I had an interesting comment this week on my recent Guildford Castle photo (thanks to Andy for allowing me to share):
An interesting point: the white balance looks “normal” but I’m wondering if the pinky colour was genuinely there (it was just past sunset, after all). Daylight WB is usually pretty accurate and will show colours the way they actually / “scientifically” were rather than the way the eye perceived it – if there was an overall pink tint the eye would neutralise it to some extent but of course the camera will blindly record what it sees…
My feeling on white balance is that it’s a very delicate thing. I tend to process in such a way that what I see on screen approximates what I saw on site. In this instance, the rosiness must have been there (scientifically), but my eyes did not really perceive it at the time, or if they did it was at a much lesser intensity than what I got when processing the image as daylight. I thought it was worth trying the alternative as well and see what people think. I must admit I liked both.
So here’s the alternative process; everything is exactly the same as the original, except that this time the white balance was set to daylight when doing the raw conversion in DPP. For side-by-side comparison, I’m including again the originally processed version on the right. So, which do you prefer?