Comments, please... lighting, color, sharpening

[Amamba]

380689

380690

[Amamba]

And this one
380691

All images were taken on an overcast day, ISO400, with 55-250IS. All sharpened with USM, some selective color (green was coming out way too yellow), some contrast / brightness / levels adjustment. The bear's white muzzle was completely blown out for some reason (too much contrast with black fur ?) so I cloned it out from a bit of fur at the bridge of his nose. How are the resulting photos ? What could be done better ? (I didn't bother much with composition so it's probably not quite there yet).

[Robert_Lay]

#1 is badly overexposed. Large areas of highlights without any detail.
I would also think that a bit more of the surroundings would help, but you can only be sure by trying it (a little late now).

I can't tell whether that is a white log or a white strip of sand in #2. Whatever it is, it's overexposed. I would crop off the right hand one-fourth of the picture. That stuff in the upper right is very distracting and takes the eye off of the bird's head.

#3 would probably look better as a vertical format - there is just too much emptiness at either side. The colors seem to have virtually no brilliance or contrast. Yes, obviously an overcast day. That's when you have to struggle to up the contrast in post processing.

I can't help but think that you are not using the Highlight Alert in the Histogram display - it should have been blinking on shots 1 and 2. It does not pay to disregard that display in that it can save a lot of pictures by warning you about blown out highlights.

[Amamba]

#1 is badly overexposed. Large areas of highlights without any detail.
I would also think that a bit more of the surroundings would help, but you can only be sure by trying it (a little late now).

I can't tell whether that is a white log or a white strip of sand in #2. Whatever it is, it's overexposed. I would crop off the right hand one-fourth of the picture. That stuff in the upper right is very distracting and takes the eye off of the bird's head.

#3 would probably look better as a vertical format - there is just too much emptiness at either side. The colors seem to have virtually no brilliance or contrast. Yes, obviously an overcast day. That's when you have to struggle to up the contrast in post processing.

I can't help but think that you are not using the Highlight Alert in the Histogram display - it should have been blinking on shots 1 and 2. It does not pay to disregard that display in that it can save a lot of pictures by warning you about blown out highlights.

Thanks !

I did have a highlight alert on. I saw what was going on but it left me puzzled. I used Evaluative Metering, +1/3 EC to make up for XTi's tendency to underexpose. Shot most pics at ISO400, Av, set to about 5 or 5.6. This is what I usually do. Yet a lot of shots that day seemed to have portions of the photo blown out while the rest seemed OK, and reducing the EC made the shadows look bad. It didn't help that I took kids to the zoo alone (wife was working) so I didn't have time to fumble with controls or take a second shot.

Anyway, what I can't figure out is why on the overcast day it would choke on slight highlights, the blown out areas weren't that bright. In the bear shot I thought it was the fur, there were some others where it made no sense at all...

[Robert_Lay]

Thanks !

I did have a highlight alert on. I saw what was going on but it left me puzzled. I used Evaluative Metering, +1/3 EC to make up for XTi's tendency to underexpose. Shot most pics at ISO400, Av, set to about 5 or 5.6. This is what I usually do. Yet a lot of shots that day seemed to have portions of the photo blown out while the rest seemed OK, and reducing the EC made the shadows look bad. It didn't help that I took kids to the zoo alone (wife was working) so I didn't have time to fumble with controls or take a second shot.

Anyway, what I can't figure out is why on the overcast day it would choke on slight highlights, the blown out areas weren't that bright. In the bear shot I thought it was the fur, there were some others where it made no sense at all...

Please search for my previous postings in which I have described the shoulder of the transfer characteristic of the XTi. The shoulder is so abrupt that you cannot get away with any over exposure at all.

[DrFil]

try shooting in M mode ;)

and listen to Bob, he knows what he's talking about.

also, just think it's funny that you want comments on "lighting"...even though you had no control over it.

[Robert_Lay]

Here is the link to my PhotoTool screen shot that illustrates the abrupt shoulder in the XTi's transfer characteristic:
http://www.zaffora.com/W9DMK/XTiResponse.jpg

[Amamba]

Thanks ! I will read this.

DrFil, I misworded my question, I was really interested in the brightness / contrast etc. of the main subject (PPd) not the lighting that I indeed had no control over.

As for the contrast of shot #3 - I tried to beef up either the contrast or the saturation but didn't like the results, it was either so contrasty the eye was blending into the brown band on his head, or there was way too much "punch" in the b/g.

A couple additional questions:

1) Would the right way to take the photo #1 be to expose for the grass (leaving the fur dark) and then try to bring up shadows in DPP (I have PS7 so no S&H tool there).

2) Do 40D/50D have better characteristics than XTi in that respect ?

[bfree32]

The reason for the overexposure is that the evaluative mode sensed a lot of black in the scene with the bear and background. When it tries to compensate for that, the lighter parts get blown out.

[Amamba]

The reason for the overexposure is that the evaluative mode sensed a lot of black in the scene with the bear and background. When it tries to compensate for that, the lighter parts get blown out.

Same question, though - wouldn't exposing for lighter parts result in lost shadow details ?

[Robert_Lay]

Same question, though - wouldn't exposing for lighter parts result in lost shadow details ?
The question is too general. A careful study of the histogram of the captured image is a good way to get answers to that issue.