# Re: When neutral 50% gray isn't 50%: elementary gamma question

From: Chris Cox (ccox_at_mindspring.com)
Date: 10/10/04

```Date: Sun, 10 Oct 2004 23:59:39 GMT

```

In article <ck2bml\$mtb\$1@netlx020.civ.utwente.nl>, Marco Al
<m.f.al@student.utwente.nl> wrote:

> Graham Jones wrote:
>
> > I'll try one more time. Let's say the 'native' gamma of a monitor is
> > 2.5, meaning (luminance out) is proportional to (voltage in)^2.5. When
> > people talk about gamma=2.2 for PC monitors they mean that a small
> > correction is applied to make the monitor behave like (pixel value)^2.2.
>
> No the problem is noone agrees with eachother :)
>
> Find the common denominator between these :

Other than the fact that they're all correct?

>
> http://www.poynton.com/notes/colour_and_gamma/GammaFAQ.html
>
> http://www.w3.org/Graphics/Color/sRGB
> http://www.mi.tj.chiba-u.jp/download/doctor/pdf/PhD_Katoh_wColor.PDF
>
> http://www.cis.rit.edu/research/mcsl/research/PDFs/GibsonFairchild.pdf
> http://www.art-si.org/PDFs/CRT_Colorimetry_Ap_2002.pdf
>
> > (sensation out) = (luminance in)^(1/1.8).
>
> Perception of differences in luminance is approximately logarithmic, so
> it should cancel out against gamma. Some of the images at the following
> page tell a different tale though :
>
> http://aim-dtp.net/aim/evaluation/perception/perception.htm

Because Timo's site is a well known misinformation web site - the
images are designed to lie.

Chris