Joseph Arkins (email@example.com)
Wed, 27 Oct 1999 21:00:51 -0400
Apropos of visual comparison as a way of measuring density differences, I'd like
to point out that early densitometers did not use photoelectric cells, but rather
used an optical device that compared the segments on a circular transparent step
wedge with the area of the negative being measured. This process was used by the
early Kodak densitometers, and in my experience is rather accurate (using a
calibrated step tablet rather than a negative as the measured entity.) (I own a
1938 Kodak Model One densitometer, which is breakdown-proof and a joy to look at,
as well, what with its crackle finished metal and wood case). Since early color
photographers used this device to balance separation negatives (a fairly critical
application), this would suggest that they are quite accurate devices, indeed.
> In a message dated 10/26/99 5:02:18 PM Eastern Daylight Time,
> firstname.lastname@example.org writes:
> << The eye is
> far more accurate in judging differences in density than a
> densitometer. I recall an article from Scientific American (back in the
> seventies) discussing the eye's capability to discriminate a
> differential of three photons or some such small amount. >>
> I think the reference is to the threshold of light against pure darkness, not
> a variation between shades light or dark. I have an article in front of me
> from a past issue of Shutterbug, written by Ctein (I don't know what issue, I
> could only find a copy reduced down to 8 1/2 by 11 ) that discusses that the
> eye does not see in continuous tones, only the brain integrates the
> information and tells us we do. Test show that we can only see a two percent
> difference in luminosity, or 0.006 desity units.
> This means a young eye can see 650 steps from pure, blazing light to no light
> at all. A trained artist's (or photographer's) eye may approach 900 steps.
> The argument that the eye is sensitive enough for not needing a densitometer
> for photo work is still correct (Ctein says if you can't see it, it doesn't
> matter anyway), but not that the eye is necessarily more sensitive than a
> I did find I could tell the difference on a print of .01 density where my
> Stouffer 30 step tablet goes from 1 to 15 and 15 to 30. The two 15 steps were
> different by .01, and the old Seagull Select FB paper would show a difference
> in that range. I had the step tablet checked on three different
> densitometers, and found they could each measure .01 difference. I could not
> see it, because these steps on the tablet are 1/4 inch apart.
> Ctein also goes on to talk about prints, where the maximum visual steps (for
> the eye) is 250 when the print is viewed under 200 foot-candles of light, and
> 350 steps under 500 foot-candles. Room light is usually 20 to 40
> foot-candles, where the darkest 25 percent of the image will lose 50 percent
> of the steps. Track lighting, and sunlight, will restore the visual step.
> Ctein gives two references, a 1980 Kodak book (out of print), and a 1975
> article from the "Journal of Mathematical Biology."
This archive was generated by hypermail 2.0b3 on Fri Nov 05 1999 - 21:26:28