Where is the headroom exactly in RAW?

[robbi@laurenson.com]

Advertisement (gone after registration)

If one wanted to do some very careful post-processing to a favourite image using a RAW converter (in my case C1 Pro) and then in CS4, which adjustments in the RAW converter actually take advantage of the 'headroom' available and which should be deferred to the greater capabilities of Photoshop?

 

My basic understanding is that RAW headroom comes into play with exposure and dynamic range, but in terms of color correction, sharpening and noise reduction etc. there are more capable tools in Photoshop.

 

 

To get the 'best' results (ignoring workflow complexity) what would the best procedure be? Here's my stab at it:

 

RAW Converter

• Lens Correction (D-Lux 4 corrected in C1)
  
• White Balance
  
• Exposure
  
• Export to 16 bit TIFF for PS
  

PhotoShop

• Chromatic aberration
  
• Sharpening
  
• Color correction
  
• Spot Removal
  
• Any other adjustments unavailable in C1
  
• Export to target format
  

What would you do?

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[dpattinson]

that's pretty much what I do, except I occasionally push the exposure comp a bit high and pull back the highlights using the C1 slider.

 

also make sure sharpening and noise reduction are switched off in C1.

[jaapv]

Add highlight recovery to the Raw converter work.

[robbi@laurenson.com]

Thanks very much for the pointers - I'll add highlight recovery to the list. Good to know I'm 'getting' it, though it is a can of worms I must say .

[stunsworth]

In Lighroom/CS4 one feature of the RAW conversion I find useful from time to time is the 'fill light' setting. This can be used to bring back shadow detail without affecting the overall image. I expect there's something similar in C1.

[jaapv]

In Lighroom/CS4 one feature of the RAW conversion I find useful from time to time is the 'fill light' setting. This can be used to bring back shadow detail without affecting the overall image. I expect there's something similar in C1.

 

Yes, it has a shadow, highlight and brightness slider which work together in a similar way. I find ACR better in this respect, though.

On the other hand, the elaborate shadow/highlight corrections, combined with levels and curves in the Luminance channel of CS 3/4 are even more versatile.

[Jamie Roberts]

Advertisement (gone after registration)

In C1, like Photoshop, the most powerful way of working exposure is the levels control and the curves control.

 

For instance, are you sick of speculars and highlights making a mess of your printing by going to paper white? Just dial back the highlight level input to around 245 RGB. You don't need the highlight slider in most cases. You also don't kill the contrast or mess with the midtones this way.

 

Same goes for shadow detail.

 

Want a nice gentle upper quarter-tone resolution? Put a pin in the middle of the curves dialog, and drag the upper quartertone down. Voila! Need "fill light" (which I personally hate on ACR and LR)? Just bump up the midtones in the curve and adjust the brightness slider.

 

Used together, you can get a lot of PS work out of the way in C1 without any highlight or shadow recovery sliders.

[robbi@laurenson.com]

Found an editorial on DPReview on RAW Headroom and I thought I'd share:

 

dpreview.com Editorial blog : Digital Photography Review

[jaapv]

In C1, like Photoshop, the most powerful way of working exposure is the levels control and the curves control.

 

For instance, are you sick of speculars and highlights making a mess of your printing by going to paper white? Just dial back the highlight level input to around 245 RGB. You don't need the highlight slider in most cases. You also don't kill the contrast or mess with the midtones this way.

 

Same goes for shadow detail.

 

Want a nice gentle upper quarter-tone resolution? Put a pin in the middle of the curves dialog, and drag the upper quartertone down. Voila! Need "fill light" (which I personally hate on ACR and LR)? Just bump up the midtones in the curve and adjust the brightness slider.

 

Used together, you can get a lot of PS work out of the way in C1 without any highlight or shadow recovery sliders.

I have a slightly different approach, Jamie. I prefer to do my contrast corrections without touching the colours. So I use the highlight ( and rarely shadow) slider to bring the maximum information into the file, export to CS4 as is and do my contrast tweaks in the luminance channel there. I would value your opinion on this.

[Jamie Roberts]

I have a slightly different approach, Jamie. I prefer to do my contrast corrections without touching the colours. So I use the highlight ( and rarely shadow) slider to bring the maximum information into the file, export to CS4 as is and do my contrast tweaks in the luminance channel there. I would value your opinion on this.

 

Except, Jaap, you can't do a contrast correction without affecting colour I know what you mean about LAB corrections, but ultimately you need to print as well. And RAW space has more malleability, I find, than even a high-bit TIFF in LAB.

 

Practically, what I usually do, actually, is balance the contrast with a curve in C1... which usually lessens contrast (and therefore saturation) in the RAW converter. Because I have the most amount of information in this stage of processing, here's where I will also tweak for skin tones.

 

I find I can even export JPEGs from there and work on them .

 

My workflow used to resemble yours: do as little in the RAW converter and work more in PS. I used to output camera-profiled TIFFs all the time--and high-bit TIFFs too--but the amount of re-work in PS is so much less if I get it right in camera and in C1 that I no longer need to do that anymore. I've proven to my satisfaction that there is no distinguishable difference in the resulting prints.

 

So in PS, adding back contrast is easy, and I too want "the maximum information" in my print workflow. But that usually means not messing with the highlight or shadow sliders, since I find they wreck midtones pretty quickly (I'd rather just clip the endpoints somewhere sensible). In PS I'm not messing with colour so much (unless I really have to do some brutal CC)--but rather dodging and burning and doing the touch up work a final print needs. I'm also squeezing that RAW information into the necessary DR for print, which, even using the best print technologies, ain't much

 

Of course, this is the general approach I take when working with many hundreds of files per wedding shoot. I'm also mainly concerned with faces and clothes, which is a completely different thing than mountain or canyon landscapes, or products, for that matter. So I still try to get basic skin tones and detail tonality (contrast) right in the RAW step, and leave PS for the grace notes.

 

If I was doing landscapes mostly, or some other kind of photography, I'd probably do the same things you're doing.

[jaapv]

Thanks, Jamie. I get you. I normally do a standard levels and curves, and then move into LAB for the final touch. The L channel, of course, allows contrast manipulation without affecting colours, and I find the A and B channel much to my taste for colour differentiation, tweaking and shifting colour balance, but I guess that is a personal thing. Then I move back to RGB, usually to desaturate a bit, as LAB manipulation often leaves the colour a bit vivid. I guess I can afford to do this as an amateur, not being under client time pressure. Your remark about the midtones and the shadow/highlight slider is noted; thanks (As is the Canyon )