Landscape Processing Opinion Please

[ajohnw]

It's worth thinking about this aspect as well. Where features are adjusted to the same size and depth is involved lenses do alter perspective.



Buried in folklaw somewhere are standard lenses around the diagonal of the format. The one shown in this shot is 55mm which isn't that uncommon. The idea is that what ever size the full frame final image size finishes up at it's about as much as the eye can take in at one time if viewed from a suitable distance. Some pundits refer to an included angle of 60 degrees but we can clearly take in more than that subject to what is in the scene. Hence wide angle plus some feature.

The eye perspective comes into it as well generally reckoned to be 100mm or some what shorter. Personally I feel that detail comes into this as well. In other words it's more inclined to show the detail in the area covered along the lines of what we see by eye. Some people describe that as having a more dramatic effect. This cropped up some where else so while out I took a shot to show the effect. 1st day I really used the camera. Tree and people are in roughly the correct proportions and detail disappears at a reasonable rate.



Digital cameras see linearly. People mention 12 bit 12 stops etc but life isn't that simple. The numbers go 0.2.4.8.16.32.64.128.256,512,1024 etc The "light change" in the lowest "stop" is a count of 2. The next one a count of 4 and so on. Things probably start getting sensible when there is a count of say 32 between "stops". Not much gradation can be shown in a count of 2 and maybe too much when the count change is 512. Relating bits like this to visual brightness changes doesn't make much sense. I've been trying to find out exactly what raw conversion software actually does to get round this. I wanna know exactly what happens to my X bits of colour. A browse round dynamic range plots on dpreview provides clues such as the dark end is heavily compressed in most cases and more recently the bright end as well. This is the only way the range can be squeezed into smaller colour spaces. Anyway this all shows why an eye view is a difficult thing for a digital camera in it's current forms to achieve.

One way I have played with to get "real" clouds is to produce a very dark image from raw and then blend that into the normally developed shot. In this case it was the only way to get the very wispy mist like clouds top right to show as they were. It also corrected the compression needed to squeeze the the shot into a jpg. The tonal range of the dark shot could also be increased if needed for a certain look as there is plenty of headroom available. I'm inclined the feel that this can be easier than selecting the sky and working on that. I've done that OK etc but. The final appearance depends on how much of each image is added into the final image and subsequent brightness adjustments. I was rather pleased with one I did properly. This was the shot I blended in. The wispy bits finished up as just being visible.



John
-