Even _I_ got it...

David J. Littleboy
Whose jokes tend to go over like lead balloons in
Tokyo, Japan

Your foot must have gotten too soggy and sore, so you've changed it again.
That must explain the "wooshing" sound again.

Your forte isn't humour, so don't even bother.
Your reputation and actions have spoken loudly for you before.
Abravise, abusive and clueless come to mind.
I decided to treat you as you have treated others.

If your were attempting (rather poorly and patheically) to make a pun
out of "Pen-F" and "ultimate", it is a total failure.

In the Olympus diagrams, they show a reduction of the off-axis angle of
the extreme rays, but still not making the rays 100% on-axis, so you would
still need something like the Leica "staggered microlens" solution.

See the diagram about half way down this page:
 http://www.four-thirds.org/en/about/benefit.html

Note how the light rays are clearly not at right-angles to the sensor when
leaving the leans, and yet magically appear exactly at right-angles when
entering the sensor!  Some artistic licence there!

David

That is precisely the myth that Olympus would like you to believe -
whilst it may be factual, it is an Olympus exaggeration that it is at
all significant within the constraints of an SLR design.

It was a well-known fact that the earth was flat, but it turned out that
it isn't.  Ancient greeks measured its radius of curvature, implying it
was round, and Magellan later sailed around it.

What was once a "well-known fact" often turns out to be no more than a
myth and that "well-known fact" is just another in a long list of
examples.

Leica did this because many Leica lenses do indeed create extreme angles
at the sensor periphery.  However the Leica is not an SLR and hence has
no significant constraints on the proximity of the lens to the sensor.

Back to your, sorry Olympus', myth again!

Soon after buying my Canon 5D I started using my old Olympus OM lenses
on it, particularly the fairly wide angle 18mm f/3.5 - about as far from
a telecentric design as you can get with its miniscule rear element just
clearing the OM mirror.  Given how hard Olympus had been peddling their
myth and how much they had invested in "avoiding" the problem (a
completely new camera system!) I fully expected that the 18mm OM lens
would vignette much more than it did with the OM film bodies.

Guess what, Alfred?  It didn't!  In fact, it seemed to be pretty much
the same.

That, and the continual repetition of the Olympus myth on this and other
forums, made me question how significant this "light fall-off" issue
with the sensor was, so I set out to measure it - INDEPENDENT of the
lens.  I posted the results here around 18 months ago.  The raw results.
You can find the thread in the group archives if you want, but the
conclusion is that across the entire range of angles that it is possible
to create within the constraints of the Canon EF lens mount (and most
other SLR mounts are more restrictive) the silicon sensor in the Canon
5D has LESS sensitivity to light's angle of incidence than film has!

You might consider that observation to be as heretical and blasphemous
as Galileo observing that the earth and other planets orbited the sun,
but that is exactly what the measurements show - the "well-known fact"
is just a myth.

In the past 18 months I have repeated the measurements many times on
several different 5D camera bodies, and while there is variance between
them, they are all better than film in terms of angular response.

How can that be?  Well, the answer is precisely the same as the cause
oft cited by Olympus and repeated parrot fashion by yourself, as quoted
above: its the microlenses.  The problem you cite occurs because the
rays focussed by those microlenses drift off the sensitive area of the
pixel underneath as the incident angle increases.  Clearly that can only
occur after a certain critical angle is reached, and that depends on the
size of the underlying sensitive area in the silicon (not all of the
pixel area can be light sensitive, some must be metal interconnections
and circuitry) and on the aspect ratio of the lenses.  That means that
the "fact" of light fall-off is very flexible and certainly not a fact
for all sensors within the relevant angular constraints imposed by other
aspects of the camera design.

There are other effects which come into play to mitigate the "well-known
fact" as well, such as the optical cross sectional area of the microlens
surface not being the same as a flat film surface as a function of angle
of incidence.  A microlens can be approximated as a section of a sphere.
A sphere doesn't change its optical cross section as it is viewed from
different angles, it is the same at all angles.  A section of a sphere
is the same over a limited range of angles.

So, whilst Olympus and their disciples state that sensor light fall-off
is a "well-known fact", it is clearly something that can vary according
to sensor design and therefore may not be at all significant, or indeed
factual, with a suitable sensor design.

The reviews at dpreview, and elsewhere, show the effect of lens
vignetting, not sensor light fall-off.  The two are separate causes even
though the end result may be the same.  The Olympus case is that the
sensor makes the end result worse than with film, but the presence of
vignetting does not imply the sensor is the cause.  I know of no reviews
which assess this comparison, sensor vignetting v's film vignetting,
with actual measurments.

No, Olympus are claiming to address it THAT way, which is only one of
several ways it CAN be addressed (you cited one other method yourself,
which Leica use).  In doing it their way, Olympus locked themselves into
the tiny x2 cropped format with poor noise performance and limited
actual resolution but, to add insult to injury, many of their new lenses
are actually LESS telecentric than the equivalent lenses from Canon,
Nikon and possibly others!  Just like sensor light fall-off, lens
telecentricity is something that can be measured independently of image
vignetting.  Try it - you will probably be just as surprised as I was
that the FF sensor in the Canon 5D had a flatter angular response than
film.

It really is time to stop peddling the sensor light fall-off myth,
Alfred.  Anyone can, if they measure a Canon 5D (and possibly other FF
and larger sensors too), demonstrate that this once "well-known fact" IS
just another myth.

No, the main reason (essentially the ONLY reason) for using the
diminutive 4/3 sensor and other cropped format sensors is because they
are much cheaper to make than full frame sensors.

Hello, Neil:

Indeed, the "Pocket Instamatic 60" was a comparatively compact and
pretty competent unit.

Here's an excerpt, from "The Young Family's Virtual Home - An
Introduction to 110 Film Format Cameras - Kodak Pocket Instamatic 60"
<http://mywebpages.comcast.net/youngds/110Cameras/Kodak60/KodakPocketInstamatic60.htm>:

"It was the Model 60 that truly awed the photographic world at the time
of its introduction, with a feature set never before seen on a camera
this small. A fully functional, coincidence rangefinder focusing system
stunned the photographic press at the time; add to that a super fast
and ultra-sharp F/2.7 glass lens, a fully coupled CdS exposure meter
system that would automatically set the aperture and the electronically
timed shutter for speeds from a high of 1/250th of a second all the way
down to 4 full seconds for low light photography, and a small, flat,
proprietary K-size battery specially developed for these cameras and
you had something that was very special."     

Correct, the Minolta Mk I. It's a monstrosity, though, and is far
bulkier than the cute, little Pentax Auto110.

More info, here:

"The Young Family's Virtual Home - An Introduction to 110 Film Format
Cameras - Minolta 110 Zoom SLR (Mk I)"
<http://mywebpages.comcast.net/youngds/110Cameras/Minolta110ZoomSLRMK1/Minolta110
ZoomSLRMk1.htm>

Oh, yes, but "engineering arrogance" is quite understandable, where
the Germans are concerned. <g>

You want "subminiature?" Kodak's "matchbox camera" of WWII, fits the
bill, perfectly!

See, "Kodak Camera X Variations"
<http://www.submin.com/16mm/collection/kodak/index.htm>.

Quoted, from the above Web page:

"Towards the end of War World II, in 1945, Eastman Kodak Company made
the Camera-X for use by the Office of Secret Service (forerunner to
the CIA).

"Officially called the Eastman M.B. for match box it has a Bakelite
inner body containing the camera's working parts, the film and the lens
assembly. This plastic body slid into an outer metal sleeve. The whole
camera (7/8x1 1/2 x 2 3/8 inches 22x38x60mm) was then to fit into a
wooden matchbox which are typically smaller 15x40x55mm."   

Hey, if it was good enough for the OSS, who needed the Minolta 16,
anyhow? :-P

Splitting "posthaste," in two...is this supposed to be a pun, of
some kind? <g>

Cordially,
          John Turco <jtur@concentric.net>

 The usage in this thread is a little different that what I've
encountered before.  As John used it ("I invite you to join me the
next time I go off on subbing "dpi" when it should be "ppi"), it
clearly means 'substituting'.  In the cases I've seen it used, it
referred to a person rather than an object or term, as in a coach
telling one of the players on the bench "You'll have to sub for Rick
until his grades improve." and then the player telling others "I'll
be subbing for Rick at forward."  In the sense that I've seen it,
this player is often called a 'sub'.  When John subs "dpi" for
"ppi", "dpi" I don't think it would be appropriate to call "dpi" a
sub, because it . . . uh, . . . lacks substance.  :)

Maybe this is part of your provblem: an inability to learn on your
own.
Are you aware of Google, and how to use it to search for things on the
Internet?
A hint: if you put "define: subbing" (sans quotes) into Google's
search box, it will find quick definitions for you.
You could even put, for example, "crop factor" (again, sans quotes)
into that same search box and find just how many respected sites
disagree with you on the usage of that term.
Of c ourse, I doubt this will change your mind; it must be open to new
things first.

Great point, but what are AI types? I can't under emphasize how
important that is.

The "care less" matter isn't as egregious as it first appears, as if you
hear someone say it, it's clear what is meant. Also, it lends itself to
silent completion along the lines of "I could care less [about the price
of eggs] only if I really tried", at least for some, it isn't so jarring.

Now, Safeway stores in California have signs for "Can Goods". Millions
of people in this State have never seen a reference to "iced tea".

 Could you 'splain on that a bit? It doesn't seem odd to me to
categorize misses as near or far or close or wide.....

"I could care less" started as sarcasm.
Try saying it in a sarcastic tone of voice, and you'll get it.
Maybe. :)

If you see me doing it, it is because I brought the wrong hat with me, and
I can't use my DSLR and flash with the visor in the front.

If you see my pants hanging down near my knees, it's because I forgot my
belt.  If I have one pant leg inside my boots, and one outside, and both
boots untied, it's because I'm sloppy.

<heavily edited, for brevity>

Hello, Neil:

You've never heard of "rally caps," before? :-P

<edited>

Definitely. <g>

Cordially,
          John Turco <jtur@concentric.net>

The focal length doesn't change either.

ploink.

In the film days, everyone used the same "sensor" (well, set of sensors)
regardless of format size.

That meant that the f/number abstraction made sense, since it told you the
flux per unit area of film, and you knew how film responded to flux per unit
area.

But it makes less sense for dcams. The FZ20 folks think they have a 450mm
f/2.8 lens, when the flux per pixel is a fraction of what the 30D sees from
a 300mm f/5.6 zoom.

David J. Littleboy
Tokyo, Japan

[]

Interesting!

Yes, the Nikon 8400 can zoom while shooting, but I don't usually use that
feature as it could detract from the content.  I'm not attempting to make
"Art Movies" here.  <G>  Stereo sound might be a nice added extra (as the
Nikon 8400 has the wide-angle 24mm lens - but stereo would be lost with at
the telephoto end of the Canon zoom, or at least the visual and audible
perspectives would be grossly mismatched), but stereo is not not a
deal-maker.

David

Pretty consistent.  We make one such sensor which can operate with APD
gains of around 1000, and in its normal operating mode with an APD gain
of 30-80 the noise factor (ie. how much the noise has increased relative
to the gain) is better than 1.1.

Yes, and other noises that may be in the system, such as quantisation
noise.

That depends on how noisy the gain is, and if APDs were just as noisy as
conventional transistor gain then there would be little reason for them.

This device (or one very much like it) was already
discussed in the newsgroup.  Let's see, 96 dB
with 2.2 micron pixels?    96 dB is magically exactly
what 16-bits is (and note you would need at least an 18-bit
converter to deliver 16 bits of real dynamic range).
And 72 dB is 12-bits, as if anyone really believed
2.2-micron pixel cell phone cameras actually delivered
12 bits of dynamic range.  Then let's talk photons.
To get 96 db of dynamic range, you need to collect
enough photons: 65535*(read noise in electrons).
So if read noise matched the current lowest read noise
consumer dcams available (~3 electrons in Canon  1D Mark III
if reports prove out; ~3.9 electrons in 30D, 1D Mark II),
then the little cell phone 2.2-micron pixels need to collect
at least 3 * 65535 = 196,605 photons.  Typical pixel
storage is 1,000 to 2,000 electrons per square micron,
so the Toshiba technology is over 40,000 electrons/sq. micron.

What's fishy about this report?  The previous thread on this,
if I remember correctly, concluded the marketing department
got carried away converting A/D bits and didn't
focus on reality.  We'll see when (if) the device
actually shows up on the market and eclipses top
end DSLRs with large pixels with dynamic ranges that
currently don't quite reach 72 dB (12 bits).

Roger