Some Cloud Doodles

[FrankB]

From time to time I delve into wild experiments with clouds. I thought some of this might be interesting and maybe in some cases inspiring to others.
So I figured I could keep this thread to post some new, crazy and unfinished experiements here.

First thing (I'm currently working on) is ripples in clouds. The image below is one of my attempts to achieve this. A lot of functions are involved, and "counter-functions" to take some of the regularity away and add randomness instead.

I've always been intrigued by rippled and feathered cirrus... never actually achieved this so far.

The hunt continues ;-)

Frank

[FrankB]

Here is another interesting one. During my attempts to create interesting planetary scale cloud patterns (which failed and nvseal's methods are waaayyyy superior), I had stumbled across this "lucky seed" (see picture).

You can give it a shot, too, by choosing a very large feature scale and a 10^3 smaller largest scale. This sometimes can produce these "whirls".

[Henry Blewer]

Intriguing images. Are you using T 2.2? The last time I tried to emulate some of your wonderful clouds, you were using the alpha 2.2. lol

[Dune]

Very nice structures, Frank.

very large feature scale and a 10^3 smaller largest scale

Would large be 1e+006 or something, and 10^3 smaller the same as 1/1000 that size? I'm just working with these large scales in an attempt to project some interplanetary structures on the background node...

[FrankB]

Very nice structures, Frank.

very large feature scale and a 10^3 smaller largest scale

Would large be 1e+006 or something, and 10^3 smaller the same as 1/1000 that size? I'm just working with these large scales in an attempt to project some interplanetary structures on the background node...

Ah, I've done the same recently. Look at the attached image. I've doodled with the background node and produced some weird fires in the sky from it. I'm sure if somebody would take this further, .... it has some potential.

And yes, your summary on the scales is exactly what I meant.

Regards,
Frank

[Dune]

Interesting and thanks. I was hoping to produce some interstellar clouds low on memory use, but they still looked too flat to be really useful. But my structures kind of looked the same. No that we're here; you might know this. The radius of the background is -2e+008 (I believe). When I change this to -4 or -1, or -2e+006, the whole background turns white. I think the background is just a huge inside out planet/ball, but why would the color change if you resize?

[Zairyn Arsyn]

the first looks really good, realistic imo, NWDA store/cloud pack material. 

the 3rd is slightly frightening. 

[FrankB]

Interesting and thanks. I was hoping to produce some interstellar clouds low on memory use, but they still looked too flat to be really useful. But my structures kind of looked the same. No that we're here; you might know this. The radius of the background is -2e+008 (I believe). When I change this to -4 or -1, or -2e+006, the whole background turns white. I think the background is just a huge inside out planet/ball, but why would the color change if you resize?

I don't have this issue. i can chnge these values just fine. But I haven't realized before that the background sphere has a negative radius... strange.

[Volker Harun]

But I haven't realized before that the background sphere has a negative radius... strange.

A TG2 sphere has its surface on the outside, so the background sphere is turned inside out for its purpose.

I really like the first and third set of clouds. I might have an idea for feathered cirrus, give me some time for it ,-)

Volker

[Dune]

So there might be another trick for interstellar gases: use an inside out planet with the negative radius of the background, invert the atmosphere, so it's 'inside' your outside and add huge cloud formations 

Perhaps you can feather clouds by using a transform shader to move the main cloud a little, add another (feathered) PF and attach to final density. Perhaps you might need to use a merge shader somewhere to add both together for the final density, or you'll loose your main cloud. If you get my drift.