Topic awaiting preservation: height fields... (Page 1 of 1)

emps talked so much about those height fields, so i got to try it out.
the big porblem is, that they are getting jaggied when you scale it up.
image

I haven't messed with HFs yet but have you tried making one at a larger resolution to get rid of the jaggies?

*agrees-with-schitzoboy*

_________________________
Who is General Failure, and why is he reading my disk?

yea yea... of course.
that would be the same if i would keep that one and zoom in with the camera so there would be no need to scale the HF up.

that one was a 5-minute-work.

It'll be something like that - it took me a while to get things right and its the little tweaks that make the difference and the size of your initial image (I'd try something large and detailled and scale it down as it looks smoother).

Emps


You're my wife now Dave

Mmmm... height feilds can cause difficulties now and then. A couple of hints (for POV-Ray):

use a large resolution, especially if it's going to be a landscape, and *especially* if it has diagonal lines in it, as the image you presented does.

Try the "smooth" keyword in the height_field. Sometimes this can make it look unrealistic, though.

I don't know of any program that takes advantage of it, but you can use the red and green channel separately to specify more than 256 heights. That is, if you use a black and white image, each pixel can be from a height of zero to 255. But if you use the red and green channel separately, then the green channel specifies a portion of the red channel. That is, As red goes from zero to 255, the height of the height field goes from zero to one. That value is added to the green channel of the image in such a way that as green goes from zero to 255, it's the equivalent of red going from zero to one, or one to two. So it's like, the green channel specifies a fraction of a single value in the red channel.