Explaining node indexing in update functions.

[JamesBremner]

Does anyone else have difficulty understanding the node indexing of the update functions?

What do the m+1 and m-1 do? Why doesn't the magnetic field not creep along the axis towrds the right, while the elctric field creep along in the opposite direction?

The answer is, of course, in figure 3.4

I had to stare at this for two days, off and on, before I got it.

I think, this figure could be more helpful. The problem, it seems to me, is that the brackets labelled 'index 0' etc are redundant since we can already see that the bracketed nodes share the indices. Maybe, the brackets should group the nodes that contribute to an update?

Here is how I have documented the update functions in the code I am developing for an acoustic simulation:

/* update velocity 
The velocity is a function of the previous velocity at this location
and the pressures at the nodes on each side of this location

Since a velocity node is assumed to be located one half the location delta 
away from the pressure nod with the same index ( k ) in the +ve direction
then the indices of the pressure locations on each side are k and k+1

Pk            Pk+1
  <<<<< Vk >>>>>
*/

/* update pressure 
The pressure is a function of the previous pressure at this location
and the velocities at the nodes on each side of this location

Since a pressure node is assumed to be located one half the location delta 
away from the velocity node with the same index ( k ) in the -ve direction
then the indices of the pressure locations on each side are k-1 and k

Vk-1            Vk
  <<<<< Pk >>>>>
*/