RE: 8 bit precision vs 16 bit
RE: 8 bit precision vs 16 bit
- Subject: RE: 8 bit precision vs 16 bit
- From: Marco Ugolini <email@hidden>
- Date: Thu, 13 Dec 2007 10:41:07 -0800 (GMT-08:00)
Mark Rice wrote:
>It is chimerical in the sense that if one has a CMYK value of 243.36, 12.55,
>34.77, and 65.88, the values going to the video card and the output device
>are going to be 243, 13, 35, and 66. I am not sure what good it does to view
>those decimal values.
It would allow one the option to view that actual values, as opposed to the rounded-up ones. And in 16-bit files, it would definitely make a difference, specially in wide-gamut spaces.
At any rate, in 8 bits, the RGB numbers would not be rounded up, since they already are in the same scale as the 8-bit architecture.
As for CMYK, Grayscale, and the L* channel of L*a*b* (all of them using 0-100 scales), the decimal values are actually not internally rounded up to 0-100 integers. Someone correct me if I'm mistaken, but they remain 0-255 integers, though viewed via a 0-100 scale -- meaning that they remain 0-100 decimals all the way through the process. I don't see how else it would work in an architecture that forces values to be one of only 255 possible ones.
Marco Ugolini
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden