At 2:02 AM +0000 3/23/03, olympus-digest wrote:
>Date: Sun, 23 Mar 2003 03:10:16 +0800
>From: "C.H.Ling" <chling@xxxxxxxxxxxxx>
>Subject: Re: [OM] Re: Some comment on digital resolution and Oly 4/3 system
>
>- ----- Original Message -----
>From: "Jan Steinman" <Jan@xxxxxxxxxxxxxx>
>
> >
> > Not really. In raw mode, some digicams will give you 10 bits --
>essentially 10 stops. Drum scanned Velvia will yield 36db of dynamic
>range -- about 14 stops.
>
>I think 36db is a wrong value, most agree slide has D-Range of roughly 4.0,
>which is 10000:1 and 80db.
This is a tricky area. With optical measurements, the rule for computing
decibels from ratios isn't quite the same as for electrical, because optical
measurements measure optical power, not the intensity of the electric field of
the light.
One measures the optical density of an object by measuring the optical
intensity with and without the object in the optical path, and computing the
ratio of the two measured intensities. Mechanically, the measurement is made
using a photcell or photodiode, a light source, and some way to measure the
photocurrent. (I have made such measurements using a desk lamp, a silicon
photodiode and a digital multimeter.)
The current coming from the photodetector is proportional to the illumination
falling on the photodiode. Illumination has units proportional to optical
power per unit area, so the measured current is proportional to optical power,
and the ratio of measured currents is the ratio of optical power with and
without the object whose density is measured.
For power ratios, decibels are ten times the base-ten logarithm of the ratio.
Optical densities are the base-ten logarithm of the illumination ratio. In
other words, the optical density is one tenth the decibel value. (Actually,
decibels are tenths of the base unit, the Bel, named after Alexander Bell of
telephone fame, so optical densities are in fact reported in Bels.)
So, a D-Range of 4.0 is 40 decibels, not 80, and 36 dB is plausible
(corresponding to a D-Range of 3.6). It would be 80 dB only if the ratio of
the electric fields of the light were 10,000:1.
Joe Gwinn
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|