Re: [Openexr-devel] OpenEXR 1.5.0, OpenEXR-Images 1.5.0 released

[Florian Kainz]

Statements to the effect that human perception of physical quantities
such as luminance, force or sound pressure is logarithmic do not refer
to data encoding.


Chris Cox wrote:
> Florian;
>
> I figured it was just a terminology problem.
>
> Normally, when describing an encoding, the baseline reference is photons.
> Linear is normally equated with gamma 1.0, and the values are proportional to 
> photon incidence.
> Logarithmic normally means that you have taken the log of a gamma 1.0 signal, 
> and that corresponds fairly well to human perception.
> Gamma encoded means that you have adjusted the gamma 1.0 signal by pow( 
> photons, 1/X), which also correspons well to human perception when X is about 
> 2.0.
>
> If the luminance is logarithmicly encoded (or gamma ~2.0 encoded), then it is nearly 
> linear with respect to human perception (L*).  But saying that is is "perceptually 
> linear" is rare because of the above conventions.
> Saying that it is "perceptually logarithmic" means that is is logarithmic with 
> respect to human perception, and would imply log( log( photons) ) ...which is usually a 
> bad thing.
>
> Chris
>
>
>
> -----Original Message-----
> From: Florian Kainz [mailto:address@hidden
> Sent: Tue 12/19/2006 7:43 PM
> To: Chris Cox
> Cc: address@hidden
> Subject: Re: [Openexr-devel] OpenEXR 1.5.0, OpenEXR-Images 1.5.0 released
>  
> I don't think I have the quantization backwards.  If you want to
> see for yourself, go to function insertChannels() in ImfRgbaFile.cpp,
> play with the settings of the the pLinear flags, and convert files
> such as WideColorGamut.exr and GrayRampsHorizontal.exr from the
> OpenEXR sample image collection to luminance/chroma format with B44
> compression (at this point, only B44 pays attention to the pLinear
> flags).
>
> Regarding terminology - I think calling luminance perceptually
> logarithmic is appropriate since perception of luminance obeys
> obeys the Weber-Fechner law, at least approximately.
>
> Florian