New Tutorial: Gamma Correction & Digital Tones

[McQ]

This article is definitely more technical than average, but it's an important concept for people really wanting to know how their image tones are recorded:

Tutorial: Understanding Gamma Correction & Digital Tones

It is also probably one of the most misunderstood. Regardless, understanding how gamma works can improve one's exposure technique, in addition to helping one make the most of image editing.

As usual, please let me know if you feel anything is unclear, if you notice any typos or just want to add something from your own experience.

Thanks!

PS: this was the tutorial that was recently referred to in this thread: Gamma: Overrated?

[Peter Ryan]

Wow Sean. Sometimes I struggle to stay in touch with some of these concepts but pleased they are here to read again and again as I grow into this tehcnology.

[Jack Hogan]

Excellent tutorial, Sean.

[Donald]

Sean

I don't know enough to know whether you've missed out any key points or if you have incorporated any inaccuracies. So, my learning is totally in your hands!

All I can contribute is that, on the basis this is making sense (but I'll have to read it a few times to get it to sink in), it is therefore good.

[arith]

Instead, we perceive twice the light as being only a fraction brighter

This may well be true; but essentially mostly eyes see everything brighter. There is an exponential relationship between actual light and perceived light.

I think what you are saying is; gamma correction is a tool for storing information uniformly, please if I have got it wrong do not despair since I'm no longer as quick as I was and this is after a first reading.

So I thought; if I move the middle slider in PSE levels, or change the gamma in Picturenaut tone mapping, if I'm using a calibrated screen can everybody else using calibrated screens see the same image?

I think maybe they can since the adjustment is the same in their monitors, 2.2 gamma, but I thought I had better check.

I must add that your tutorials are much better than any I've seen anywhere else.

[Andrew Mckay]

thanks Sean
Just what I was looking for.
cheers
Andrew

[rick55]

Hi, Sean;

Great subject to tackle, and great material. Very clean and clear.

It may be unnecessary, but I wonder if it would be helpful to structure it more around the whole concept of calibrating your system, or achieving a faithful result. That information is there, but it seems (to me) a bit woven into the narrative, and not very obvious.

If closing that loop of scene to reproduction with gamma of 1.0 is set up as the processing goal, it would make it easier to bring in other considerations in other tutorials like multiple monitors and printers.

Cheers,
Rick

[McQ]

It may be unnecessary, but I wonder if it would be helpful to structure it more around the whole concept of calibrating your system, or achieving a faithful result. That information is there, but it seems (to me) a bit woven into the narrative, and not very obvious.

If closing that loop of scene to reproduction with gamma of 1.0 is set up as the processing goal, it would make it easier to bring in other considerations in other tutorials like multiple monitors and printers.

Hi Rick -- thanks for the feedback. Yes, the narrative definitely follows a lot more of the theory/definitions than putting it to use in practice. The monitor calibration tutorial was partly aimed at the applied aspects of display gamma, but it certainly doesn't address all of the topics that you mention. What I'd like to do is add more "practical" articles to the color management section, and in doing so have gamma be one of the things that is addressed. However, I'll see what I can do to better emphasize the system gamma = 1.0 goal of calibration, because as you say, this will better set up subsequent topics.

[arith]

I don't think the graph is helping much; I don't really know what is meant by 100% luminance unless it means the brightest thing we can see.

Technical Note: Gamma is defined by Vout = Vingamma , where Vout is the output luminance value and Vin is the input/actual luminance value. This formula causes the blue line above to curve. When gamma>1, the line arches upward, whereas the opposite occurs with gamma<1.

If Vin is a fraction, and a percentage is a rational less than 1 with representative a fraction for (simple) percentages then 1/4 to the power 1/2 equal to the square root of 1/4 =1/2, as illustrated approximately in the above graph for 1/(2.2).

This seems to contradict what you are saying.

[McQ]

This seems to contradict what you are saying.

Thanks, the >,< signs were flipped. The page has been updated.

[_GUI_]

Hi Sean,

As I already privately commented with you before the tutorial was published, my main disagreement about most gamma related articles I found so far, is that they focus on the way our eyes perceive light as the reason for gamma to exist, instead of the non-linear response of typical output devices.

Although you begin the tutorial referreing to our eyes non-linearity, a careful reading of your article makes things clear. I specially celebrate one of the last comments: "Is Gamma Required? No, linear gamma (RAW) images would still appear as our eyes saw them -- but only if these images were shown on a linear gamma display. However, this would negate gamma's ability to efficiently record tonal levels."

Strictly speaking, the only real usefulness of gamma today is to achieve optimal image encoding when using integer formats such as 16-bit TIFF and specially 8-bit JPEG.

Regards