Digital BW, The Print

After way too much screwing around with a print densitometer (X-Rite 414, Status A, for anyone interested) and a scanner (Epson 2400) to measure print densities, I'm beginning to think the very best density measurement tool is a digital camera.

You need one with raw output, and reasonably low noise. I've been having some success with a Nikon D70.

Whats really wrong with the densitometer is the way it illumiates and reads the test target. The densitometer (and for that matter, some really nice spectrophotometers) hit the paper with a tightly columated beam of light, perpendicular to the paper, and read with sensors at a 45 degree angle to the paper. Some use a fiber optic light sourfce and illuminate at only 45 degrees, and read perpendicular.

The whole perpendicular vs. 45 degrees thing is ingrained into the color industry, and I'm beginning to think it's to blame for much of what is going wrong in the field. It's sure what's going wrong with the QTR curves I'm building...

On matte papers (and on platinum prints, for that matter) there's image density "beneath" the image that the densitometer is reading, and that a human eye can't see. If you want to see what I mean, take an image printed with QTR on epson enhanced matte ona a 2200, and have a look at the shadow detail with the picture illuminated normally. Then hold it so some light shines through the paper. The shadow detail is there, it's just not visible from the front, because the curves overinked the paper in the shadow details. And why did that happen? Because the perpendicular/45 degree densitometer read deep into the paper, exactly the way that humans don't!

Scanners are almost as bad. The light source on the 2400 is far from perpendicular, and the optics assembly is perpendicular in the center of the bed, and diverges from perpendicular towards the edges...

Now, a camera, at a reasonable distance, and with a big, diffused light source, sees density much more like a real eye. And the curves I build from it don't lay down as much ink in the shadows.

End result, the density doesn't change, relative to what a human sees, but there's more shadow detail. Although the lying densitometer says I've lost considerable DMAX, you can't see it.

Now, if I just had a colormetric digital camera, to get the grays neutral...

> From: koloshor [mailto:wiz@...]
>
> After way too much screwing around with a print densitometer
> (X-Rite 414, Status A, for anyone interested) and a scanner
> (Epson 2400) to measure print densities, I'm beginning to think
> the very best density measurement tool is a digital camera.
>
> You need one with raw output, and reasonably low noise. I've been
> having some success with a Nikon D70.
<snip>
> Now, a camera, at a reasonable distance, and with a big, diffused
> light source, sees density much more like a real eye. And the
> curves I build from it don't lay down as much ink in the shadows.
>
> End result, the density doesn't change, relative to what a human
> sees, but there's more shadow detail. Although the lying
> densitometer says I've lost considerable DMAX, you can't see it.

You probably don't need to take an actual picture, just use the camera's
light meter. The density range you need to cover corresponds to about ten
stops, which is well within what any decent camera can do.

--

Ciao,               Paul D. DeRocco
Paul                mailto:pderocco@...