HDR question

I'm new to doing HDR processing and have a question...
If I'm merging two photos, is it better to make adjustments (histogram, contrast, etc, etc) to each photo individually prior to merging them, or is it better to merge them "as-is" and then make adjustments to the merged photo?

Good question, Bob. I've done minor adjustments beforehand, like making sure white balance is the same, then doing most other adjustments post-merge. But why not try it both ways to see what you get? (And then let us know! )

Bob,

By no means an expert so FWIW

- If you are blending images by layering them and using masks to paint in the parts you want I think you can do the adjustments beforehand without causing too much trouble.

- If you are merging them, with something like photomatrix, etc. I'd play it like Todd, try to match white balance but I'd wait to do everything else until after the merge. You might make it harder for the program to decide which image to use for highlights/midtones/shadows.

I pretty much follow Aleksei's process.

Thanks guys!

I don't have photomatrix, but for my manual blends I get them as close as possible first and then blend. Seems to work better for me.

I don't do HDR with Photomatix much any more (tonemapping always seemed to leave me with a less than natural looking output and it was too much work to make things look natural again) and I don't think that it takes into account the ACR sidecar files anyway, but I do use what's now called "Exposure Fusion". I sharpen and/or reduce noise, synchronize all images then export as TIFFs which I then process in Photomatix.

Hi Bob. I've done a little bit of experimenting both ways and have found that I get better results in Photomatix with RAW files instead of slightly adjusted TIFF files. I've never quite sure understood why since logic would dictate, at least to me, that the merging of better optimized files would produce a superior result to merged RAW files, but in my experience this hasn't been the case. I haven't exhaustively studies the issue though, and I'd be curious to hear of your experience if you do some comparisons between methods.

I'll go along with the crowd here - if you're using an HDR tool of any sort to do the blending, you're best off using files as close to the originals as possible, and editing afterward. If your HDR tool does image alignment and can suck in RAW files, use RAW files. If not, match the white balance and align the different exposures without making any exposure adjustments and use the resulting files in the HDR.

HDR tools assume the files all come from the same device in the same basic conditions; altering the files outside of the absolute basics may degrade your HDR tool's ability to comprehend the scene.

Only if you're doing hand-blending is it a good idea to do pre-blending adjustments beyond the very basics IMHO.

You can do an HDR blend with only one photo if the highlights and shadows are less than 3 to 4 stops apart. Open the photo in ACR and adjust the luminance/brightness for the highlites. You can usually decrease 1 to 2 stops of exposure. Then open it in Photoshop, go to Layer/Smart Objects/New Smart Object via Copy and it will open the RAW file as a new Smart Object layer sitting above the original. You can then open it in ACR and adjust for shadows. Voila, two different exposures perfectly aligned. How much you can tweak the highlites and shadows varies of course depending on the shot, but if you are shooting RAW you can almost always separate the two layers by two stops of exposure, often three, and sometimes four.