Re: DSLR vs Superzoom - help please

Paul Mitchum wrote:
Roger N. Clark (change username to rnclark) <username@xxxxxxxxx> wrote:

Paul Mitchum wrote:
Roger N. Clark (change username to rnclark) <username@xxxxxxxxx> wrote:

The spread of light gathering ability of large pixel DSLRs versus small
pixel cameras is currently about a factor of 16 (more with the newest
crop of sub-2-micron pixel cameras). That's an 4-stop difference! So a
large pixel DSLR working at f/5.6 gathers more light than a small pixel
camera working at f/2 (if such a camera existed).
Such a camera existed; I just mentioned owning one in the paragraph
you're responding to. f/1.8, in fact. And you know what else? The
sunny/16 rule applied to it, just as it applies to the DSLR with a
larger sensor, which wouldn't be true if you knew what you were talking
about.
So you have a camera that has 24mm f/1.8 to 300 mm f/1.8 lenses?

Your doubt was whether a point and shoot camera was ever made that had
an f/2 lens, not whether it was a constant aperture long zoom.

Look at the SUBJECT line: superzoom. If I was questioning
whether point and shoots had fast f/ratio one focal length
lenses, I would have compared it to the common SLR lens:
50 mm f/1.2, which in a tiny camera would need to be f/0.3 to
deliver the same light level per pixel.

F/1.8 on a small pixel camera still collects less light per pixel than a
large pixel camera at f/4.

What you're saying here is both misleading and irrelevant. f/4 is always
f/4, regardless of the size of the sensor, because the 'f' in f/4 stands
for 'focal length,' not 'sensor size.' Of course it's true that less
light is absorbed by a smaller sensor, because, well, duh: It's smaller.
But that doesn't transform f/1.8 to f/4 in any sense.

The fact that you think it is irrelevant shows you do not
understand. I agree that f/4 is "always f/4," but that does
not mean they produce the same image, or the same light levels.
Example: why do you think astronomers build very large telescopes,
even though they are f/4 to f/10 lenses (typically)? Hint:
the larger lenses deliver more light to see fainter.

AND STILL, on top of of that, the smaller sensor has an ISO rating,
doesn't it? So it can take the photo you want to take at a given ISO
rating, plus aperture and shutter speed. The sunny/16 rule still works.
It all boils down to: Know Your Equipment. Know what it does, and why
you'd want to use it or not.

Did you know that the ISO ratings of different sized electronic sensors
(e.g. CCDs and CMOS) are defined differently for each sensor?
In digital cameras, ISO is rated as a fractional accumulation
of electrons relative to the full well (sensor saturation).
A large pixel sensor, like the Canon 1D Mark II has a
maximum signal of 50,000 electrons at ISO 100. So a normal
exposure (as given by the camera's meter) of an 18% gray card
results in about 9,000 photoelectrons. Canon S70's small
pixels only hold a maximum of 4100 electrons at ISO 100, and
the normal exposure is defined on the 18% gray card to be
about 740 electrons.

Let's try explaining another way so you can see the light: [..]

I understand completely. I just think it's totally irrelevant. If you
want to argue that smaller sensors require more 'fudging' in the form of
sensor amplification and excessive noise reduction in processing as a
reason to avoid P&S cameras in general, then I'd agree with you. But the
OP was about whether to choose a 'prosumer' P&S or a DSLR, and the big
difference there is that with a DSLR you can choose lenses.

It has nothing to do with fudging. It is fundamental physics,
and not only is it relevant, it accurately defines performance.
Example, the above 1D Mark II versus S70 illustrates that
physics. Even though the same f/ratio lens is used on
each camera, looking at the same gray card illuminated by the
same light source, with the same exposure time, results in
about 9,000 electrons in the large pixel camera, and
740 electrons in the small pixel camera. The signal/noise
ratio (S/N) in the ideal case in this example (which is the
case for all modern digital cameras) is the square root of
the number of photoelectrons. So the 1DII gives S/N = 95,
and the S70 S/N = 27. Now, guess what? The ratio of the
signal levels (9000/740) = 12.2, and the ratio of the
pixel areas (8.2*8.2/ 2.3*2.3) = 12.7 (pretty close).

So, in summary, the signal levels in a digital camera are
proportional to the area of the pixels, and the signal
to noise ratios are proportional to the linear size of a pixel.
Camera manufacturers adjust gain and ISO definition to
give the same value in the digital output file.
These fundamentals accurately describe the performance
of digital cameras.

All this accurately explains why large pixel DSLRs have such
high ISO performance relative to small pixel cameras.
In the above example of the 1D II versus S70, if the 1D II
ISO were increased to ISO 1200, it would record the same number
of photoelectrons as the S70 working at ISO 100, using lenses
with the same f/ratio, you would get 12 times faster
shutter speeds with the 1D II.

Roger
.

Relevant Pages

  • Re: Any thoughts /news on Foveon sensors?
    ... >the design of the sensor and the fact that the pixel count hasn't changed. ... The X3 sensor doesn't suffer from colour moire on B&W images, ... >anti-alias filter of a Bayer sensor camera. ...
    (rec.photo.digital)
  • Re: The Joy of Pixel Density
    ... critics here have no viable working model in their heads of what a sensor ... That is the only way to compare pixel density, ... that higher pixel density lowers DR or increases noise. ... Who said I was trying to tell people what camera to buy? ...
    (rec.photo.digital)
  • Re: SLR concerns, owners experiences please.
    ... pixel size, and in particular what is called the Unity Gain ISO ... Therefore I want a large sensor with as few ... such a camera does not exist. ... I'm thinking about fill-in flash. ...
    (rec.photo.digital.slr-systems)
  • Re: Measure distances w 1mm precision
    ... For a good camera system you need at least 3000-5000 Euro (camera and lenses). ... With a 20 mm lenses and a sensor with 1280 x 1024 pixel you will have an field of view of 850 mm at a distance of 2 m. Lenses with a smaller focal length are also really expensive and not accurate enough for measurement systems. ...
    (sci.image.processing)
  • Re: Factor Effecting High ISO Pictures - Camera or Lens
    ... One of my complains about this combination is the high ISO ... of what lens is on your camera. ... It is the camera that determines how noisy ... keep dust off your sensor. ...
    (rec.photo.digital)