16 bit post-processing (RAW)

[marc_schmidtmayer]

Hi, I've read several articles/topics regarding RAW-files and their advantages.

But still I've got some questions.

 

So, I'm well aware of the advantages like being able to set/change white

balance, sharpening, saturation and exposure in the RAW-convertor. But let us

forget these listed advantages and just discuss following 'topics' : Bit Depth

and Dynamic Range.

 

I've read that the higher Bit Depth decreases the susceptibility to

posterization (which is important to me, as I work in B&W with relatively much

post-processing). Are there other advantages in working in 16bit vs 8bit ?

 

Also, RAW files tend to have more "dynamic range" than a JPEG file. But is this

(also) linked to the (possible) higher Bit depth or just with the lossless

compression (or something else) ?

 

In Camera Raw, one has the possibility to open (convert) the picture in 8 bit

or in 16 bit. When opening the picture in PS in 8 bit or 16 bit ... what's

the 'difference' ?

If I convert the RAW file and open it in PS in 8 bit ... do I loose all the

advantages regarding the (higher) "dynamic range" and (less) posterization ?

 

Is there an advantage (left) when converting the RAW file to 8bit ?

 

Does it mean that I must do all the post processing in 16 bit ? And if so, is

there a moment I need to convert it to 8 bit ?

 

All help is appreciated ! Thanks, Marc.

[orensztajn]

You can do a test, take a picture and conv to 16 and then to 8.

After this you can examine the histogram.

 

The 16 should have more info.

 

I work very much in 16 and I can see the diff against the 8 bit ones (that I used before).

[neil_grant]

An image with a greater bit depth uses more binary digits to describe the level of each pixel.

When you perform density transformations on an image these are recalculated- and rounded

up or down if necessary. This 'rounding' is a form of error and loses tonal descrimination.

A 16 bit file is less effected by this than an 8 bit one. To answer your last question, perform

your image adjustments on the highest bit file you have, then convert to 8 bit when you've

finished.

[luminous world]

Typically, conversion to 8 bit occurs during the part of the workflow where you prepare the file for final output. I save the post processed PS file as a master and then duplicate it before flattening, converting to 8bit for output to inkjet or lightjet, resizing to target size, sharpening and then adding a levels layer for adjusting output levels depending on the printer/lab being used. I then save as a TIFF for use again later.

[gdanmitchell]

A few responses:

 

Much of your understanding is correct.

 

I'm not sure of the reason for the greater dynamic range of RAW files, but it definitely is

there. Perhaps this is mainly an issue when you "push" the boundaries of exposure,

especially when trying to recover details from shadows. You potentially have more levels

to work with when you expand that region in particular.

 

I always leave my files in 16-bit form until the printing stage. While this might not be

terribly necessary on photos that you adjust little, you'll have better results when making

significant changes to levels, colors, etc. if you have 16-bits to work with. As you mention,

this can reduce effects like posterization.

 

"Post-processing" in 16-bits is, as I understand it, what I've been referring to above: "post

processing" the image from the camera in PS as a 16-bit image rather than converting it to

8-bits (or, worse, jpg) and then making changes.

[dave_powell2]

16-bit is also an advantage when upscaling an image in size.

 

Dave

[marc_schmidtmayer]

16 bit it will be ! ;-)

 

Thanks very much for your answers,

Marc.