Gamma curve
1 - 13 of 13 Posts
Yes gamma is the rate of transition from black to white.
The basic power formula is dead simple
% Luminance = % simulus ^ (gamma power)
so at 50% stimulus at 2.2 = .5^2.2 = .217 or 21.7% of total luminance
same at 2.4 gamma .5^2.4 = .189 or 18.9% of toal luminance
If you've still got a graphing calculator laying around from high school y=x^2.2 versus y=x^2.4 over x=0 through 1.
The basic power formula is dead simple
% Luminance = % simulus ^ (gamma power)
so at 50% stimulus at 2.2 = .5^2.2 = .217 or 21.7% of total luminance
same at 2.4 gamma .5^2.4 = .189 or 18.9% of toal luminance
If you've still got a graphing calculator laying around from high school y=x^2.2 versus y=x^2.4 over x=0 through 1.
Lately I've seen some questionable measurements from different light reading instruments posted by someone who is a calibrator. I'm concerned that if a professional's instruments can have substantially different results that I can't be confident that anything, including gamma is being correctly set by these calibrators. It's concerning because I very much want a calibration done for color and gamma, but I'm very skeptical the more research I do on this.
A good calibrator should have good gear in good shape. We have a ton of gear come through our calibration lab. A lot of it is pretty close weather it's coming from a factory or from the field, all of it is close when it heads out the door.
Ask your calibrator when the last time they had their tools calibrated, if they have current NIST certificates for their gear. NIST certs are good for a year, that should give you some certainty on the quality of calibration you'd be getting.
A good calibrator should have good gear in good shape. We have a ton of gear come through our calibration lab. A lot of it is pretty close weather it's coming from a factory or from the field, all of it is close when it heads out the door.
Ask your calibrator when the last time they had their tools calibrated, if they have current NIST certificates for their gear. NIST certs are good for a year, that should give you some certainty on the quality of calibration you'd be getting.
rec709 and smpte don't have prescribed home veiwing gamma.
sRGB does have a specific gamma formula, but it's a little more complicated as it has a linear tail for the first 8% or so.
If you download CalMAN and pull out a luminance graph you can play with changing the gamma setting and see how they look different on the chart.
Quote:
Originally Posted by clsidff3g /forum/post/20885929
I wanted to see,in graphic form, the difference between a gamma of 2,2-2,3,2,4,rec 709,smpte 240,srvb,it's possible?![]()
rec709 and smpte don't have prescribed home veiwing gamma.
sRGB does have a specific gamma formula, but it's a little more complicated as it has a linear tail for the first 8% or so.
If you download CalMAN and pull out a luminance graph you can play with changing the gamma setting and see how they look different on the chart.
Joined
·
34 Posts
According to Poynton Rec.709 describes the transfer function. See http://www.poynton.com/PDFs/GammaFAQ.pdf
However it does not specify any values for gamma which is the real problem, since deviations in gamma in the presentation (wrt the gamma used during colour grading) will cause deviation in hue/tint and colour satuation in the presentation compared to the expected outcome.
However it does not specify any values for gamma which is the real problem, since deviations in gamma in the presentation (wrt the gamma used during colour grading) will cause deviation in hue/tint and colour satuation in the presentation compared to the expected outcome.
Joined
·
3,458 Posts
Another problem is that measuring windows to look at gamma does not necessarily result in what Poynton is talking about in his theory-based discussions of gamma. Typical window measurements cause changes in average picture level, which can affect the light output of some displays. Various displays react to changing average picture levels differently, so identical gamma graphs may not actually result in similar gamma, by the way he uses the term.
Another problem is that measuring windows to look at gamma does not necessarily result in what Poynton is talking about in his theory-based discussions of gamma. Typical window measurements cause changes in average picture level, which can affect the light output of some displays. Various displays react to changing average picture levels differently, so identical gamma graphs may not actually result in similar gamma, by the way he uses the term.
I take this opportunity to ask you what you think of videoequalizer http://www.avsforum.com/avs-vb/showthread.php?t=1042160 have you had good results,this seems a good idea to have a linear gamma.
Joined
·
34 Posts
Tint will be affected by choosing a wrong gamma which can be seen by calculating the angle between the chromaticity coordinates (x; y) wrt the whitepoint (or by calculating H) for the same RGB stimulus with two different gamma values.
As far as I can tell the only colours that will not be affected in the gamut are those placed on the straight lines stretching from the whitepoint to the primaries and secondaries.
Tint will be affected by choosing a wrong gamma which can be seen by calculating the angle between the chromaticity coordinates (x; y) wrt the whitepoint (or by calculating H) for the same RGB stimulus with two different gamma values.
As far as I can tell the only colours that will not be affected in the gamut are those placed on the straight lines stretching from the whitepoint to the primaries and secondaries.
1 - 13 of 13 Posts
Join the discussion
