16 Bit Tiff Advantage

[dman65]

I hate to ask a stupid question, but I am questioning what converting RAW to 16bit tiffs achieves exactly.

I just converted an existing 8 bit tiff I have to 16 in PS and put my cursor over a white area and it showed red, green, and blue to all be 255. Shouldn't they all show 65535 if it is 16bit? (2^16=65536) Is the extra data actually being utilitized? Does it result in greater detail in the shadow area? And then, I hear that people are converting them to 8 bit when they finish manipulating them so that extra data is lost at that point. What exactly is the advantage in the 16 bit conversion? Is the software being used to convert the 16bit to 8 bit in Adobe a lot better than what is being used in the Canon software to convert the 12 bit data to 8bit?

I converted one raw image to both 8 and 16 bit tiffs and I did not see a lot of difference. Of course, my video adapter only has 24 and 32 bit modes. The 24 bit would coincide with the 8 bit mode. The 32 bit mode should be able to show a little more of the combined 48bits of the 16 bit tiff, but I don't know if my actual monitor is capable of getting more than 24 bit performance and I also don't know if my printers can handle the 48bits (Olympus P400 and Epson 1280).

I am just curious as to what the benefit is and whether or not I need to start using 16bit.

[Roger_Cavanagh]

This Bruce Fraser article explains the benfits of 16-bit with some pretty pictures:

http://www.creativepro.com/story/feature/7627.html

Also bear in mind that some of the constraints that Bruce mentions in what PS can do with 16-bit images don't apply any more. For instances, he says you can't sharpen or use any filters, which is not the case in PS6 or PS7.

The D30 produces 12-bit (per channel), so strictly speaking there is some wasted space. However, you are definitely throwing away data by converting to 8-bit first. There's no point to convert from 8-bit to 16-bit since PS cannot make up the extra data.

However, for some reason PS only allows us 256 values (0-255) for each colour channel whether we are working in 8/16 bit. Quoting Bruce:

... we can see at the start that the 48-bit image and the histograms look identical to the 24-bit one. But there's an important difference: In the 24-bit file, each channel contains 256 possible levels, from 0 to 255; in the 48-bit file, each channel contains 65,536 possible levels. Our monitors don't display them, and our histogram doesn't show them, but they're there, and they make a big difference.

It is hard to see any difference, but Bruce Fraser's article proves that it does have some benefit. I also found that it is better to work in 16-bit when you are upsizing an image to any great degree. At some point, whatever you do, artefacts are introduced because, after all, PS is just making up pixels, but these artefacts show up much, much sooner with 8-bit than 16-bit files.

For myself, I stay in 16-bit as long as possible. Data is (or for the pedants, are :D ) data, and it doesn't seem right just to throw it away.

Regards,

[dman65]

Roger,

Thanks for the link. I will check it out.

Darrell