What do you think of HDR?

Rayt

Rayt

Nonplayer Character
Local time
2:07 AM
Joined
Jan 19, 2006
Messages
3,077
I am a LR and PS newbie but thanks to the free time I have forced myself to scan my negatives and learn how to use it. Putting stuff on the Flickr and Apple cloud is a kind of insurance I guess.

Once I have saved my files into TIFF I make a small Jpeg copy for Flickr, Photo App for sharing. I have the PS app on my phone and it's so convenient I am always tempted to do stuff to the photo and one feature I find useful is HDR. Now I hate photos that overdo HDR looking like razorblades to the eyes but some mild HDR is ok I guess. On the other hand I like that full tonal look from a darkroom print and I am drawn to it when I do digital processing. Here are two one with mild HDR and the other one in plain vanilla. Please let me know what you think.

Agra44 by ray tai, on Flickr

HDR by ray tai, on Flickr
 
:) I’m looking at them from mobile phone.
First is fine, second is less natural.
 
Thanks for asking.

Sometimes when you stick your neck out by asking for comments, it can get cut off.

Whatever your vision, the looks you want to achieve means the most.

My process is to make the photographs using RAW capture with the camera.

Then I start the process stage with my iMac using Photoshop, keeping the RAW file after I work on it saving my work as a jpeg.

If I need to, I will save my RAW file as a psd which I can bring up later revealing the layers I did my processing.

To sum up, my work flow on computer, is work the RAW file, save my work as a psd then flatten the file and make a jpeg.

I don’t usually spend a great amount of screen time working on a file as I work hard to get it correct when I make the image in camera.

HDR is not for me nor are making tiffs.
 
I am just getting acquainted with the Monochrom and the files are certainly more flexible for tweaking. I’ll just keep the Raw files and am sure each attempt will hopefully look better as my computer skills improve. Scanned negs don’t have that pliability especially when less than perfect negs. I am glad I finally got the Monochrom.
 
The second is more of a 'modern' look, which may or may not be what you want.

In my brief excursion into real estate photography, I experimented with exposure fusion in Lightroom, which seems to produce a less artificial looking result than HDR. I quite liked many of the exposure fusion images I made.

For film, you could probably do the same fusion process with scanned images.
 
Done well, HDR can produce some amazing results and can be a good way to save highlights and shadows; done poorly, it's pretty ugly (in my eyes, anyway). Play with it and see if it gets you what you want.
 
HDR, like any other tool, is neither good nor bad. The question should be what do you think about the sensibilities of most HDR users, because that is what determines the quality of the results. If done well, HDR can produce stunning and very natural images. But if done poorly, the final photos can range from merely unnatural to abysmal. I'm afraid much of what's on public display is from the latter category.
 
I’ll find the file later but for exposure errors like two stops underexposure due to back lighting I have rammed up HDR to save the image when the shadow slider wasn’t enough. I suppose with the posted example if there is no need to use HDR then I should refrain. Thanks for your comments.
 
If done correctly it can look great. With exposure bracketing. Some people like sugar in their coffee or wear socks in their sandals. Enhancement sliders in post just doesn't sit well with my eyes.
 
Your image doesn't need HDR. It's not a high contrast scene. You've introduced quite a bit of noise, however. And there is quite a bit of dust/ dirty neg. artifacts that I would take care of first... YMMV. The first photo is perfect as is, in terms of contrast and dynamic range.
 
I think local dodging and burning can be a better approach than HDR in many cases. Dodging and burning can easily be limited to just the areas that need it without adversely affecting the rest of the image. The few HDR programs I have experience with seem to default to global application.

In the first image in this thread, I just don't see any need for HDR -- almost no lost shadow or highlight detail.

Below is an image I used to test what amount and quality of detail could be dug out of deep shadows using just dodging and burning without having to resort to hdr.

Leica M240

Straight out of camera.
M2403855 4 by Brusby, on Flickr


After a little dodging and burning. Far from perfect, but at least useable.

M2403855 by Brusby, on Flickr
 
brusby said:
I think local dodging and burning can be a better approach than HDR in many cases. Dodging and burning can easily be limited to just the areas that need it without adversely affecting the rest of the image. The few HDR programs I have experience with seem to default to global application.

In the first image in this thread, I just don't see any need for HDR -- almost no lost shadow or highlight detail.

Below is an image I used to test what amount and quality of detail could be dug out of deep shadows using just dodging and burning without having to resort to
Click to expand...

Thanks. I’ll spend a few more weeks with Lr tutorials before going any further.
 
Brusby, that is an astounding difference. I wouldn't have guessed that there was detail to be salvaged from the first image.

- Murray
 
Rayt said:
Thanks. I’ll spend a few more weeks with Lr tutorials before going any further.
Click to expand...

Here's a tip for a general approach that works well for me in Photoshop. First on the main layer, set the primary exposure and contrast so most of the image looks right. Then make one layer just focusing on the shadows and get them to look correct. Then do another layer just for the highlights. Finally, blend the layers to appear as seamless and natural as possible, adding only so much of the shadow and highlight layers as absolutely necessary. Keep the emphasis on looking natural or as close to how things appeared to your eye when you took the photo (if an approximation of reality is your goal).

I'm sure something similar could be done in Lightoom, but it's much easier with layers.

CMur12 said:
Brusby, that is an astounding difference. I wouldn't have guessed that there was detail to be salvaged from the first image.

- Murray
Click to expand...

Thanks Murray. It was a very high contrast scene and getting really dark. The sky in real life was pretty dark and starting to take on a bunch of almost surreal colors but there was still detail in the grasses in the foreground. So, I exposed the sky as bright as I dared, almost overexposed, just before it lost detail so I could get the most info in the shadows. Then in post, I burned in the sky and dodged the foreground using the method I described above.

In the finished image the sky looks almost fake, but it's as close as I could get it to the actual image and hyper vivid colors I saw when I took the photo.

jarski said:
and its from camera released in 2012 ;)
Click to expand...

It surprised me too.
 
M9 files are like that, too. As long as you're at a low enough ISO, you can push underexposed M9 files a few stops and they look acceptable. The amount of detail you can pull out of shadow areas is astonishing. One trick I like for sunset scenes like above is to raise the exposure for the whole image, then use a gradient filter to lower the exposure for the sky, or use the Dodge tool for the same effect.
 
The second picture is better to my eye. I like the shadow detail in the man's jacket and the gradation in the stonework or cement in the structure behind. The top picture is just a little too dark, to me.
 
The second image looks good, but it doesn't have the "feel" of the first, it's more stark. The HDR look in B&W isn't necessarily for high-dynamic range situations, it's more to pull out a lot of tonal contrast. I've used that look before, but I come back to those images and don't care for them as much as some others. An example,

P1050496 copy by Andrew Lossing, on Flickr

The processing, which was mostly done in On1 Effects, is a good way to pull a lot of detail out of a 16MP M4/3 Panasonic sensor, but it looks very overdone to me now. I think the key is to keep a neutral one on hand, like the OP first image, and consult it when editing the HDR version. If it were me, I would try to retain the soft dark feel of the man's clothing in the first image, while going just far enough to pull some detail out of it. My strategy is to edit the way I used to like, but then dial it down until it seems like just too little of an effect. In the longer term I feel that ends up looking best.
 
HDR is an ambiguous term. There are several ways to produce a final image when the dynamic range of a scene exceeds the camera's dynamic range.

Most of the time HDR means blending multiple images with auto-bracketed exposures and then rendering them with tone mapping.Tone mapping is used to compress the higher dynamic range of the blended image into a 16 bit format so it can be displayed and printed. The choice of tone mapping parameters is highly subjective. At one extreme one can use parameters that attempt to render an image that appears as though it is not from a blended image. The viewer is deceived. The viewer can't tell if whether or not the dynamic range for the scene's luminance was greater than the camera's dynamic range for a single exposure. At the other extreme are highly stylized images with surrealistic or symbolic renderings. The viewer is deceived because the rendered image does not attempt to portray reality. The image depicts someone's interpretation of reality to produce an emotional impact.

Surrealistic HDR rendering can be generated in seconds. It can be automated. Unfortunately, this degree of convenience often results in sloppy images rendered with muddy low-contrast regions, obvious color shift and color cast errors, and with halo and ghosting artifacts around high contrast borders or in high contrast regions.

HDR can also refer to blending multiple exposures without resorting to tone compression. This method was used by architectural photographers long before tone mapping techniques existed. Multiple exposures are manually blended using some form of layer masking (usually luminance masks). Eventually interiors and architectural photographers decided to call this technique exposure blending to differentiate it from tone mapping. Exposure blending is more time consuming and less convenient than tone-mapped HDR. The blending is usually selective and done region by region. Again, the goal is for almost all viewers to assume the final image is from a single exposure.

I used either weakly tone-mapped or non tone-mapped exposure blending for almost all my photographs of residential and commercial building exteriors. There were rare occasions when a space was incompatible with off-camera lighting and I had to use subtle tone mapping.

However, I can not abide gratuitous, surrealistic HDR. I can not abide the look of interiors photography images that use tone mapping primarily to eliminate the inconvenience of using off-camera lighting.

It turns out I benefited from tone mapping because my interiors photography clients were not tempted to use lower-cost photographers who worked very quickly making six exposure bursts per room and delivering automatically rendering low-quality, tone-mapped images.
 
You must log in or register to reply here.
Back
Top