underexposure compensation

[Jonathan Wienke]

If you're going to do noise reduction, do it immediately following RAW conversion. I don't do anything special if I think I might do double noise reduction; my goal is to always kill as much noise and as little detail as possible. I usually make a new noise profile for each image; an accurate noise profile seems to be key to effective noise reduction. After Neat Image, I run Focus magic and my sharpening action, do any color correction if necessary (usually not, due to having Camera RAW calibrated to my cameras), curves to adjust tonality, etc. If sharpening brings out too much noise, I'll run a second Neat Image pass making a new noise profile (the sharpening and stuff changes the noise characteristics) and then call it good. I may fade the last Neat Image pass to dial it back if it gets too aggressive.

[Jonathan Wienke]

I bought Neat Image before Noise Ninja came out, and after comparing them concluded that while both are very good at what they do, the differences between them probably don't justify owning both. If I owned neither, I might get Noise Ninja, but Noise Ninja doesn't offer enough advantage over Neat Image to justify buying it IMO. Both programs are capable of excellent results.

[DarkPenguin]

Exposure compensation can work.  The shadow highlight tool can work.

The best solution is to pay someone to beat you with a stick everytime you click the shutter and do not check the histogram.

[Jonathan Wienke]

Basically windows is 2.2 gamma. 1.0 gamma and 1.4 gamma and 1.8 gamma can be used depending on the amount of underexposure It it is worth the bother tracking down the information to learn this technique

Don't bother. You can change the gamma by moving the upper-middle slider in the Levels dialog, and tweaking gamma will not make the highlights white. The Exposure adjustment in the RAW converter is a far better option.

[digitaldog]

False profiles in PS will rescue an underexposed image Dan Margulis in Professional PS goes into the topic in detail On the disc that comes with it he provides the profiles You can create them yourself but it is fiddly Basically windows is 2.2 gamma. 1.0 gamma and 1.4 gamma and 1.8 gamma can be used depending on the amount of underexposure It it is worth the bother tracking down the information to learn this technique

Not really. When you convert from this profile, the degradation will at that time be introduced. The benefit of using a profile to change the color appearance is speed. It’s a heck of a lot faster to Assign a profile to correct the color appearance than use all kinds of edits. But there’s no free lunch. While the image is in the current color space with this profile, you’ve not changed the numbers but when you convert, that’s when the time comes to pay the piper.

[bobrobert]

Not really. When you convert from this profile, the degradation will at that time be introduced. The benefit of using a profile to change the color appearance is speed. It’s a heck of a lot faster to Assign a profile to correct the color appearance than use all kinds of edits. But there’s no free lunch. While the image is in the current color space with this profile, you’ve not changed the numbers but when you convert, that’s when the time comes to pay the piper.

You obviously haven't read Dan's book When you use the false profiles you are assigning the profile It is accessed from Mode > Assign profile Michael Kieran in PS Color Correction covers the same ground It is as I previously stated worth learning

[bobrobert]

Are you saying what Dan Margulis says on False profiles is wrong and should be diregarded?

[digitaldog]

Are you saying what Dan Margulis says on False profiles is wrong and should be diregarded?

No, did you read this in any post?

I don’t agree with the term false profile. It’s a matter of semantics but the profile isn’t false. You could called it right, preferred or a tweaked profile.

I don’t agree that the use is lossless which I don’t think Dan claims. When the profile is used for conversion to any color space, the data loss will now be produced. Do it on high bit to minimize this.

I don’t believe Dan’s opinion that editing in high bit is of no use and in fact if you subscribe to his color theory list, the last few days discussion have been very interesting. I’ve uploaded a file and tests that appear to confirm that there IS a quality advantage to editing in high bit on a real world image.

[DarkPenguin]

Let he who has never committed this sin strike the first blow.  Be honest, now.  

I did it just after my post on the subject.  (Two shots, tho.)

BTW, did EC or Shadow/Highlight or whatever the other suggestion was help?  Rescue any shots?

(Never mind.  I replied while reading page 2.  Should have read page 3 first.)

[Mark D Segal]

Jonathan, I've been using Neat Image for dealing with film grain (processing my legacy negatives) and do exactly what you do - let it profile each image, tweak if necessary to get the right balance between grain reduction and preservation of image detail, then sharpen immediately afterward, except I use PK Sharpener Pro. For my digital work I'm using Noise Ninja instead of Neat Image. Have you compared them for digital noise?