Blue Node Playground

Started by KlausK, May 15, 2017, 08:44:03 pm

Previous topic - Next topic

bobbystahr

Cool thread and given a case of insanity I may try it, heh heh heh, cuz it'll make me nuts anyway...Interesting work Klaus...
something borrowed,
something Blue.
Ring out the Old.
Bring in the New
Bobby Stahr, Paracosmologist

KlausK

Hi there,
I`ve been playing with the blue nodes again. So I thought I put my experiments in this old thread again to have it all
- especially the useful comments - in one place. Anyway...

Right now I am trying to understand what the "Get" nodes all do.
It is quite often not really clear to me how to use these nodes in the context of a scene. Let alone how to hook them up in a network,
how to use the data they provide and what kind of data they actually provide (like scalar, vector or whathaveyou).
It would be nice to have some rudimentary example networks for (in my case actually most) blue nodes.

First one I tried is the "Get Camera Position".

The Wiki says: "Provides the current position of the main camera in world space, independent of current shading context."
So what to do with this?
Since it is the main cameras position I thought that an animated camera would be the best way to visualize this.
The (at least my) default scene has a Simple Shape Shader connected to the Mask input of  the Fractal Terrain node.
The picture of the network shows how I fed the Get Camera Position to the Simple Shape Shader.
The Multiply by Constant Scalar helps to magnify the cameras position values to values big enough to be noticable (x 1000 in this case).

The animation file:
At the start of the movie the Camera is far away from the Simple Shape Shaders position.
What happens when you get closer and decrease the distance between Camera and Simple Shape Shader is what is shown in the animation.

Nice and smooth shape shifting.

CHeers, Klaus
/ ASUS WS Mainboard / Dual XEON E5-2640v3 / 64GB RAM / NVIDIA GeForce GTX 1070 TI / Win7 Ultimate

Dune

Interesting. The SSS is just (soft) white, so it seems and keeps that area as it is, with a gradient to the changing terrain. The altitude (green) of the camera position is what seems to change the terrain.... intriguing. So if you'd multiply the (changing) cam-pos with something like a pf, it can be very wild I suppose.
Thanks for sharing this, Klaus.

DocCharly65

In fact a very interesting and surreal effect. Looks cool in the video!

KlausK

Thank you, guys.
I am trying to come up with something more interesting and useful...looks like it takes some time  ???
The tgd of for the video is here now also.

CHeers, Klaus
/ ASUS WS Mainboard / Dual XEON E5-2640v3 / 64GB RAM / NVIDIA GeForce GTX 1070 TI / Win7 Ultimate

the first seer


KlausK

Thank you, "the first seer".

CHeers, Klaus
/ ASUS WS Mainboard / Dual XEON E5-2640v3 / 64GB RAM / NVIDIA GeForce GTX 1070 TI / Win7 Ultimate

KlausK

Here is something that came out trying to layer and mix voronoi noise patterns.
Now I try to find a reliable way to soften the displaced edges without almost always changing the rest too much.
I uploaded a few more pictures because the camera distance makes a great difference to the overall look of the structures.

The first two pics differ a little bit in displacement values. Flipping between them makes this mor obvious.

CHeers, Klaus

/ ASUS WS Mainboard / Dual XEON E5-2640v3 / 64GB RAM / NVIDIA GeForce GTX 1070 TI / Win7 Ultimate

KlausK

...some more views.
CHeers, Klaus
/ ASUS WS Mainboard / Dual XEON E5-2640v3 / 64GB RAM / NVIDIA GeForce GTX 1070 TI / Win7 Ultimate

KlausK

...and the Network.
CHeers, Klaus
/ ASUS WS Mainboard / Dual XEON E5-2640v3 / 64GB RAM / NVIDIA GeForce GTX 1070 TI / Win7 Ultimate

Dune

Nice experiments, Klaus. I wonder why you have attached the seed input to the get node (seed needs just one number, afaik), and I think the log doesn't do anything (no input to 'log')... any particular reason, or is it just 'random' linking, seeing what it does?

KlausK

Hi Dune,

it is mostly a mixture of checking my knowledge / understanding / assumption of a function node and trying to verify that.
"seeing what it does" is always a part of this "process".
Also part of that is not really having grasped how the data flows through the network.
I often have trouble rebuilding mathematical formula because i have the feeling that there are sometimes inputs or outputs missing.
But that could - of course - also being me just doing it wrong.
It would be very helpful to be able to visualize the data / numbers / values that a node network produces.
But I cannot for the life of me find out how the "Debug" nodes work...sigh.
Anyway:

Log Scalar:
my thinking was log versus linear. I was hoping to scale the noise less regularly.
But you are right: the missing input on the Log Scalar is a mistake.
The Constant Scalar has to be connected to the Log Scalar and to the Main Input of the Add node.
I must have disconnected it somehow.

The Get Position from Texture is first connected to the X to Scalar node - this goes into the Voronoi.
I wanted the narrower / squeezed and elongated voronoi cells. (hope, that`s understandable...)

Does that answer it?

The attached pics of the changing connections illustrate the differences are quite clearly.
Perhaps someone can use this as a starting point.

CHeers, Klaus

/ ASUS WS Mainboard / Dual XEON E5-2640v3 / 64GB RAM / NVIDIA GeForce GTX 1070 TI / Win7 Ultimate

Dune

Thanks Klaus, it does.

The voronoi cell is very hard edged, maybe you can use the smooth voronoi (color) instead and get interesting results....
Ther are some nice squares in the preview, so maybe those can be rotated and used on a lateral area.

KlausK

Thanks for the suggestion. I`ll have a go at it sometime.
CHeers, Klaus
/ ASUS WS Mainboard / Dual XEON E5-2640v3 / 64GB RAM / NVIDIA GeForce GTX 1070 TI / Win7 Ultimate