cgardner
Offline
• • • • •
Upload & Sell: Off
|
WCMTWS wrote:
Cgardner, I agree. I am not real big on post processing images. More of a what you see is what you get.
Then I suspect your monitor brightness is set too high because your images "by the numbers" are underexposed rendering things like specular highlights seen as 100% reflectance and frost / snow which reflects 95-98% of the light much too dark in the images you are posting.
Here's your second shot opened in Levels:
The gap on the right indicates at capture you underexposed the file by about 1-1/2 stops below nominal / normal with highlights pegged at their correct reflective values. Specular highlights in photos should be at 255 (clipping), solid whites like snow in highlights at 245-254, and snow in shade in the 230-245 range in the file to reproduce what is seen in person by eye.
What can happen if your monitor it overly bright is you see correctly exposed files as being over and underexposed files as being correctly exposed. That results in you either capturing them underexposed so they look "normal" on the monitor, or taking correctly exposed files and making them underexposed in PP.
Evaluating the the file in Levels is the first thing I do when I suspect an exposure problem. With an out of camera file the width of the histogram represents sensor range which for most digital cameras is about 6-7 f/stops of scene brightness. That how based on the size of the gap I see I estimate 1 to 1-1/2 stops underexposure.
Scene ranges vary with lighting conditions. On a sunny day the limited range of the sensor results in loss of detail in shadow because sunny clear days have a range of 10-12 f/stops depending on the content and angle of the sun and shadows it casts on darker objects.
But on a foggy day the same scene will only have a range of 6-7 stops, nearly the same as the camera so theres really no reason not to exposed it to capture a full range of detail. That's done by exposing guided by the playback clipping warning and histogram. Raise exposure until you see the brightest areas in the photo (in this one the upper left corner) start to clip.
What I did in my edit of your first shot is mostly correct the underexposure at capture to render in the file the actual reflectance values that where there when you took the shot, anchoring the correctly on the specular refections on the road, which in a print would be reproduced with the white paper base and in a digital file get pegged at 255.
In Levels to find the point where clipping starts I hold down the alt/opt key and move the highlight slider left. The clipping starts to occur when the slider is moved to the point where the bars on the histogram have some amplitude.
Each bar on the graph represents 1/256th of the tonal range of the sensor. So if the sensor and record 7 stops between 0 black and 255 white every 36 units on the scale equals one stop of sensor range. I blocked the number with the arrow but its 200. That means if what you posted is the out of camera capture your highlights at capture where 55 units or 55/36 = 1.52 stops under exposed "by the numbers".
More significant in that shot, given the low contrast lighting conditions, is the fact that by underexposing the highlights by 1.5 stops you pushed the shadows in the scene below the point where the camera could record detail in them.
In Levels it is possible to bring out more detail in midtones and shadows where the camera recorded some detail above black by moving the middle slider left as I did above, with the results seen below...
But nothing can be done in PP to add detail into the shadows.
The lesson here? When possible always try to capture the full tonal range of the scene. First exposed the highlights per the clipping warning and take a shot. Look at the LEFT side of the histogram in that test shot. If the bars are high on the left and running off it indicated SCENE EXCEEDS SENSOR and loss of shadow detail. When that occurs the options are: 1) add flash fill to the foreground (if feasible); 2) Shoot on a tripod and bracket exposures, then blend in PP, or 2) recompose the scene so what is in the viewfinder doesn't exceed sensor range.
In those those lighting conditions had you exposed the highlights correctly there likely would have been a gap on the left indicating the scene not only fit the sensor it was even shorter, a rare occurrence outdoors except in heavy overcast and fog. You should of had no problem capturing detail everywhere from highlight to deepest shadow had you simply exposed the highlights nominally.
If you can capture the file with a full range of detail is is very easy to manipulate it in PP to get the desired result. It's not a perfect capture medium so its not realistic to expect the best results without some manipulation. Even Adams manipulated results. He started by recording a full range of tone as I'm suggesting here, but within the range between highlight and shadow detail he would use red filters at capture to alter contrast of the sky, green filters to make foliage brighter and dodge and burn prints to alter tonal values in much the same way I did in my edit.
Adams called the process "pre-visualization". Where you start with the baseline of what your eyes and brain perceive see as they scan from highlight to shadow, the add a creative twist like making the sky abnormally dark — all based on understanding how to produce those desired result by manipulating the technical capture and printmaking processes. Pressing the shutter isn't the end of the creative process its just the starting line. Where you finish is only limited by your technical skills and your imagination 
Here's your first shot revisited...
Here I adjusted Levels until with the opt key held down as I moved the slider I forced clipping in the reflections on the road. That resulted in the brighter sky getting washed out...
But then in Photoshop I layered the adjusted copy with the original and blended them with a mask on the original....
Here's the result of that few minutes of moving slider and erasing the mask...
That's probably closer to how it is perceived in person — I've seen similar conditions in person — but the camera can't record what the adaptive eye and brain perceive due to its shorter fixed range.
What you could have done at capture to get results similar to the edit is use a graduated ND filter over the lens to darken the sky, or shoot from a tripod and bracket exposures. Either would have been better than how you captured the image because there would be enough exposure in the shadows with either to record the shadow detail lost in yours and that can't be brought back in PP.
|
