Search Unity

  1. Welcome to the Unity Forums! Please take the time to read our Code of Conduct to familiarize yourself with the forum rules and how to post constructively.

VFX GRapn noise range disabmiguation

Discussion in 'Visual Effect Graph' started by deadfallgame, Sep 27, 2020.

  1. deadfallgame

    deadfallgame

    Joined:
    Oct 1, 2014
    Posts:
    16
    What exactly means range in noise Operator? Suppose I need noise values within rande [0,1]. When I use this range in block params I get different result.
     
  2. JulienF_Unity

    JulienF_Unity

    Unity Technologies

    Joined:
    Dec 17, 2015
    Posts:
    284
    Hi, range would be the range of the output of the noise operator. If you set it to [0,1], you'll get an output between 0 and 1 etc.
     
  3. deadfallgame

    deadfallgame

    Joined:
    Oct 1, 2014
    Posts:
    16
    no :) see my picture below
     
  4. deadfallgame

    deadfallgame

    Joined:
    Oct 1, 2014
    Posts:
    16
    Here some test
     

    Attached Files:

  5. JulienF_Unity

    JulienF_Unity

    Unity Technologies

    Joined:
    Dec 17, 2015
    Posts:
    284
    Try to increase the frequency or the range of the input. Here you are sampling between 0 and 1 with a frequency of 1, which means you're likely not hitting the whole range
     
  6. deadfallgame

    deadfallgame

    Joined:
    Oct 1, 2014
    Posts:
    16
    Changed frequency to 1000, results cganged but slightly. New pic in attach
     

    Attached Files:

  7. JulienF_Unity

    JulienF_Unity

    Unity Technologies

    Joined:
    Dec 17, 2015
    Posts:
    284
    Well value and perlin noise are in the correct range.

    Cellular looks strange though. It's expected that the distribution along the range is not linear., cellular noise is essentially a distant field. 1 should be the theorical max but it really does seem far from being hit here.

    Anyway, range property is the guaranteed range of the output noise, it does not ensure values are uniformly distributed within it.
     
    Last edited: Sep 29, 2020
  8. deadfallgame

    deadfallgame

    Joined:
    Oct 1, 2014
    Posts:
    16
    Well, perlin values slightly differs from range and celluar differs sufficiently. By the way these results I collected when one octave set. If I set more octaves even value noise will deviate from bounds. Seems like normalization problem. Should I report a bug?