Olympus-OM
[Top] [All Lists]

Re: [OM]: 16MP DC vs film again

Subject: Re: [OM]: 16MP DC vs film again
From: Joe Gwinn <joegwinn@xxxxxxxxx>
Date: Tue, 24 Dec 2002 14:41:28 -0500
At 3:02 AM +0000 12/24/02, olympus-digest wrote:
>Date: Tue, 24 Dec 2002 10:33:02 +0800
>From: "C.H.Ling" <chling@xxxxxxxxxxxxx>
>Subject: Re: [OM]:  Dipping our Toe Into Digital, 16MP DC vs film again
>
>Joe Gwinn wrote:
> > 
> > 
> > A color camera being used to produce black&white pictures cannot have 
> > better resolution than the color with the greatest resolution, typically 
> > green.
> > 
> > Because digital cameras are either 1:1:1 or 1:2:1, green is always a good 
> > choice for determining spatial resolution.  So, I use the green channel, 
> > and also compare it directly with B&W only pictures as well.
> > 
>
>Yes, I know about the 1:2:1 color filter distribution, but the
>resolution seem not much affected by this, I suspected it will only
>affect the color accuracy (wrong interpretation of color on fine
>details) but every pixel will contribute to the scene details.  

If all colors don't contribute equally, one gets color noise, so one cannot 
just count the total pixels regardless of their color.


> > > The linearity is probably why good digital cameras are reported to handle 
> > > skin, sparkles, eyelights, et al, so well.
> > 
>
>That is the point and I heard long ago in DC news group that some
>professional photography (at least they said they live on this) think
>their "new" Nikon 990 is better than their Hass for studio portrait on
>output quality.

One wonders how this can be so...


> > >
> > >http://www.accura.com.hk/50-02.jpg  (200K)
> > 
> > I looked at the picture, but I don't know how to interpret it.   What is 
> > the purpose of a lens test slide?  To test slide duplicators?
> > 
>
>The test was performed in around 1990 to test the performance of my
>Zuikos, all shots were taken at around 1:50 or something similar. To
>interpret it is easy, you should download it in photo editing
>software, there should be dimension information on your software.
>Check which point you think you can still identify the lines. I would
>say at least the "200" position. Measure the width of this group of
>lines, there is total 15lp and at the point "200" the width is 0.2mm,
>so 5*15=75lp/mm.

Ah.  The markings are the width in microns.


> > To my eye, the E-10 shot is nicer-looking, but the scanned film has higher 
> > resolution.  Sort of as expected.  So, there you have it.
> > 
>
>You really see the scanned film has higher resolution? Yes, it may
>look to have more details but to me they are just film grain :-)

If you look at very narrow features (like the various metal rods) in the two 
pictures, the E-10 will show a wide but smooth rod-like object, while the 70mm 
photo will show a narrower but somewhat noisy  image of the same rod-like 
object.  What's happening is that the E-10 is doing a gaussian blur to reduce 
noise and to supress the CCD pixel pattern (which would otherwise produce 
jagged edges).

The most basic test of resolution is to photograph a high-contrast knife edge, 
and then to measure the illumination versus distance (perpendicular to the 
edge) function.  The film will win in the present test case, ecen though the 
film is somewhat noiser (grainier).


> > If you have the energy, could you make test shots like these, only with 
> > wider dynamic range, and including some specular highlights?  Thanks.  The 
> > claim is that digital does better than film in such situations.
>
>Ok, I will try it out later but to my experience the E-10 has very
>good shadow details (although a bit noise), but for very bright spots
>most DC are much poorer than film, it blow out and enlarged the spots.

This blow-out is called "blooming" in the CCD literature, and is supposedly 
solved in modern CCD chips.  Blooming is caused by photoelectrons overfilling a 
CCD pixel and spilling over into adjacent pixels.  The newer chips have what 
amount to overflow drains to divert the excess photoelectrons.  (If one wants 
to know more than is healthy abbout CCDs, Janesick's recent tome on the subject 
is the scientist's bible.)

If one reduces the exposure, the blooming should be reduced, while the 
linearity and dynamic range of the CCD allow shadow detail to be retained, as 
the theory goes.  So, in the experiments, it might be useful to bracket by one 
and two stops, for the experience.  I guess the problem will be the noise in 
the shadows.

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 >


<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz