TIFFs: 8, 16 or 32-bit?

[Robin Springall]

What are the advantages or disadvantages of saving images as 16- or 32-bit TIFFs please? I've always just used flattened 8-bit ones in the past.

Thanks

[jgr]

What are the advantages or disadvantages of saving images as 16- or 32-bit TIFFs please? I've always just used flattened 8-bit ones in the past.
Advantage - Color depth:

1 bit - Black and white only
8 bit - 256 shades of grayscale, or 256 colors
16 bit - High Color, 65,536 colors
24 bit - True Color, 16,777,216 colors
32 bit - True Color, 4,294,967,296 colors
Disadvantage - file size:
File sizes increase dramatically as you increase the color depth.

[Stephen Owades]

What are the advantages or disadvantages of saving images as 16- or 32-bit TIFFs please? I've always just used flattened 8-bit ones in the past.

Thanks
You should be sure of the terminology being used, by the way. Most programs work in 8-bit-per-channel color, so a greyscale image will be 8 bits per pixel, an RGB color image will be 24 bits per pixel, and a CMYK color image will be 32 bits per pixel. But those might all be described as 8-bit by some software.

It is possible to work with more than 8 bits per channel these days in some programs, particularly if the source is a RAW camera image. This can be helpful especially if you're going to manipulate the tonality of the image, since previously smooth gradations can suffer from banding if you change the curves enough. But 8 bits per color is usually enough in the final image.

[Robin Springall]

OK chaps, thanks to you both!