Does anyone know if converting a 16bit RGB file to 8bit before sending to say an Epson Inkjet?

I thought I heard Scott kelby say that after editing in 16bit convert file to 8 bit THEN priint.

It might be worthwhile determining if your Epson printer even accepts 16-bit data. Maybe Epson's drivers are already converting pixels to 8-bits by default?

Human vision (luminance latitude of the eye) can only see about 7 bits at any moment. Many DSLR's are now recording 12-bits or more. I'm not sure what the current status of point & shoot cameras is; most users don't even care.

The first two least significant bits are always noise. So a 12-bit DSLR really records only about 10 good bits. A point & shoot that records 8-bits only possesses 6 good bits.

So, why worry about all this techno jazz? More color depth gives your computer more latitude for human error, or more latitude in contrasty shots. Color depth is the 3rd dimension of resolution that everyone forgets. The other two resolution dimensions are height and width - which is what we commonly refer to as resolution (areal resolution).

Real life has about 17-bits (i.e., 17 "ƒ-stops") of luminance - like during a morning sunrise. Midday shots probably have a few stops less. Somehow, you've got to capture all that light, and then compress it down somehow to suit your tastes.

If the printer can print 10-, 12-, or 16-bits, I wouldn't reduce the color depth resolution. If the drivers reduce the resolution "on-the-fly," I wouldn't bother doing it manually, unless you want to determine if PhotoShop or the Epson drivers do a better job.

Viewing an RGB monitor will display a lot more nuances of color than viewing printed ink on paper via reflected light. The same goes for transparencies vs images on photographic paper. Even with 6 and 7 color printers, the range of ink on paper is not that great. Most printers are set up in the sRGB color space, and 8 bit is all they can handle.

Thanks Paul.

When Scott said convert 16 bit files to 8 bit before sending to a printer... here's what he meant....

1. When sending the file to your Epson... There is no need to convert. That's the best way to proof your images.

2. When sending it to someone else it is better to :
a. Flatten the file first (that way any adjustments or filters you have made will be applied at 16 bit) then convert to 8 bit and the proper colorspace.

That's a good and safe workflow.

Jb.

quote:

Originally posted by Junebug:
When Scott said convert 16 bit files to 8 bit before sending to a printer... here's what he meant....

1. When sending the file to your Epson... There is no need to convert. That's the best way to proof your images.

2. When sending it to someone else it is better to :
a. Flatten the file first (that way any adjustments or filters you have made will be applied at 16 bit) then convert to 8 bit and the proper colorspace.

That's a good and safe workflow.

Jb.

Thanks again.
I think it was the pro/con of sending it to an epson that raised the question