16-bit linear conversion

I have Photoshop CS and a Nikon 8800 camera that I am shooting raw images with. I have found adobe's camera raw to not quite do what I would like it to, and wanted to investigate using dcraw as a possibility. dcraw outputs 16 bit linear psd files. Are there any turorials for how to adjust 16 bit linear files to something that I can get a decent printout of? Googling has yielded nothing but flamewars about linear gamma and 16 vs 8 bit, and a couple of relevant things that were Canon specific.

thanks,
Andrew

I am curious about 16 bit files too, and what is the advantage to using them.
My scanner creates images in 16 bit but I can never do much with them in PS unless I change them to 8-bit.
So am I wasting my time scanning in 16? only to throw it out later? How does bit size affect the resolution, if in deed at all?
Will CS2 have more filters/processes available to 16 bit images?

"Andrew Crabtree" wrote in message

I have Photoshop CS and a Nikon 8800 camera that I am shooting raw images with. I have found adobe's camera raw to not quite do what I would like it to, and wanted to investigate using dcraw as a possibility. dcraw outputs 16 bit linear psd files. Are there any turorials for how to adjust 16 bit linear files to something that I can get a decent printout of? Googling has yielded nothing but flamewars about linear gamma and 16 vs 8 bit, and a couple of relevant things that were Canon specific.

thanks,
Andrew

On Tue, 3 May 2005 15:52:41 -0400, "KatWoman"
wrote:

I am curious about 16 bit files too, and what is the advantage to using them.
My scanner creates images in 16 bit but I can never do much with them in PS unless I change them to 8-bit.
So am I wasting my time scanning in 16? only to throw it out later? How does bit size affect the resolution, if in deed at all?
Will CS2 have more filters/processes available to 16 bit images?

The benefit of using CD files is when adjusting the image (using layers and so forth) you have 65356 values to work with instead of 256. If you look at the histogram of an image before and after you've worked on it, large changes will often show gaps in the histogram with only 256 to play with. With 16 bit, even if you make alterations large enough to make gaps in the histogram, when you have achieved your final image and saved it as 8 bit the missing areas will have little effect on your image because you're using a larger subset set of image values than with 8 bit.

--

Hecate - The Real One

Fashion: Buying things you don't need, with money
you don't have, to impress people you don't like...

In article <lCQde.162189$>,
says...

I am curious about 16 bit files too, and what is the advantage to using them.
My scanner creates images in 16 bit but I can never do much with them in PS unless I change them to 8-bit.
So am I wasting my time scanning in 16? only to throw it out later? How does bit size affect the resolution, if in deed at all?
Will CS2 have more filters/processes available to 16 bit images?
"Andrew Crabtree" wrote in message

I have Photoshop CS and a Nikon 8800 camera that I am shooting raw images with. I have found adobe's camera raw to not quite do what I would like it to, and wanted to investigate using dcraw as a possibility. dcraw outputs 16 bit linear psd files. Are there any turorials for how to adjust 16 bit linear files to something that I can get a decent printout of? Googling has yielded nothing but flamewars about linear gamma and 16 vs 8 bit, and a couple of relevant things that were Canon specific.

thanks,
Andrew

You may find some of the discussions about 8 vs 16 bits on my web site useful. Follow the tips link on the home page.
16 bits is useful in the early stages of tone and color adjustment if large-scale adjustments are needed. Having more steps prevents posterization which produces a blocky look.
Many people can't see the difference as my example shows. Once you have made the large-scale adjustments you can convert to 8 bit for printing and display as this is all output devices support. Most people find an "S" shaped curve the best way to transform from a linear space. You could experiment by shooting a picture of a gray step wedge (or similar scene) and then using a curves layer to bring the image to a visually similar appearance to the original.

--
Robert D Feinman
Landscapes, Cityscapes and Panoramic Photographs
http://robertdfeinman.com
mail:

"Andrew Crabtree" wrote in news:elOde.4866
$:

I have found adobe's camera raw to not quite do what I would like it to

What is missing?

"Eric Gill" wrote in message

"Andrew Crabtree" wrote in news:elOde.4866
$:

I have found adobe's camera raw to not quite do what I would like it

What is missing?

Early reports suggest it may be improved in CS2, but I like the ability to just say "convert *.nef" and come back an hour later (I realize I can probably automate a lot of this if I want to). I also wanted to compare image quality between dcraw and adobes just to see if it was noticable.

Andrew

so scan at 16 bit, use levels, and then switch to 8 bit? I bought a new slide scanner just before we went fully digital and have hardly used it.
I have loads of old slides I would love to scan but not sure if I will ever get back to it.

"Robert Feinman" wrote in message

In article <lCQde.162189$>,
says...

I am curious about 16 bit files too, and what is the advantage to using them.
My scanner creates images in 16 bit but I can never do much with them in PS
unless I change them to 8-bit.
So am I wasting my time scanning in 16? only to throw it out later? How does
bit size affect the resolution, if in deed at all?
Will CS2 have more filters/processes available to 16 bit images?
"Andrew Crabtree" wrote in message

I have Photoshop CS and a Nikon 8800 camera that I am shooting raw images
with. I have found adobe's camera raw to not quite do what I would like it
to, and wanted to investigate using dcraw as a possibility. dcraw outputs
16 bit linear psd files. Are there any turorials for how to adjust 16 bit
linear files to something that I can get a decent printout of? Googling has yielded nothing but flamewars about linear gamma and 16 vs 8 bit, and a
couple of relevant things that were Canon specific.

thanks,
Andrew

You may find some of the discussions about 8 vs 16 bits on my web site useful. Follow the tips link on the home page.
16 bits is useful in the early stages of tone and color adjustment if large-scale adjustments are needed. Having more steps prevents posterization which produces a blocky look.
Many people can't see the difference as my example shows. Once you have made the large-scale adjustments you can convert to 8 bit for printing and display as this is all output devices support. Most people find an "S" shaped curve the best way to transform from a linear space. You could experiment by shooting a picture of a gray step wedge (or similar scene) and then using a curves layer to bring the image to a visually similar appearance to the original.

--
Robert D Feinman
Landscapes, Cityscapes and Panoramic Photographs
http://robertdfeinman.com
mail:

You may find some of the discussions about 8 vs 16 bits on my web site useful. Follow the tips link on the home page.
--
Robert D Feinman
Landscapes, Cityscapes and Panoramic Photographs
http://robertdfeinman.com
mail:

Impressive web site Robert. Thanks.

Brian

In article <4%7ee.4917$>, Andrew Crabtree
wrote:

"Eric Gill" wrote in message

"Andrew Crabtree" wrote in news:elOde.4866
$:

I have found adobe's camera raw to not quite do what I would like it

What is missing?

Early reports suggest it may be improved in CS2, but I like the ability to just say "convert *.nef" and come back an hour later (I realize I can probably automate a lot of this if I want to). I also wanted to compare image quality between dcraw and adobes just to see if it was noticable.

You could do that (batch convert) with ACR 2.0 in CS1.
CS2 makes it even easier to batch process RAW files.

Chris