"Fake" HDR question...

I was just wondering if I make 3 copies of the same RAW file, can I adjust the exposures in PP and then merge to HDR? Or does it have to be shot in the various exposures?

You can definitely do it that way. You don't even need to make 3 copies, just develop and output jpegs from the same file multiple times.

Ok thanks, I'll try it out.

No, you can use one exposure and rely solely on the headroom provided by the single RAW. Many people do not regard this as true HDR, but rather 'pseudo-HDR'. Personally, I find pseudo to work well under certain circumstances. Infact, the following came from a single JPEG (not even RAW), so you can definitely achieve a typical HDR-looking image from just one file.

Byte size: ZERO | Content warning: NOT AN IMAGE

Byte size: ZERO | Content warning: NOT AN IMAGE

^ Those are great, would you mind sharing some of your technique in using just one JPG? I am okay in CS3 but I would not be able to pull off those shots, they are wonderful!

Thanks TeamSpeed! Each shot is worked a slightly different way, but it's a combination of a set of curves adjustment layers (with each selectively masked over different areas of the shot), while also playing with the Shadows/Highlights tool and the like. The shots are actually shot in RAW, where I then pick an 'optimum exposure' to convert to JPEG (or TIFF if you prefer) before doing this workflow in CS4. I always use DPP rather than ACR as I get punchier results. The most important thing is setting up for the workflow in-camera. The histogram is key, and things like a CPL can help alot (it certainly did for the second shot).

Sometimes if the DR is too much, I do a two-shot bracketed handheld sequence (this is where a high FPS body can actually come in handy for landscapes!) and then manually merge in CS4, mainly using masks. Here's an example:

Byte size: ZERO | Content warning: NOT AN IMAGE

Awesome examples Collin

Awesome, thanks collin. The one thing I've noticed from trying from one RAW is massive clipping in white areas and blow outs in some other areas. Is there any way to prevent that or is it just the shot? What histogram shapes should I look for / stay away from? Thanks.

Doing this from one Raw file is the same in principle as using multiple exposures -- multiple exposures will have, say, one "middle" exposure where maybe highlights are too bright and shadows are too dark, then (for a three shot sequence) you would have an "underexposed" shot where the highlights have desired detail but the shadows are much too dark and an "overexposed" shot where the shadows have good detail but the highlights are blown. You put each in a separate layer in Photoshop then use masks and brush the "good" parts into one image.

Using one Raw file you can use your Camera Raw adjustments to take advantage of the range of the pre-processed Raw file to get results that aren't quite as much but close to what the three separate exposures give you: if your original exposure is "medium", you can load that as one layer, then you reload the Raw into Camera Raw and lower your exposure by say a stop and load that into another layer, then again run the original Raw file into ACR and raise the exposure by a stop.

The resulting three layers can be merged (automatically align pixels) if need be, then you add layer masks so that only the background is visible then paint in the highlight details from the lower exposure and the shadow details from the higher exposure.

Like I said, the results won't be quite as good as three separate exposures but can still work quite well.

An alternative is to use the local adjustments in Camera Raw if you are using CS4: gradients and brushes can be applied to lighten the shadows and darken the highlights as needed and all is contained in the data for the single Raw file. No photoshop needed if you use a good touch with the tools.