Home Products Download Order Contacts


Subject: Re: ACR v3.2 D2x white balance still not correct

Not to stir the pot, but hopefully to offer an explanation for the disparate WB numbers in various software packages.

The CIE definition of CCT (coordinated color temperature) in Kelvin degrees is based on spectral measurements of various light sources. This is not technically a blue/yellow and green/magenta axis, but when plotted the Kelvin line forms a curve across both axes on a gamut diagram. In other words, 3200K and 6500K each define a unique point on each of these axes.

That said, even the CIE will say that the standard illuminant matches are rarely achievable in nature or with artificial light. Even the standard blackbody radiator is based on a mythical material. So it does make sense for software to accommodate separate controls for each axis. And I can easily accept approximations in the numbers displayed.

The ISO TIFF/EP standards do support a LightSource tag (33423) for Kelvin degrees recording. But no one is using it. Private MakerNotes are not reliable unless they are publicly documented. My opinion is that we should be pressing for better “open” standards.

On the other hand, practical white balance techniques and settings are not that difficult.

Cheers, Rags :-)


View All Messages in adobe.photoshop.camera.raw

ACR v3.2 D2x white balance still not correct =>


Copyright © 2006 WatermarkFactory.com. All Rights Reserved.