Author Topic: Rock study  (Read 13051 times)

Offline TheBadger

  • Member
  • *
  • Posts: 7416
  • what's the difference?
Re: Rock study
« Reply #30 on: August 27, 2015, 03:27:41 AM »
Kadri,
I did see that you were playing with the idea! As for me, I gave up on trying, I am just waiting for the rectangular noise.
I thought to use the file for Islands too. For my greek/roman city which I wanted to be for Poseidon the way Athens is for Athena... Still can't come up with a name for a city for poseidon that sounds good the way Athens does for Athena... if you have any ideas.

Thanks for the file!

It has been eaten.

Offline AP

  • Member
  • *
  • Posts: 2688
Re: Rock study
« Reply #31 on: August 27, 2015, 03:37:56 AM »
The latest test looks nice as well. I would imagine that would look very cool once the procedural erosion is available. I am working on some hero mountain tests right now. I will probably make a dozen of them and post the settings also.

Offline Matt

  • Planetside Staff
  • *
  • Posts: 2852
  • I'm the crazy one
Re: Rock study
« Reply #32 on: August 27, 2015, 03:50:46 AM »
Kadri,

The combination of multiple fractal warp shaders is really interesting, and I didn't realise it produced this sort of stepping. But I don't understand what the Twist and Shear is doing. When I try it it doesn't change the character of the rocks, it just shears it like I would expect it to shear any other displacement. I'm not seeing where it fits into your renders.

Matt
Just because milk is white doesn't mean that clouds are made of milk.

Offline Matt

  • Planetside Staff
  • *
  • Posts: 2852
  • I'm the crazy one
Re: Rock study
« Reply #33 on: August 27, 2015, 04:01:16 AM »
Oh, you're compressing it down by making the lean vector point downwards. You know you can get the same effect just by making the stone tallness 0.4 instead of 2?

Matt
Just because milk is white doesn't mean that clouds are made of milk.

Offline AP

  • Member
  • *
  • Posts: 2688
Re: Rock study
« Reply #34 on: August 27, 2015, 04:35:13 AM »
If i am understanding this correctly. The more warp intensity, the more displaced stepping forms along the x and z axis.
« Last Edit: August 27, 2015, 06:08:05 AM by Chris »

Offline Kadri

  • Member
  • *
  • Posts: 7865
Re: Rock study
« Reply #35 on: August 27, 2015, 06:07:26 AM »
Oh, you're compressing it down by making the lean vector point downwards. You know you can get the same effect just by making the stone tallness 0.4 instead of 2?

Matt

I do mostly something and then add another thing to see what it does as i go.
There are so many options,possibilities it just happens without any deep thinking Matt. Just tests.
Some parts-values are leftovers from other tests.
Sometimes there seams to be things found when you play with different approaches.
I just asked myself here what would do a negative twist do as i played with different things.
So there are probably some parts-numbers that doesn't do anything like the high warp settings (not sure).
It looked to me as if there is a limit.So i added other warp nodes to see if they help for example.

I am rendering another test with an Alpine fractal now and had not thought to use a negative displacement before with that node.
It looks like kinda a canyon that way. With 6 Warp shaders this time.It is rendering now.
« Last Edit: August 27, 2015, 06:18:00 AM by Kadri »

Offline AP

  • Member
  • *
  • Posts: 2688
Re: Rock study
« Reply #36 on: August 27, 2015, 06:09:07 AM »
Well, that should be interesting.

Offline Dune

  • Member
  • *
  • Posts: 12827
  • Corkscrew Bird
    • www.ulco-art.nl
Re: Rock study
« Reply #37 on: August 27, 2015, 06:13:13 AM »
You're heading into dangerous terrain, Kadri  ;) You get a lot of small displacements probably intersecting eachother. I wouldn't dare to put a RT reflective shader on that. But... really interesting structures!
In case you still haven't seen enough of my work: www.ulco-art.nl

Offline Kadri

  • Member
  • *
  • Posts: 7865
Re: Rock study
« Reply #38 on: August 27, 2015, 06:16:55 AM »

From far away it looks interesting but not sure how it will look closer Chris.

Ulco yes it is rendering even just with that nodes...uhh...a little slow but i was curious :)

Offline Kadri

  • Member
  • *
  • Posts: 7865
Re: Rock study
« Reply #39 on: August 27, 2015, 06:20:45 AM »

Michael there are so many names you might choose... Helike for example?

Edit: you might have seen this probably  https://en.wikipedia.org/wiki/List_of_ancient_Greek_cities
« Last Edit: August 27, 2015, 07:06:03 AM by Kadri »

Offline Kadri

  • Member
  • *
  • Posts: 7865
Re: Rock study
« Reply #40 on: August 27, 2015, 06:34:59 AM »

The first image does have 3 warp nodes after an Alpine fractal.
The scales in the warp nodes are different and goes like 100,82,63.

The second image does have another 3 warp nodes.
"Scale" values are 47,19,7 in those added nodes.
I changed the erosion values to higher numbers in this second image.


Offline Matt

  • Planetside Staff
  • *
  • Posts: 2852
  • I'm the crazy one
Re: Rock study
« Reply #41 on: August 27, 2015, 06:44:45 AM »
I do mostly something and then add another thing to see what it does as i go.
There are so many options,possibilities it just happens without any deep thinking Matt. Just tests.
Some parts-values are leftovers from other tests.
Sometimes there seams to be things found when you play with different approaches.

That's a good way to experiment with a complex system. I didn't mean to criticize that approach at all.

The Y component of a twist and shear shader can be used to exaggerate or reduce the height of a terrain. It wasn't specifically designed to adjust height, but it's perfectly valid to do. In some cases it will be really useful. In your setup with the stone warped by fractal warp shaders, I think it's useful to know that the twist and shear shader isn't necessary to achieve the effect, so you can simplify the setup and make it easier to understand how it works.

Matt
Just because milk is white doesn't mean that clouds are made of milk.

Offline Kadri

  • Member
  • *
  • Posts: 7865
Re: Rock study
« Reply #42 on: August 27, 2015, 06:54:18 AM »
...
 I didn't mean to criticize that approach at all.
...

No problem Matt.I didn't saw it that way.
It was just me trying to express myself, as clear as i can in another language, probably a little too lengthy as always ;)
« Last Edit: August 27, 2015, 07:26:12 AM by Kadri »

Offline Matt

  • Planetside Staff
  • *
  • Posts: 2852
  • I'm the crazy one
Re: Rock study
« Reply #43 on: August 27, 2015, 07:06:31 AM »
So there are probably some parts-numbers that doesn't do anything like the high warp settings (not sure).
It looked to me as if there is a limit.So i added other warp nodes to see if they help for example.

The shader limits its own warp amount (with a kind of soft clamp) so that the displacement doesn't become discontinuous, so the cliffs should always be less than 90 degree slope. Higher warp amounts just push the cliffs closer to 90 degrees and make the shallow slopes even shallower. Using additional nodes will exaggerate this but won't overcome that limit. But they can change the character of the surface and make it more complex, as your results show.

To create overhanging cliffs you could add some horizontal displacement afterwards, maybe using a Twist and Shear Shader :)  Then you could reduce the warp effect and perhaps get some natural-looking structures. Because the problem with these extreme warps is that they create unnatural stretched cliffs.

Matt
« Last Edit: August 27, 2015, 07:08:03 AM by Matt »
Just because milk is white doesn't mean that clouds are made of milk.

Offline Kadri

  • Member
  • *
  • Posts: 7865
Re: Rock study
« Reply #44 on: August 27, 2015, 07:17:48 AM »
So there are probably some parts-numbers that doesn't do anything like the high warp settings (not sure).
It looked to me as if there is a limit.So i added other warp nodes to see if they help for example.

The shader limits its own warp amount (with a kind of soft clamp) so that the displacement doesn't become discontinuous, so the cliffs should always be less than 90 degree slope. Higher warp amounts just push the cliffs closer to 90 degrees and make the shallow slopes even shallower. Using additional nodes will exaggerate this but won't overcome that limit. But they can change the character of the surface and make it more complex, as your results show.
...


Good to know for sure.Thanks :)


...
To create overhanging cliffs you could add some horizontal displacement afterwards, maybe using a Twist and Shear Shader :)  Then you could reduce the warp effect and perhaps get some natural-looking structures. Because the problem with these extreme warps is that they create unnatural stretched cliffs.
...


Yes that was my thought too. Just with those nodes it is not enough.
I saw it in a small closer test render in the file i shared.
Next stage maybe all together :)
« Last Edit: August 27, 2015, 07:20:39 AM by Kadri »