openexr-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Openexr-devel] f-stops and steps. How are they related?


From: Yves Poissant
Subject: Re: [Openexr-devel] f-stops and steps. How are they related?
Date: Sat, 8 Jan 2005 11:22:56 -0500

As a follow-up, I found it interesting that in order to view this page correctly:
http://www.sgi.com/misc/grafica/setup/index.html
I have to set my gamma (as measured on a gamma strip pattern) to 1.3. The gamma setting I naturally feel most confortable with anyway.

I just observed that I already had stored gamma presets in my video card color settings. 1.3, 1.8, 2.2 and 2.6. But I don't remember why I set them that way. The more I read about Gamma setting the more I'm confused. I'm not even sure which setting I used for RAW digital photo processing anymore because I have kept no note of it and I may very well have changed the gamma setting since.

I know that most photographers sites that care to supply a gamma adjustment instruction, do recommend a gamma of 2.2. But Frankly, I find this way too dark. And I suspect this is to compensate for the inherent gamma curve that are embedded in photos, be it through digital camera gamma transform or through scanners gamma transforms. This camera embedded gamma transform was *extremely* obvious when working with RAW digital camara data which is, as I understand it, the most linear data we can get from a digital camera. So I suspect I probably must have used a gamma of 1.3

As a note, we tested 3 brand of pro digital cameras and 3 different Canon models (we found Canon makes the best camera for RAW data quality BTW, mainly because they use CMOS sensors which have a very good s/n ratio in very low levels) and we found that the sensor data is not that quite linear as evidenced when comparing the intensity distributions from camera to camera but it was linear enough for our needs. For those who wonder, RAW data from good pro level digital cameras gives you at least 2 additional f-stops in the low range portion and 2 f-stops in the high range portion of the captured photo.

Also, for convenience, the pattern you instructed to do in your post, that is :
    BBBBBBBBBBGGGGGGGGGG
    WWWWWWWWWWGGGGGGGGGG
    BBBBBBBBBBGGGGGGGGGG
    WWWWWWWWWWGGGGGGGGGG
    BBBBBBBBBBGGGGGGGGGG
    WWWWWWWWWWGGGGGGGGGG
    BBBBBBBBBBGGGGGGGGGG
    WWWWWWWWWWGGGGGGGGGG
    BBBBBBBBBBGGGGGGGGGG
    WWWWWWWWWWGGGGGGGGGG
    BBBBBBBBBBGGGGGGGGGG
    WWWWWWWWWWGGGGGGGGGG

Can be found in several places on the net:
http://graphics.stanford.edu/~ericv/gamma/gamma.gif
http://www.spurgeonstudio.com/Oldindex.htm

Also, when I was working on RAW digital camera photos, I found this site
http://www.normankoren.com/digital_tonality.html
Where you can find this ingenious calibration chart:
http://www.normankoren.com/makingfineprints1B.html
The advantage I found of this new chart is that it will work on any monitor. Not only on good quality monitor.

Yves Poissant
Hash Inc.
www.hash.com

P.S.: The work on RAW digital data I mentioned above was not related to Hash Inc.




reply via email to

[Prev in Thread] Current Thread [Next in Thread]