Difference between Fit Range / Unclamped

[Georgie]

I'm sure it's been brought up before but for whatever reason I can't find a single answer to this on Google. Searching for "fit range unclamped" returns only 2 results o_O

There's no documentation on this VOP either. 

 

Correct me if I'm wrong but Fit Range works as a fit expression. You feed it the source and the destination and it basically stretches or squeezes the values. So if I have a Sparse Convolution noise pattern and I feed it into a Fit Range VOP and set the Source Min / Max to -1.7 and 1.7 respectively, I'll get that remapped in the 0 to 1 range, correct ? 

If I set the Source Min / Max to -1 and 1 , that'll simply 'cut out' the values under -1 and above 1, correct ?

 

So what does the Fit Range Unclamped do exactly ? I suspect the whole 'source' thing confuses me and I think of it the same way I think about the regular Fit Range >.> 

How does it do the remapping ? 

 

Thanks :}

[rayman]

It does the same thing as regular fit() without clamping the result if the input is outside the input range.

[ranw0477]

I am also confused by this. So can anyone explain in what cases the unclamped behavior can be useful? 

[Toledor]

Clamped is useful to fit and make sure no values exceed your new limits. Unclamped fits and extrapolates.

Here you have a small clip showing a basic visual explanation. You can see that both clamped and unclamped have the same slope but clamped gets blocked between newMin and newMax.

The uses for this are many, I hope it's a bit clearer now.

Cheers!

JT_clampedUnclamped.mp4

[acey195]

if you have a value of 2 and you want to fit it from 0,1 -> 0,10

clamped will return 10, unclamped should return 20.

Its useful to to know a relative ratio of a value to another value in 2 different scales

[akelian]

Does anyone know an alternative to efit in hscript ? I want the same behaviour to scale down the sensitivity of an asset parameter, with unclamped range.

[f1480187]

@akelian:

(((value - omin) * (nmax - nmin)) / (omax - omin)) + nmin