>"Neuman - Ruether"
>news:3a3b5da6.36233559@newsstand.cit.cornell.edu...
>> On Thu, 14 Dec 2000 22:05:09 GMT, "Alexander Ibrahim"
>>
>> >"Neuman - Ruether"
>> >news:3a3add0a.3305058@newsstand.cit.cornell.edu...
[...example removed that didn't make any sense to me, sigh...]
[...other examples, not responded to, also removed for the sake of brevity...]
>> How? the zoom lens is more likely to lose most of a stop
>> as it is zoomed long. The imaged brightness of an object
>> in its view does not change more than this with zooming OR
>> moving (and not at all if you just move, without zooming).
>> You are misusing physics, methinks....;-)
>I don't know how you can misuse physics. I could be wrong, but then I am
>pretty good at math...lots and lots of practice.
I am mystified... Maybe we are saying the same thing, but
not understanding each other's terms... (The key, I think,
lies near the end of this, but I plow forward, anyway, since
some of this may be useful to others...;-)
>Zooming loses about a stop in most, but not all, lenses. Canon has a
>70-200mm f2.8 L-series lens that I happen to love. 70mm or 200mm you can
>open up to 2.8. You knew that though. Right ?
Yes. "Constant-aperture" lenses maintain the relative
f-stop by actually changing the actual diaphragm
size to compensate while zooming. Most zooms don't do
this now due to cost/size considerations, but most
expensive and most early zooms did use diaphragm
compensation to maintain a constant lens "speed" with
zooming.
>Zooming itself does not cause you to 'lose' light.(Which is what dropping a
>stop means.)
If you maintain a constant diaphragm area, by the
definition of relative aperture (f-stop =
FL divided by aperture diameter [in a "simple"
lens]), zooming with a lens that does not use a
compensating diaphragm will result in lens "speed"
loss when zoomed long (or, in practical terms,
"a loss of light"...). Other factors apply in the
lens design, and the loss is unlikely to be in
a direct 1:1 relationship with the FL change, but it
WILL relate to it.
>Go to your camera store and get Canon's brochure on EF lenses.
>Many of the lenses have a little diagram next to them showing all the lens
>elements...there are a lot of them. All those pieces of glass have nearly a
>1:1 ratio.
???
>They add up to f2 or worse. This is usually the lower number on a
>zoom lens.
??? This makes no sense - sorry...
Complex lenses have absorption/reflection losses,
but this is true for lenses regardless of speed,
and it is generally far less than a whole f-stop.
I have seen only one instance in hundreds of lenses
I've checked where the losses approached one f-stop,
and that lens hasn't been made for at least 30 years.
Methinks you may be drawing conclusions where none
can be drawn...
>Like holding a magnifying glass at the wrong distance, it gets harder to
>focus the light coming through as you move these lens elements closer
>together...so you have to use a smaller aperture (higher f-stop) to inrease
>your field of depth simply to obtain accurate focus.
???
Again, this has no basis in optics...
I am mystified...;-)
Optical design does not work with major
approximations (only minor ones...;-);
the above would result in uselessly soft
lenses of VERY slow speed, if true...
Elements are designed to work properly
with small spacings, and even with none.
The Tessar-type lens I mentioned earlier
in this thread has only four elements, two
of which are cemented together, leaving
two close-spaced element sections for two
air-spaces - yet these lenses are often quite
sharp
at the widest stop...
>Indeed zooming towards the telephoto range causes the light that does get to
>the imaging plane in focus to be light from a smaller relative angle of
>view. You collect more of the photons from that smaller area for focusing.
???, again...
Photons are reflected from subject-points in
a mostly spherical way (limited or modified by
the subject shape and reflectivity characteristics).
All the subject-points combined give the reflectivity
characteristics of the object photographed. By
selecting a lens FL and a light receptor size (the
film or CCD size, after masking), you select the
angle of acceptance of the group of the photons
emitted by the subject (in straight lines). Nothing
about this changes the photons-per-unit-area
of the subject collected; the correspondence is
the same between subject emission density and
subject reception density over the angle of
acceptance chosen by choosing a FL; one photon
emitted by the subject, if accepted by the light
sensor, is by definition accepted in a 1:1 fashion
(barring losses due to atmosphere, lens
absorption/reflection lens speed, etc.)...
There is no inherent loss due to distance between
the point of reflection of light in the subject
and the arrival of that point of light in the image
(assuming .
>So...with the same framing on a subject, you can maintain the same exposure
>as your imaging surface becomes more distant. If you are using a prime lens
>you will have to change your exposure. Hence spot metering systems on film
>cameras.
Again, this does not follow...
[...]
>Seriously, any understanding you can gain of your tools and how they work
>will aid you in your craft in the long run. It does not matter what craft
>you are practicing.
>
>I mean, that is why you have camera reviews on your site right ? So people
>can understand their choices in tools.
>
>At the end of the shoot your most important tools are, in order, light and
>your camera. If you fail to accurately understand these tools you will fail
>in your tasks.
This is generally true, though many great photographers
had little technical understanding of their medium. While
this often damaged the technical excellence (in arbitrary
terms...) of their work, the aesthetics of their technical
shortcomings often didn't interfere with, and often enhanced
the work (Bill Brandt, Julia Cameron, etc.).
>Right now it seems to me that you know a lot of the "rules of thumb" that
>can be used in most circumstances. You don't seem to know where these rules
>break down or why. The lighting issue we are discussing is one of those
>places.
We may disagree here...;-)
>It so happens in this case that CCD technology, and the physics of light
>combine with a low light scenario to break the ability of these fine
>cameras. It is in failure that we learn the most about these and other
>tools. In order to learn from our tools when they fail we must understand
>them and the circumstances of their failure.
>
>Stop arguing with me and do the test. Your school has a theatre no doubt. Go
>there and set up a test. Then try to explain it with the notions you have. I
>think you'll find physics works better.
>
>Here is a good one.
>
>Light a piece of sheet music on a stand with a 250watt household bulb on
>one side of the stage. Set your camera on a tripod right next to the stand.
>Make sure the light, the camera and the music are very close to the same
>height. Set aperture priority and use the highest F-stop that gives an image
>you can see detail with. (As close as you can to underexposure, meybe even a
>little underexposed.) Bring down the house lights. Now go from 3-4 feet to
>10 feet. Then go out to 20 feet. (use a tape measure.) Do not zoom. You may
>have to use a waveform monitor on your composite output to get a solid
>measurement. The peak on the waveform monitor will drop as you back away. On
>a monitor you will see things get dimmer.
Ahah! ;-) You have given an example that will work to
solve the misunderstanding, I think... In this example,
small illuminated areas are surrounded by darkness. By
your observation, moving away from the illuminated areas
appears to reduce the brightness of those areas. I
think it does not, and that is easy to both demonstrate
and prove. But, the AVERAGE light level of the whole scene,
as the illuminated areas lose relative importance to the
overall darkness, WILL go down (as sensed by you, and
as indicated by the monitor [that cannot sense the
ever smaller peak values in the scene]). But, if you
choose to fully compensate by the amount your theory
would have you do, or even by the more moderate amount
an averaging meter would indicate, you would overexpose
in an obvious way the illuminated areas - so you choose
moderate compensation to increase the impression of
overall scene brightness as its average brightness goes
down... But, the brightness of the illuminated areas does
not actually change with distance, but it "appears" to
with changes in proportion - and compensation as you
advise (excluding lens losses with zooming) will result
in the brightening in the image of the ever smaller
illuminated subject areas.
>I have little else to say on the topic in this forum.
>
>If anyone likes, email me, and I will write articles for my site discussing
>optics and lighting.
Ummmm...;-)
While adding compensations that do not follow absolute
rules of illumination to our exposures are often adviseable,
assigning shakey "absolutist" justifications based on
faulty views of physics may not be adviseable.....;-)
Have fun!