I found this brief tutorial to be interesting, summarizing some issues related to trying to print 16-bit photographic images.
http://www.earthboundlight.com/phototip ... nting.html
----------------------------------
I wonder if digital photography will some day switch from integer arithmetic to 32-bit or 64-bit floating-point data, eliminating concerns about quantization errors. Some parts of Photoshop CS5 can now handle 32-bit pixel data, presumably allowing things like HDR processing to work with fewer artifacts.
tutorial for 16-bit printing
Moderators: rjlittlefield, ChrisR, Chris S., Pau
tutorial for 16-bit printing
-Phil
"Diffraction never sleeps"
"Diffraction never sleeps"
- rjlittlefield
- Site Admin
- Posts: 23561
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
Re: tutorial for 16-bit printing
Zerene Stacker uses exclusively 32-bit floating point for its internal representations. This gets reduced to 8 or 16 bits only for output to standard image files or screen. The representation increases memory requirements, but in addition to the advantages you mentioned, floating point also turned out to run faster when I tested that during initial prototyping. Modern computer systems have some behaviors that aren't always intuitive!DQE wrote:I wonder if digital photography will some day switch from integer arithmetic to 32-bit or 64-bit floating-point data, eliminating concerns about quantization errors.
--Rik