LOL, Ajax, you are absolutely right. But that's why I said
"There are times when it does, but that's usually only required for very close-up images..."

In the vast majority of cases, refraction doesn't come into it (in image production terms). After all, how often do you create an image that sees both sides of a piece of glass?

I'm not talking strict physics here, only practical image production. Usually, refraction can be ignored for reflective images. If you have an object passing through an object that is both (to some degree) reflective and refractive then both come into the equation and might need to be taken into account. If in close-up. Depending on the material. Taking glass as an example, the difference can't really be seen from more than a few feet away (I think - not sure about that: not seen many real-life examples of pencils embedded in glass..).

As for water, well (no pun intended), yes, the eqation needs to balance. If, that is, you're going for a super-realistic image (and extended render times)..

I apologise. My previous post was based on a pragmatic approach, not a detailed one. I should have made that clear.. But the thread was about mirrors, so that's how I responded...

Cheers,
Diolma

Message edited on: 02/10/2006 15:55