Cloud Altitude Offset Limitation

[RRMessiah]

I've been experimenting with the new cloud altitude offset function. In short, if you want your entire cloud layer visible you're constrained to a multiplier of about 25% of the cloud depth. This is quite limiting if you want real differences in altitudes in your clouds.

Anything above 25% and your cloud map will begin clipping out and you will lose information. Is this on purpose or some sort of bug?

[RArcher]

Have you taken a read through the tutorial Martin created?  You can find it here:  http://www.planetside.co.uk/content/view/61/99/

My guess as to what you are experiencing is that you are running into the top constraint of your cloud altitude.  Say your altitude is set to 2000 and your depth is set to 500, this would give you a max altitude of 2250.  If you are pushing the clouds up beyond that point they are going to be cut off.  Increase the depth and you should be able to make the clouds as big as you want.

[RRMessiah]

Yes you are correct and yes I've read it. Depth has to change, but all you're doing is making the clouds thicker. The steepness of altitude is constant/limited no matter what. It's evident in his video:

http://www.youtube.com/watch?feature=player_profilepage&v=d-X8jImO6Wo

The depth of the cloud changes, but you can't increase the curve profile without clipping the clouds due to the interaction between altitude and depth.

[Tangled-Universe]

The steepness of altitude is constant/limited no matter what.

With my best intentions I can't figure out what you mean here

You mean that the altitude offset function cannot contain too much contrast because the altitude offset would differ too much in a small area?

The depth of the cloud changes, but you can't increase the curve profile without clipping the clouds due to the interaction between altitude and depth.

Reading this along with your idea that you MUST use 25% depth modulation (why do you think so???) I'm pretty sure you're driving your function through the boundaries of your cloud layer.

You need to create a thicker cloudlayer and use less low depth modulation. This still gives you a thick cloudlayer with enough curve profile.
The thicker the cloud and the smaller the depth modulation, the greater curve profile you'll have.

[RRMessiah]

With my best intentions I can't figure out what you mean here Smiley

Sorry I wasn't clear. It's a tough thing to explain.

Reading this along with your idea that you MUST use 25% depth modulation (why do you think so???) I'm pretty sure you're driving your function through the boundaries of your cloud layer.

Yes! That's the problem!  If you're affecting the altitude offset, why should the boundary be considered at that point?

If I wanted one particular cloud within a cloud distribution shader to be offset by 500 meters, I have to make those clouds super deep (or create a new cloud altogether) to do it. To me, that's the wrong way to do it. If I like how deep my clouds are and I want to change the altitude range, I shouldn't have to change depth at all.

[Tangled-Universe]

No, it's actually quite logical.
The cloud attitude and depth parameters are necessary to create a "box" where clouds can exist in.

This depth is then being modified so that you can alter the altitude.
As long as you don't offset the clouds too much and keep them within the "box" boundaries it's all fine. Without these boundaries TG2 would need to calculate the cloud functions for the entire depth of the atmosphere or more.

That's why I started with and repeated a lot in my tutorial that altitude and depth are critical to understand.

The reason for this system and approach in altitude/depth modulaire make perfect sense then.