HDR creation, two approaches?

[Shiva_42]

Forum members, I'm an old-school film guy just getting into digital, but have a seasoned view of photography in general. Have a specific question about an HDR image:

I'm shooting RAW with my Nikon D90, and am curious about the following two approaches to an HDR image...

1) Shoot multiple exposures at the full range of dynamic ranges, then merge them together.

2) Shoot ONE image of the scene, then develop that image for all exposure ranges with my RAW software, then merge them together.

Assuming that the full HDR of the expected image is POSSIBLE just with the developing options from the single RAW exposure, which of the two approaches is expected to give the best results? For the sake of this discussion, let's assume that I don't care about the cost of any software involved, nor the time/complexity/trouble that's required to accomplish the image, I'm only concerned about the ultimate final image quality and look. TIA!

 

[ann]

If one image covers the contrast range it is NOT HDR. you can take that file and tone map it to get a similar look , or you can take that file and make three exposures, but why?

It is my opinon that one needs as many images as it takes to cover the dynamic range, and the results will depend on how it is processed after the merge.

There is a wide variety of software available these days, with Nik providing a new product that handles moving objects very well. I must have at least four different pieces of software, each with their own strengths and weakness.

Photomatrix keeps improving their product and provide free updates and those updates are often so they aren't resting on past history.

It is also MHO that the workflow involves more than a few clicks of a mouse to produce the "ultimate quaility and look". Takes practice and work.

 

[Shiva_42]

Thanks Ann for the feedback. That is why I was careful to outline that the ONE image would be developed (during the RAW conversion and processing) to cover the needed exposure range(s). I'm still curious about expectations given that premise.

 

[ann]

You can make it look like what ever you wish, but it is not consider HDR by those who take this style very seriously, it is just a single exposure that has been tonedmapped and manipulated. Which doesn't mean it can't be terrific, but it is not HDR , at least imho.

If your more serious about doing "true" HDR, then that implies a longer dynamic range than 1 image holds.

Does that answer the question, or do you mean will one exposure give you
acceptable results.?
It also all comes down to what your vision is and how you want to express that to others. Use what the range calls for, but don't be surprised to get some "flak" when using one exposure and calling it HDR.

 

[Bynx]

Ann is quite right. One image does not cover the whole range. Even a raw file. Every shooting situation is different so you have to shoot for the conditions youre in. Very bright light source to very dark shadow areas will need a lot more than 3 shots to capture the whole range. Shoot as many as you need so there are no blown out highlights or filled in shadows.

 

[Shiva_42]

So... if I'm interpreting this stuff correctly... the expectations (and probably the literal reality) is that this will REQUIRE separate exposures to create the stunning effects that we see in the "best" hdr images found.

If I wished to follow the process I describe to "fix" a problem exposure, the manipulation suggested during RAW processing will help, but would never be an acceptable substitute for multiple exposures when you KNOW or EXPECT ahead of time that a real HDR image is your goal.

That's basically the response I was expecting for this question, but the overall dynamic range of these DSLR's is so superior to my film experience, I was just trying to get grounded on the subject before I started attempting the images.

Appreciate the responses! Will monitor this thread for any other advice, and look forward to being a routine contributor here in the forums!

Thanks again folks!

 

[Bynx]

5 minutes doing it is better than an hour talking about it. Nothing to lose.

 

[ann]

study the light, make as many exposures as necesary to cover the range; think shoulder to shoulder on the histogram. You can never have too many exposures.

One method is to take a spot metering reading from the highest highlight and a spot reading from the darkest shadows with detail you wish and "run" the exposures between.

Use a tripod and mirror lock up if at all possible as the images will be very sharp with that addition to the process.

my suggestion would be too

. with the RAW file, check with CA issues and spoting, apply to all images at the same time with batch correction,

. convert to tiff files,

. merge with the program of your choice.

. tonemapp and make all other corrections within that program and then with what ever other program your using.

others workflow may vary , just "play around a bit" to find out what works for you, and as Byinx indicates doing is much better and a faster way to learn.

 

[demandread]

If you're doing a normal photoshoot and discover that you have some lost shadow or some blowout in a part of your RAW file (after the fact), then tweaking the RAW file can help you save some of it, and is a fairly common approach (but don't call it HDR ).

The biggest difference in results between single exposure and true HDR is the shading of saturated areas, which cannot be properly extracted from the RAW file even after playing with the exposure/balance settings. If you ramp up your tonemapping on the file, it'll appear distinctly "flat" compared to a multi-exposure image.

Try it! Take a triple-exposure set on your DSLR and see what you can do with it, and then try to do the same things with (any) one of the exposures alone. That's the quickest way to learn what you can and cannot do.