FW: Re: FW: Re: [ATM] re:Arbitrary ray origins

["mlbrown@everstrive.com" <mlbrown@everstrive.com>]

[ I'm not getting all emails] 
Yes, you have to turn N around so that N dot I > 0.  That is, if N dot I <
0 then use N = -N

In addition, if the argument to the square root is negative, you get total
internal reflection.

=Matt

Original Message:
-----------------
From: Richard F.L.R. Snashall rflrs@rcn.com
Date: Thu, 14 Oct 2004 23:12:25 +0100
To: atm@atmlist.net
Subject: Re: FW: Re: [ATM] re:Arbitrary ray origins


mlbrown@everstrive.com wrote:

> T = gI + [g(-N dot I) - sqrt( 1 + g²((N dot I)²-1))] N

The domain interrogation I mentioned is:

        N dot I < 0

If true, the ray hits the surface in the wrong direction
to be refracted.  This actually happens in some fisheye
designs.  However, it also means that the user has to
have some way of defining which way is "out".

-- 

Rick S.

http://users.rcn.com/rflrs

_______________________________________________
ATM mailing list http://www.atmlist.net/


--------------------------------------------------------------------
mail2web - Check your email from the web at
http://mail2web.com/ .


_______________________________________________
ATM mailing list http://www.atmlist.net/