adding displacement cancels other displacements

[Matt]

If the surface is displaced negatively at any point, it will fall into shadow because the ray-traced version of the object - which is used to cast shadows - isn't displaced accordingly.

If you're using a Default Shader, negative displacement can happen if you use a negative displacement multiplier, or if your displacement function has negative values in some places.

Which Wiki page were you reading about offset?

Matt

[Matt]

If I understand your thought process correctly, you are maybe thinking that if you displace outward by 1m, then use a displacement function/map to add *negative* (inward) displacement of -1m *or less* (i.e. it does not displace *below* the original surface), then perhaps it will avoid the artifacts you're seeing. I honestly do not know if that is the case, but it should be easy enough to try. Matt, assuming my understanding is correct, does my explanation of the idea make any more sense to you?

If you add a Power Fractal displacement then you can get negative displacement, which you don't want. So yes, now I think I see why displacement offset came into the question. Adding a positive offset in the Power Fractal should allow you to prevent it from becoming negative in some places.

Badger, can we get a render showing what you're seeing, so I'm not left trying to guess from the OP, which is a completely separate question?

Matt

[Dune]

Ah, this

(If the surface is displaced negatively at any point, it will fall into shadow because the ray-traced version of the object - which is used to cast shadows - isn't displaced accordingly.)

was probably the case in my snowy scene: http://www.planetside.co.uk/forums/index.php/topic,19591.msg192702.html#msg192702

I also tried using an offset in the displacement of worn areas of a pillar, but that makes the eroded areas stand out from the pillar, not eroded inward.

[TheBadger]

Badger, can we get a render showing what you're seeing

Ok, I will. But it won't work, because:

If the surface is displaced negatively at any point, it will fall into shadow because the ray-traced version of the object - which is used to cast shadows - isn't displaced accordingly.

What I thought I was trying to do, was get around this. By raising the surface first, then displacing it negative (back down at least to the original surface) I thought that I could trick it.

I had thought maybe this would work because I was dealing with such small displacement values (0.002). And that the camera any normal distance away from the subject would look real

Does that make better sense of what I was posting about? I will try this one more time now that this thread gave me a better understanding of things. If there are any results to speak of I will post them. The reason I did not post an image before was because I got no results at all (that I could identify) perhaps because of my small displacement values. I will use much bigger number to force a obvious result.