Playing around with noise texture in nodes, i had improper behaviour from what i expected.
As for some other nodes in wich i found bugs by examining them through my oscilloscope the noise texture node has a strange behaviour; though am not sure it is a bug…
When i hear ‘noise’ i understand randomness and unpredictability.
In noise texture nodes, there’s indeed unpredictability ( i didn’t find repetition ) but there are some strange behaviour when you examine the output data.
Lets talk only about the value output ( fac ) with seems to be an unsaturated version of the color value.
1st of all it has to be mentionned that the output value never reaches 1 nor 0 !!!
I noted the limits of the output wich are about 0.25 for low and 0.75 for high. ( yellow marks.
Moreover, you can see with this high density view that there seem to be some ‘steps’ in the signal. this is more obvious when dilating the time base:
i noted the ‘steps’ in yellow.
And as there are those kind of deterministic leves, there are also pattern repetitions like the ones i surrounded in red or green.
What is you opinion about this ? Am i wrong to be surprised of the ‘unexpected’ behaviour of this noise texture ? Or is it simply normal this texture would raise only 3 or 4 peaks in a Fourier transform ?
Finally i simply found the real noise texture wich is called white noise texture.
This one gives proper and expected output: a real random noise generator
Thanks for your lit answers
And happy blending !