 2011 10:26 am

Re: Osmos follows the laws of physics - almost

Postby Ayjay » Mon Apr 04, 2011 10:46 am

The gain in kinetic energy comes from the perfectly elastic collisions with the side of the map. As soon as ejected mass hits the wall, the direction of motion completely reverses. If this rejoins with your mote, you've just gained kinetic energy for "free". If the motion of the map itself were modelled and considered, total energy would be conserved. Given that, in reality, most boundaries are fixed to the earth (an effectively immobile mass), this is a decent simulation.

With regards to the fuel for the action where the mote shoots off a bit of mass, that's just not modelled in game. If the mote is using the mass it collects to power the expulsion (rather than some internal store), then the amount of mass required would be really tiny. Anyway, it's not really relevant to the conservation of kinetic energy - when you stand still and throw a ball, yes, a teensy amount of mass is converted to the kinetic energy of the ball, but it is so infinitesimal it's pretty forgivable to ignore in most contexts! :D


Return to “Hemisphere Meta”

Who is online

Users browsing this forum: No registered users and 8 guests