Question about clipping indicators in ACR/false recovery of clipped values

[AFairley]

I keep meaning to test this for the last week, but I only think about it when I'm not around ACR, so I'm just gonna ask even though it's lame to do so when I really could just find out for myself. 

Say I have a RAW file that has some highs that are indisputably, unrecoverably, clipped.  Is it the case that no matter how far I pull down the overall exposure in ACR that these highs will show as clipped with the colored indicator and stay at 255-255-255; or at some point will the adjustment pull the values down below 255-255-255 and introduce false grey/color into them?  Will the answer be different depending on the adjustment used to pull the exposure down (exposure/brightness/recovery, e.g.)?

Thanks

[AFairley]

I did remember when I was near ACR, and some quick experiments showed me that it is possible to pull a fully clipped area of an image into grey, but it takes some pretty extreme adjustments to do it.  I had a textured white wall next to a dark case; if the image was not too badly overexposed, I could pull the blank area of the wall down so you could see the texture in it; if the wall was too overexposed, I could pull the image down enough to introduce some grey around the edges of the blank area, but there was not texture in the grey (only distinct banding).

The bottom line for me is that the adjustments required to produce the effect are so extreme that you really could not apply them to an image that was anywhere near normal, so there is no practical consequence for me. 

The thing I really took away from my little experiment, though was just how much detail there can be in areas that appear to be blown out (of course, if it's really blown, nada).  I'm reminded of Schewe's example of the falls in his ETTR article.

[madmanchan]

Raw image values that are clipped to pure white will remain white in ACR as you reduce Exposure and/or Brightness. This is the intended behavior of the controls.

(You can actually reduce the white point, effectively making those clipped whites turn gray, by adjusting the rightmost point of the point curve. Usually undesirable, though.)

[digitaldog]

Raw image values that are clipped to pure white will remain white in ACR as you reduce Exposure and/or Brightness. This is the intended behavior of the controls.
(You can actually reduce the white point, effectively making those clipped whites turn gray, by adjusting the rightmost point of the point curve. Usually undesirable, though.)

I saw an interesting feature request on one of the Adobe forums whereby no matter the initial settings of those sliders, IF indeed the raw image values were clipped, you'd get the option of a new kind of new color clipping overlay. This would tell you, don't mess with the sliders, you blew your highlights.

[AFairley]

I saw an interesting feature request on one of the Adobe forums whereby no matter the initial settings of those sliders, IF indeed the raw image values were clipped, you'd get the option of a new kind of new color clipping overlay. This would tell you, don't mess with the sliders, you blew your highlights.

That feature would be cool.  Eric, thanks for the response.

Bearing in mind what you said, I fiddled some more, and, lo and behold the exposure and brightness sliders do not create false density in blown areas.  The false density effect I saw was the result of using the recovery slider in addition to those adjustments.  So from this, I gather that I need to be careful when I'm pushing things with the recovery slider.

I'm attaching some shots (shadow and textured wall in sunlight) I did with the S90

1. Apparently overexposed file:  Bright area show red at ACR default, but shows texture when developed with Exposure -4 and Brightness -10

2.  Badly overexposed file:  Bright area shows red at ACR default, stays red with Exposure and Brightness sliders all the way to left

3.  Same badly overexposed file:  With Exposure and Brightness all the way to left, Recovery all the way to right, false density is introduced at edges.

4.  100% crop of #3 showing banding, just for kicks.

[Peter_DL]

2.  Badly overexposed file:  Bright area shows red at ACR default, stays red with Exposure and Brightness sliders all the way to left

3.  Same badly overexposed file:  With Exposure and Brightness all the way to left, Recovery all the way to right, false density is introduced at edges.

Interesting,

I was not really sure if Recovery could do anything special,
which couldn't be done the same way with -Exposure setting in ACR + other tonal controls.

Unlike with Fill Light for example,
where it seems (to me) obvious that some sort of pixel-selection-based "exposure blending" happens behind the scenes.


Peter
--

[Orlando Photographer]

I could stand to be corrected, but in my experience if their are nearby values and you push the recover slider all the way, you can begin to recover some of the information from neighboring pixels to get to or at least near a usable image.

Sometimes I have played pushed recovery all the way, raised fill and lowered exposure to get a salvageable black and white image in some more extreme situations. How salvageable a given image is, is certainly a subjective thing, but barring that, if an image is overly blown out, no adjustment will bring it up to a level that I find acceptable. Not that I have a lot of blown out images at this point, but there have been a couple occasions where I was glad to have additional control.

I tend to work in Lightroom and only revert to ACR via PS if I know I might need more control with the use of a smart object, but I can't imagine there is any difference between the results I get with LR when compared with the results from ACR, since LR is basically ACR with cataloging.

[maria85]

thnx.. ws looking fr ths..