Planetside Software Forums

General => Terragen Animation => Topic started by: Kadri on March 21, 2022, 08:26:47 AM

Title: Rock study v3 animation test
Post by: Kadri on March 21, 2022, 08:26:47 AM

Just an animation test for this work https://planetside.co.uk/forums/index.php/topic,29942.0.html

Title: Re: Rock study v3 animation test
Post by: Dune on March 21, 2022, 08:42:00 AM
Quite incredible rock. Love that movement.
Title: Re: Rock study v3 animation test
Post by: WAS on March 21, 2022, 02:03:08 PM
This is pretty darn cool, Kadri! I haven't seen anything like it before in Terragen.
Title: Re: Rock study v3 animation test
Post by: Hannes on March 21, 2022, 03:59:06 PM
Stunning!
Title: Re: Rock study v3 animation test
Post by: WAS on March 21, 2022, 04:36:03 PM
PS when I say never seen, I mean this type of displacement overhang and time based erosion. Matt scene in The Time Machine will always be forever dear to me as far as just time erosion goes. Even if today it seems so simple. That sh** opened my eyes to a world I wanted to be apart of.

Did you use image maps or something? I am actually very puzzled by how you did such unique weathering effect.
Title: Re: Rock study v3 animation test
Post by: Kadri on March 21, 2022, 04:46:37 PM
Quote from: WAS on March 21, 2022, 04:36:03 PMPS when I say never seen, I mean this type of displacement overhang and time based erosion. Matt scene in The Time Machine will always be forever dear to me as far as just time erosion goes. Even if today it seems so simple. That sh** opened my eyes to a world I wanted to be apart of.

Did you use image maps or something? I am actually very puzzled by how you did such unique weathering effect.
When i do a scene like this, that movie comes always to my mind too Jordan :)
I always wish that there was much longer and more animations like that.

I used no image maps. It is all with nodes and actually easy. Wait i will show the node network.
Title: Re: Rock study v3 animation test
Post by: WAS on March 21, 2022, 04:52:34 PM
Yeah I definitely agree. 

And wow, I actually suspected some image Magicka. Surprised there. 
Title: Re: Rock study v3 animation test
Post by: Kadri on March 21, 2022, 05:10:42 PM
:)

This is the setup. As i said in the image thread the scene is based on my old sinkhole scene.

Very much basic as you see. Of course i used a good looking part.
It doesn't look like that in other parts ;)

What i saw from old tests for this kind of movement using less animated parameters as possible is the best.
For me at least. Here it looked like animating the "Fractal warp shader" would be kinda enough.
Warp amount goes from "0.5" to "1" and Roughness from "1" to "0.2".

Title: Re: Rock study v3 animation test
Post by: WAS on March 21, 2022, 05:19:22 PM
That's pretty ingenious, and really simplistic in node setup. Going to have to play around with something like it.
Title: Re: Rock study v3 animation test
Post by: Kadri on March 21, 2022, 06:30:01 PM

By the way i used "Raytrace everything" in the render settings
with some different settings in the "Render Subdiv settings node" like "Ray detail multiplier" at "1".
It takes nearly 3 times longer to render and does have a different kinda triangled polygonal look.
But you get much less bad displacement then using high different settings like "Displacement tolerance" "3" in the planet node for example.
And this takes more then 8 times to render. At least for this scene.
For rocky scenes i sometimes use this method.
You can see the difference below

Title: Re: Rock study v3 animation test
Post by: Kadri on March 21, 2022, 07:33:32 PM

This gave me the opportunity to test the "Ray detail multiplier" setting to see what will happen when you use something like "3".
As you can guess it works and takes much much longer to render. The crop render you see  below was at 51 minutes when i stopped it.
The setting with "1" above took 9 minutes for the full render.
Title: Re: Rock study v3 animation test
Post by: Dune on March 22, 2022, 04:36:24 AM
Did you do this on the planet? Might be interesting on a cube too.
Title: Re: Rock study v3 animation test
Post by: Kadri on March 22, 2022, 08:53:11 AM
Quote from: Dune on March 22, 2022, 04:36:24 AMDid you do this on the planet? Might be interesting on a cube too.
It is on the planet.

I made many different tests with cubes. The problem is when i take this scene directly on a cube it looks more like chaos.
And when i lower the displacements etc. It doesn't look at all like above.

There are so few nodes but there is a delicate balance between parameters that it takes time to get it right in at least some places. And then adapting... I don't know but using displacements on a flat surface and 6 sided objects looks quite different...
Maybe Matt would say "as it should be" :)

But a kinda similar adapted step by step new approach might be usable maybe. I might give it a try later Ulco.
Title: Re: Rock study v3 animation test
Post by: Hannes on March 22, 2022, 09:02:10 AM
Although I have no idea, how all this together produces such nice shapes, I'm very surprised, that there is only one blue node. Really cool stuff, Kadri!
Title: Re: Rock study v3 animation test
Post by: Dune on March 22, 2022, 09:25:32 AM
Kadri is a magician!
Title: Re: Rock study v3 animation test
Post by: Kadri on March 22, 2022, 10:44:31 AM

Don't do this guys :)

I will share this scene or a similar one (with another different render i made before) later.
But i haven't exhausted some possibilities and things i have in my mind so it might take a while guys.
Title: Re: Rock study v3 animation test
Post by: Hannes on March 22, 2022, 10:50:10 AM
Take your time. 
(Can't wait!!!)  ;D ;D ;D
Title: Re: Rock study v3 animation test
Post by: Kadri on March 22, 2022, 10:56:43 AM

 ;D
Title: Re: Rock study v3 animation test
Post by: Kadri on March 22, 2022, 04:11:31 PM

I tried to play around with a cube. Until i rotate the cube there is no displacement with strata (powerfractals too).
And even then it looks wonky as with every parameter of rotate some parts gets displaced and looks wrong to me.
Is this a bug? Somehow it reminded me about the population rotation bug (is it still around?)...

Can anybody confirm this?
Title: Re: Rock study v3 animation test
Post by: WAS on March 22, 2022, 04:19:05 PM
I know that the cube has the edge radius, which needs to be adjusted relative to both scale and displacement. But I notice the XYZ coordinates of this area are just wonky in general, and I'm not as pleased with the results. I'd probably be easier to use sphere. But with strata, it always wants to follow the normal, lateral wise, so not really helpful for both cubes and spheres. Tough I am sure you can use like the terracing function and base it off of finals and Y to get the desires effects, well, similar at least.
Title: Re: Rock study v3 animation test
Post by: Kadri on March 22, 2022, 05:12:26 PM

Yes the edges of a cube are problematic after a certain displacement amount.
Or better said it is tolerable to a certain degree.
With masking, restricting it can be manageable depending on the scene of course.

In this scene you can see those edges when you know what you are looking for:
https://planetside.co.uk/forums/index.php/topic,26357.msg262446.html#msg262446
But it is kinda ok there. To me at least. Better options would be great of course.