Olympus-OM
[Top] [All Lists]

[OM] Re: DOF question

Subject: [OM] Re: DOF question
From: Moose <olymoose@xxxxxxxxx>
Date: Thu, 26 Apr 2007 22:09:51 -0700
Ali Shah wrote:
> Does the camera's sensor have an affect on DOF? Say
> you take a photo in which you want good DOF using an
> Olympus Digital lens on an E-series body. Take the
> same photo (using the same exact settings) with a
> N*kon or C*non dslr using their respective lenses.
> Will the photo be more crisp and sharper from end to
> end with the N*kon or C*non because of the better
> sensor? 
When I read this, I groaned. Perfectly legitimate question or questions, 
but about one of the most complex and hardest to explain aspects of 
photography.

With any luck, you have now received  plenty of references and 
explanations of the theory of DOF.

----------- Those with any form of limited attention span may skip to 
the bottom - if yo don't want to know how complicated this all is. :-)

I'd like to say something about the other questions that seem implicit 
in your questions. As Chuck so clearly and simply described, there is 
only one distance from the camera at which objects are in perfect focus, 
and DOF is simly a measure of how far from that distance one may go and 
still have the resulting image APPEAR to be in focus.

The other question that interacts with your question about "crisp and 
sharper", is how crisp and sharp is the image of the part of the subject 
that is exactly at the focal plane? The sharper the image, i.e., how 
close to a perfect, dimensionless point is the image of the star in 
Chuck's example, the better the apparent sharpness will be throughout a 
given DOF range.

The DOF tables assume a perfect lens, which is not what we are working 
with, and a perfect film/sensor system, which is also not the case. So 
before even leaving perfect focus, the star image is already a blob with 
less than perfectly sharp edges, almost certainly slightly asymmetrical 
and with the different colors focusing at ever so slightly different 
distances from the lens. All of these effects diminish apparent 
sharpness, which is, in itself, a complex and not fully defined thing.

So, as to one aspect of your question about Oly vs. N and C, one really 
has to ask which lens you are using and which models of N and C and 
which lenses you might be using.

Oly's ZD lenses are generally quite good and, at any given level of 
price/target user, at least as good as those of the others, and probably 
better at the low end than the competition. So one answer is that, given 
lenses from the same quality tier within each manufacturer's line, you 
won't gain anything from switching due to lens quality.

Another confounding factor is diffraction as the lens aperture gets 
smaller. If it weren't for the same phenomenon of optical physics that 
makes pinhole cameras work, you could just stop waaay down, and get 
everything in focus on stationary subjects like your ship. 
Unfortunately, apertures below a certain size start to lower the 
sharpness of the lens due to diffraction of light along the edge of the 
aperture. You may have noticed that the lenses for smaller sensor 
cameras stop down less far than those for 35 mm film/sensors. That's 
because of this limit.

The next issue is how well the various sensor systems work. The E-500 is 
a decent, but not outstanding sensor system now a few years behind the 
leading edge. ONe of the things reviewers talk about is "pixel level 
sharpness". Imagine again Chuck's pinpoint image of a star, this time 
hitting only one pixel of the sensor. Ideally, the resulting image 
should have one bright pixel surrounded by black pixels. In practice, 
this isn't what happens. For physical design, optical and electronic 
reasons, their are reasons that the brightness 'leaks' into immediately 
surrounding pixels, causing a result sort of like being slightly out of 
the focal plane due to DOF issues.

Current N and C APS sensor size camera bodies have less of this effect 
than the E-500. So, given equal quality lenses and good technique to 
minimize blur caused by inaccurate focus, camera and subject vibration & 
movement, and a large apparent image size, whether maginfied via large 
print, considerable cropping or pixel peeping, it is likely that an 
image from the C or N camera will have somehwat sharper appearing images 
with slightly better perceived DOF than the E-500.

Next up is image processing. Apparent sharpness is some difficult to 
define combination of resolution, contrast, saturation, local 
contrast/accutance and who knows what else. How the image is processed 
has a considerable effect on how sharp it looks, as opposed to how many 
lines of a test pattern it can resolve. So a further question is how the 
camera image parameters are set and what post processing is done.
---------------------
OK, have I made it clear that this question or questions are really 
highly complex, and essentially unanswerable without empirical testing? 
So there is no clear, simple answer in any absolute sense.

However, if your question, as I suspect, is more or less "If I dump this 
Oly and buy a C or N (As all these folks around me say I must to get  
good images.), will my problems with getting really great, deep DOF go 
away?" - the answer is no.

In fact, your example image is really an exceptional example of great 
DOF put to good use. Post processing can greatly increase the apparent 
DOF, both far and near. 
http://www.moosemystic.net/Gallery/Others/AliShah/MarineStar.htm
> Is this because the sensor absorbs more light?
>   
I actually have no idea what you are talking about here, the sensors 
will all "absorb" the same amount of light. I can say with assurance 
that it has nothing to do with the question(s) you asked.

Moose

==============================================
List usage info:     http://www.zuikoholic.com
List nannies:        olympusadmin@xxxxxxxxxx
==============================================

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