Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - Campi

Pages: 1 ... 4 5 [6]
I feel like I am missing something:

A lot of Substances I have looked at have a very small area of features.
For example a brick wall tile might be  set up in a fixed way of let's say 4x8 bricks on a tile. Maybe a little more.
But never in a way that it is built to cover an entire wall with unique bricks on each wallsegement.

I can understand why this is common - people work on details and it is much easier this way and the value ranges of substances sliders by default are made to work best with a certain amount of 'cells' on a tile so you tend to work in that cell range.

I know it is easy enough to increase the tiling within the project but that's just it: All it does is tile the small set of features you constructed in your original 'tile'.

I feel with all these possible randomization SD allows it should be very easy to globally change the computed amount of cells without affecting size realtionships between cells.

Or in other words:
If I have made a tile with 4x8 bricks, I set up all the detail scale, size relationships and gap sizes for the bricks (random) and randomized seeds for each brick, what is the easiest way to - after the fact - change the tile to a 40x80 brick tile without just tiling the original one.
Do I need to go into each FX Map and readjust Offsets etc ?
Or is there somewhere a slider I can use to just increase the 'cell count' ?

What I could imagine happening is a 'Quadrant' Node that allows you to set the amount of 'Quadrants' that should be generated.
Bu changing the Quadrant amount you could generate more cells.
I know there probably isn't a quick one slider way to do this at the moment but it almost seems like there should be a way to mark  nodes and FX Maps as 'global'.
So 'global' ones get recalculated for each cell even when tiled, while non-global ones are allowed to 'repeat'. This way, while pattern offset for example would still be repeating, at least the randomization of each brick could be unique.


Oh I think we have the same way.
The if Statements are directly connected. Just a little awkward layout of the Lesser/Equal Nodes on top of the connection between the ifs.

Edit: Never mind. You are right. Yours would be different. I'll try that out as well just to get the idea.

The EDIT of my last post actually triggered the solution.
So this is working now. Thanks for the idea !

Hm. Still not sure I am doing it right.
In my book Input 4 (green) should only have a roughly 5 percent chance of being picked.
Yet in a range from 0 - 1500 randomization I seem to be hitting green a lot.

I read the graph as:

- Generate random value of 0 - 1
- If Value less than 0.3 Output 1, else output random value
- If Value less than 0.6 Output 2, else output random value
- If Value less than 0.95 Output 3, else output random value
-If Value less than 0.99 Output 4, else output random value

I would almost need to figure out the equivalent of a 'break' statement, that if one if condition becomes true, that it stops and outputs the value.
Because at the moment it almost seems that even though conditions are met earlier in the chain the value (1,2,3) that is output doesn't get through the chain ?
This visual coding is confusing me, I want an expression node :)

EDIT: Tried as well feeding the the Output of the second to last if statement into the ELSE of the last if statement since I assumed that if any of the previous ifs returned a value greater than 1 (2,3) that I would get this value through the chain this way but no joy

That makes sense, I give that a go.


Ok, not sure if I am doing this right.
In the top image each value hasa probability of 25% of being chosen right ?
In the bottom image by adding more uniform blends to preferred values I am increasing their chances to 30%.
So Subsequently I will need to add in a couple more steps to weight things differently.

Is this the way to do it or is there some easier (possibly built in) way ?

Substance DesignerSubstance Designer - Discussions - Weighted Randomizing
 on: April 18, 2016, 01:27:01 pm 
If I want to weight different values with a higher percentage in a function -- how would I go about that ?
So instead of completely random I want the values to be chosen in a weighted manner:

Value 1 = ~ 50%
Value 2 = ~ 20 %
Value 3 = ~ 20%
Value 4 = ~ 10%

So if I have a really unique element that is clearly identifiable as repeating I can keep its occurence to a reasonable minimum.
If I were to script this I would do a for loop but I don't see a way in the function graph to do loops.
Is it even possible to do this ?


question - I can't seem to login into Substance Feedback anymore.
Whenever I click the 'sign in' button I get 'My Account' opened -- where I am already signed in.

Am I doing something wrong there ?


Thanks for the ultra-fast answer !
(was refering to specular reflectance value btw when I was taking about reflectance of 0.6)

I have a little bit of an Implementation Question regarding Metalness Workflows.

In cases of Metalness = 1.0 the reflectance value is taken from the albedo map - so is the specular color to tint the highlights.
My question is if the specular color is taken at the same value that it is in the albedo or is it set to 1.0 first and then used ?


A Pixel in my Albedo Map has a Hue/Sat/Value of 0.6/0.6/0.6
With metalness = 1 the reflectance value is 0.6.

But is specColor

a) 0.6/0.6/0.6


b) 0.6/0.6/1.0 (to not dim the specular) ?

Thanks, Jens

Pages: 1 ... 4 5 [6]