On 10/24/2010 3:32 PM, John Hudson wrote:
> As to a lens's optical quality and the reference to "high end Zuiko lenses" I
> was infering a consideration of the "negative" from which a large hard copy
> print could be made. I think it can be safely assumed that a negative created
> with say these Zuiko film camera lenses for example .. MIJ 50 / f1.8, 85 /
> f2, and 90 / f2 ......would, all other things being equal and favourable,
> generate a sharp large hard copy print second to none.
I'm going to leave aside silly issues like what you mean by a large print and
whether you might limit the "second to
none" part to similar or smaller film/sensor formats. "all other things being
equal and favourable ", a 4x6 foot print
from 8x10 film is going to show more detail than a comparable print from 35mm
film, no matter what.
More practically, I have seen Bob's large prints from 35 mm film (OM), 4/3
sensor (E-1) and 24x36 mm sensor (D3). I can
say pretty unequivocally that the D3 prints are the sharpest, cleanest, most
precise, whatever term(s) apply.
> Likewise with most, if not all, latter day Leica lenses. With digital there
> is obviously no film "negative". The comparable image storage medium is a
> digital file created by the camera's sensor generated by the light entering
> the camera through the lens. What is common here is the "light entering the
> camera through the lens". What is different is the medium on which the latent
> image is stored.
>
> What interests me, and what provoked my original question / observation, is
> the relative contribution provided by the camera lens to the generation of
> the latent image.
Although that wasn't how I understood your original question, I did answer it
as best I could. Saying that the G12 lens
outperforms the sensor system, particularly in the realm you mention of
sharpness. We have a situation here that
sometimes happens, where sensor alone is changed, with the same lens, in a
fixed lens digicam. Dpreview, in their G11/12
reviews, states and shows examples that the same lens resolves slightly more
detail on the 14 MP G10 than the 10 MP G11/12.
Pretty definitive, the case of the specific camera you talk about.
> On a side by side comparison of two large hard copy prints, one generated by
> a "high end" Zuiko film camera lens and one generated by a lens on the G12,
> which one would be an obljectively judged better print [colour rendition,
> sharpness, etc, etc].
I don't think this is a definitively, quantitatively answerable question. If
you are talking only sharpness/detail, I
suspect, but am not sure, that "all other things being equal and favourable .",
using proper film and a good copy of a
Zuiko like the 50/2, that the film print might well beat out the G12 print.
To even start to get a meaningful answer, as opposed to tossing about
generalities, you would need to take the same
shots with both at the same time and place. Then there are sooooo many
variables twixt shot and print. Is the film shot
to be wet printed by a master printer? Is the film to be scanned, then both are
to be post processed and printed by a
master digital printer? The film results simply won't look the same.
Then, who's to do the judging?
> In combination, if the G12's lens AND sensor generate a superior print is
> that superiority more attributable to the optical quality of the lens or to
> the technology of the sensor? If this is so, what is the relative
> contribution of each of the lens and the sensor?
I hope I've already answered that above. The mix will differ with different
cameras.
> If the film camera lens image is superior, is there a way in which the
> optical quality of the lens can be compared against that of the digital
> camera ?
Sure. Take the G12 apart and test both lenses on an optical bench, then adjust
the results to the same final image size.
> Hopefully, the above line of thinking and enqury makes sense.
Not that much, frankly. See below.
> For someone who was brought up believing that the quality of the lens optics
> was paramount it is challenging to redirect that thinking into believing that
> the quality and technical superiority of the sensor can also make a
> significant contribution to ultimate hard copy image quality.
It's not that simple. Put a crappy lens in from of a good sensor or good film,
and you get poor resolution/sharpness.
That is the exact point of the Holga, Diana and other cameras and of pinhole
photography.
Put a great lens in the world on an early DSLR and ultimate sharpness will be
sensor limited. Still, other IQ aspects
may go the other way. Put it on a 24 MP DSLR, and you may find lens
imperfections that weren't obvious on film. Folks
like AG are still using E-1s for their color rendition and other qualities
other than resolution.
The truth is that the latest sensor systems, even APS size, simply outperform
35mm film in the area of resolution of detail.
Your particular choice of camera for this discussion seems odd to me. 35 mm
film has an active image area of 24x36 mm,
864 cm². The G11 sensor is 7.6 x 5.7 mm, 0.43 cm². How is it meaningful to
compare them?
What you can get from 5.7x7.6 mm film (About half a 110 format frame.) will be
junk in a large print, compared to 35 mm
file. In this case, the sensor is greatly outperforming film, on a comparable
basis. The lens requirements are very
different, too. The digilens needs to resolve much finer detail on the sensor,
but over a much smaller area.
> On the assumption that this redirection of thought is necessary
In the case of interchangeable lens cameras, perhaps it is. For fixed lens
cameras like the G11, it seems to me to make
no sense. the camera is, in effect, a black box, and should simply be judged by
what comes out of it against other
cameras. It shouldn't matter which parts of it's internal parts and processing
create the output, as there is no way to
change them.
> how much of that paramountcy should be directed towards the sensor and
> whether this should be at the expense of accepting a degredation in the
> quality of the optical glass used to manufacture the lens?
Now I think you are getting off into the edges of silly land. In the case if
interchangeable lens cameras, that need not
be a trade-off, if you can afford a good lens on a good body, or may be an
personal trade-off depending on available
funds and the primacy of needs in other areas.
In the case of fixed lens cameras, those decisions have already been made, and
your choice need only look at results.
And what's this about "accepting a degredation in the quality of the optical
glass used to manufacture the lens? "? High
end compact cameras from the major makers use the same optical technology in
those lenses as in their top DSLR lenses.
It's not all about glass, as mechanical sophistication, precision, reliability
and many other factors are involved in
determining and maintaining optical performance.
But lets talk expensive optical glass. Lets assume the glass for a particular
element in a lens for 35 mm costs $100.
Just as the reduction in linear dimensions of about 22% becomes a reduction in
area of 95%, it becomes a reduction in
volume of a lens element of 99%. The cost of glass for the same element in the
same lens formula, scaled down for the
small sensor of a G12, is about $1.10.
Hm. Perhaps the difference between $.50 and $1.00 in the cost of a $500 camera
may be a smaller factor than the
difference between $50 and $100 in the cost of a $1000 lens?
A final comment about digital sensor systems. I often carry both a 12 MP G11
and a 12.7 MP 5D. If you believe in simple
stats, there should be little difference in IQ between them. Yet all pixels are
not created equal. The G11 images are
amazingly good, but even at base ISO of 80 show noise, whereas the 5D doesn't
have as much noise 'til maybe ISO 800. The
5D also has better dynamic range and better 'per pixel sharpness', clearer
demarcation between pixels along an edge.
The G11 also has some odd little artifacts viewed at 100%. They aren't visible
at small to medium print sizes, but are
bound to show up eventually as print size increases. They are one reason I'm
not sure about the G11 vs Zuiko on film in
large prints question.
Moose
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|