"knuttle" wrote in message
news
With a chemical camera the resolution is limited to the grain size in a
film. However with a print the quality of the paper the images is printed
on will also affect the resolution in the print

With a digital in my opinion has a large range of light conditions under
which you can get good images.

With all of the above, in both types of camera it is the lens system. Poor
quality lens gives poor quality images regardless of the film or CMOS. As
an example I have a cheap phone with a 1.3 megapixel camera. It gives me
consistently better pictures than my tablet which has a 2 megapixel CMOS.
This is evident in that with the phone I can easily get readable images of
printed pages, but impossible with the tablet.

The other thing with digital is that the quality of the image is affected by
the post-processing and the amount of noise that the sensor generates. Noise
increases with increased amplification (higher ISO setting) and with reduced
pixel size: a phone with a small sensor (so each pixel is smaller) will
produce more noise than an SLR with a larger sensor with the same
resolution.

Often this is masked by post-processing which manifests itself as localised
blurring of detail.

My SLR at 3200 ASA produces a less noisy picture than my phone camera at a
much lower ISO setting. The SLR's lens is also better, but that's a separate
issue. One other factor is that phone cameras are often a fixed focal
length, so if you zoom in you are using a progressively smaller area of the
sensor which increases noise and (even more so) decreases resolution - just
like making a print from a progressively smaller part of the negative.

Digital also has the advantage that it is much easier to correct for
different colours of light (sunlight / cloud / daylight fluorescent / warm
white fluorescent / LED / tungsten), either manually with presets or
automatically. And the sensitivity of the sensor doesn't change at very
short or very long exposures: with film you had to make corrections both for
exposure and colour cast due to "reciprocity failure" whereby the normal
rule of "reduce shutter speed by one stop requires opening up aperture by
one stop" no longer applies. With negative film it wasn't too much of an
issue because neg film can produce a usable print from a negative with more
under or over exposure, and colour cast can be corrected at printing,
whereas slide film has much less exposure latitude and has no opportunity
for correcting colour cast, apart from by copying onto a new slide with a
filter in place, or by scanning to digital.

I was surprised at how much correction scanning does allow. I took some
night-time photos of an illuminated building and grossly overexposed (I was
guessing). The slides are very pale. When I scanned them (about 30 years
later!), I could correct for this increasing the contrast so the darkest
pale tones became nearly black and the lightest, almost clear film, became
white. Given that exposure at night is very subjective anyway (there is no
one "correct" exposure) this was good enough to produce better copies than
the original. If I'd been shooting on digital, I'd have seen the results of
my guesses immediately and corrected accordingly, either by looking at the
result or looking at the histogram (proportion of pixels with each
brightness - should look *very roughly* like a symmetrical bell-shaped
curve, assuming a typical scene, which night pictures often aren't because
of bright lights or shadows which are outside the range of what you want to
reproduce well (ie it's much more acceptable have some parts which are
totally black or bleached maxed-out white).