bit depth question

rbiemer

rbiemer

Unabashed Amateur
Local time
4:15 PM
Joined
Sep 17, 2004
Messages
5,092
Location
Cortland, NY
View My Gallery
I want to enter a photo contest and am having trouble with the submission guidelines.
Specifically this:
...images should be 8 bit (NOT 16 bit) saved on CD as tiff, jpeg or psd files and must include a color or black-and-white printout...
Click to expand...
8 bit? I use several programs depending on how involved any editing might be: Picasa, Sagelight, the GIMP, and Fastone. I don't find any way to get down to 8 bit in these programs. 24 seems to be the default--which is not 16 so I guess I could just save the files to disc at that setting and submit but I don't want to be rejected 'cause I didn't follow the rules.
So, buy Photoshop? And if so, which version? Is there some, hopefully free/cheap, alternative way to get what I need for this?
Appreciate any help here, folks!
Rob
 
Hmm. I have never really understood this. I was thinking the OP was referring to the two options in PS. 8 and 16-bit. I thought 8-bit was the default (e.g. an out-of-camera jpg comes as an 8-bit file in whatever color space for which the camera is set). Hard to believe that the GIMP can't convert one to the other . . .
 
hteasley said:
They're referring to per-channel bit depth. They want 24-bit images.
Click to expand...
OK, that makes sense. I probably will email and ask them to be sure.

Hmm. I have never really understood this. I was thinking the OP was referring to the two options in PS. 8 and 16-bit. I thought 8-bit was the default (e.g. an out-of-camera jpg comes as an 8-bit file in whatever color space for which the camera is set). Hard to believe that the GIMP can't convert one to the other . . .
Click to expand...
Frankly, I've never given this much thought; and it is a bit confusing to me too!
Rob
 
8 bits in binary can hold decimal values of 0 thru 255. When talking about 8bit they are referring to colour channels being 8bit. There are three colour channels in an RGB file. 3 times 8 = 24. So correctly speaking, standard RGB files are 24bit files with 3 8bit colour channel.
16 bits in binary can hold decimal values of 0 thru 32768. When talking about 16bit they are referring to colour channels being 16bit. There are three colour channels in an RGB image. 3 times 16 = 48. So correctly speaking, 16bit RGB files are 48bit files with 3 16bit colour channels.

Big difference in the number of shades between an 24 bit and 48 bit file. But most printing devices can't (or just don't) print 48bit so that is why they ask for 24bit files with 8bit colour channels.

Default for most software is 24bit files.
 
The JPEG statndard also allows for 12-bit-per-color images, but rarely encountered.

Most JPEG implementions are 8-bit per color. Maybe the contest is covering itself.
 
Benjamin Marks said:
Hmm. I have never really understood this. I was thinking the OP was referring to the two options in PS. 8 and 16-bit. I thought 8-bit was the default (e.g. an out-of-camera jpg comes as an 8-bit file in whatever color space for which the camera is set). Hard to believe that the GIMP can't convert one to the other . . .
Click to expand...

The source of confusion here, I think, is that there are 8-bit files out there (think GIF), which have only 256 colors available (or 255 and an alpha register). Those are inappropriate for photographic purposes. They asked for 8-bit/channel files in a somewhat sloppy manner. PS refers to 8-bit files as "Indexed color": the 256 colors available in 8-bit files are a lookup index into 24-bit colorspace (really old VGA spec was a lookup into 18-bit colorspace, but that's long ago, in a galaxy far, far away).

The emoticons you can add to your posts are 8-bit files.
 
You must log in or register to reply here.
Back
Top Bottom