Digital photography, changing the world

In article , Mxsmanic
wrote:

The Reids writes:

Not by that method, its simplistic, as you know. You could get a
close approximation by doing it properly, but at the end of the
day it does not matter. Why should Photoshop produce the same
false image as yellow filtered black and white? Why should a
digital camera produce the same variations on reality as a given
film stock? This is all a nonsense and nothing to do with
pre/post filtering colour film..

As I've said, most people don't understand the problem.

There are even people who understand it, follow this discussion with
bemusement, and don't really care 8-)

--
Mary Loomer Oliver (aka Erilar)

You can't reason with someone whose first line of argument
is that reason doesn't count. Isaac Asimov

Erilar's Cave Annex: http://www.airstreamcomm.net/~erilarlo

In article 1102334757.3120edfaf4f39faec8ec02207ff0e5f7@teran ews, Tim
Challenger wrote:

On Mon, 06 Dec 2004 08:45:26 +0000, The Reids wrote:

..............

Are you
going to tell us PS isn't made of light sensitive layers?

Don't do that when I've got a mouth full of coffee! Yuck!


A computer program made of light sensitive layers? I had finished my
coffee already...8-)

--
Mary Loomer Oliver (aka Erilar)

You can't reason with someone whose first line of argument
is that reason doesn't count. Isaac Asimov

Erilar's Cave Annex: http://www.airstreamcomm.net/~erilarlo

"Mxsmanic" wrote in message
...
Keith Willshaw writes:

From the dictonary on my desk

Why are definitions from the dictionary on your desk "actual," if other
definitions are not?


Because dictionaries codify conventional meanings thats
why we buy em.

Keith

"Mxsmanic" wrote in message
...
Keith Willshaw writes:

Erm no, the act of NOT filtering doesnt remove data.

The act of converting a continuous spectrum to three discrete RGB values
does, however, and in a big way.


Its a good job that cameras dont do that then

The full spectrum is still present at this point.

The full spectrum enters the lens, but it is lost as soon as the image
is recorded.


The record contains the mixture of the 3 primary colors
needed to reconstruct the image in reasonable detail,
ultimately some precision is lost whatever the media used.

Applying the equivalent of a yellow filter is trivial, removing
it is more difficult but not impossible.

You can't apply the equivalent of a yellow filter once the data in the
full spectrum has been lost.


You can - I provided a link to the software used to do just
this, you have chosen to ignore this I notice.

Yes it does, a filter simply screens out certain spectra.

But there is no spectrum in an RGB image.

Or in a color image recorded on film , it uses
3 layers

A full spectrum contains an infinite set of numbers.


So !

The reality is that a 35 mm Ektachrome slide has approx
the same image resolution as a Canon Digital SLR,
some degree of approximation is inevitable in any
recording method, digital cameras are now as good as
high end 35 mm film and everything I used to do with
filters and in the darkroom I van now do digitally.

Keith

"bogus address" wrote in message
...

Why don't you explain the visible manifestation of this invisible
radiation?
There isn't any. That's why it's invisible.
Wrong answer. I'll help you out, because this is getting old.
Black and white film records values from white to black depending on the
intensity of the light that hits it (to make things simpler I'm going to
ignore the "negative" effect andtalk about the final positive image). The
brighter the light, the closer you get to white on your print. However,
the film also responds to some frequencies of light that are outside the
range captured by color film (or digital camera sensors, or the human
eye - everything has its own profile in this regard).

Colour film is also sensitive to UV. There are some flowers that
look red to the human eye but are brilliantly reflective in the
near-ultraviolet to attract bees. Photograph one on normal colour
film without adequate UV filtration, and they'll show up blue or
purple on the print. Scan that print and there will be no overall
filter you can apply to the image that will make that flower the
colour you see without distorting the colour of everything else -
any Photoshop trick you use will have to be local manipulation, the
digital equivalent of hand-tinting.

Trying to take colour photos under domestic fluorescent light is
a quick education in how differently film and the human eye respond
to a discrete spectrum (fluorescents emit most of their light at
only a few frequencies). If your photo comes out bilious green
you can sort of patch up the result after scanning, but the result
is never as good as if you have time to select an appropriate filter
to go with the lights, or preferably use thermal-spectrum lights
instead.


Or use a digital camera that compensates, as many modern
devices do.

Keith

In article , Keith Willshaw
wrote:

"Mxsmanic" wrote in message
...
Keith Willshaw writes:

Erm no, the act of NOT filtering doesnt remove data.

The act of converting a continuous spectrum to three discrete RGB values
does, however, and in a big way.


Its a good job that cameras dont do that then

The full spectrum is still present at this point.

The full spectrum enters the lens, but it is lost as soon as the image
is recorded.


The record contains the mixture of the 3 primary colors
needed to reconstruct the image in reasonable detail,
ultimately some precision is lost whatever the media used.

Applying the equivalent of a yellow filter is trivial, removing
it is more difficult but not impossible.

You can't apply the equivalent of a yellow filter once the data in the
full spectrum has been lost.


You can - I provided a link to the software used to do just
this, you have chosen to ignore this I notice.

Yes it does, a filter simply screens out certain spectra.

But there is no spectrum in an RGB image.

Or in a color image recorded on film , it uses
3 layers

A full spectrum contains an infinite set of numbers.


So !

The reality is that a 35 mm Ektachrome slide has approx
the same image resolution as a Canon Digital SLR,
some degree of approximation is inevitable in any
recording method, digital cameras are now as good as
high end 35 mm film and everything I used to do with
filters and in the darkroom I van now do digitally.

How can PS penetrate into water, like a circular polarizing filter? On
the flip side do you use the paint brush to enhance reflection on
surfaces....

jay
Tue Dec 07, 2004



Keith

Mxsmanic wrote:
Keith Willshaw writes:
Erm no, the act of NOT filtering doesnt remove data.

The act of converting a continuous spectrum to three discrete RGB values
does, however, and in a big way.

Not in any meaningful way if the granularity is finer than the human ability
to discern differences, in the expected viewing mode.

miguel
--
Hit The Road! Photos from 32 countries on 5 continents: http://travel.u.nu

Keith Willshaw writes:

Its a good job that cameras dont do that then

Image sensors (such as those in "digital" cameras) do exactly this. So
does film.

The record contains the mixture of the 3 primary colors
needed to reconstruct the image in reasonable detail,
ultimately some precision is lost whatever the media used.

A _great deal_ of precision is lost. Almost all the information in the
original image is sacrificed. This severely limits the transformations
that can subsequently be applied to the image.

You can - I provided a link to the software used to do just
this, you have chosen to ignore this I notice.

I know better.

Or in a color image recorded on film , it uses
3 layers

Correct. The only place where the full spectrum exists is in the
original scene. So many transformations effected by optical filters are
possible only because those filters act on the continuous spectrum
reflected or emitted by the original scene. No amount of simulation can
duplicate their effect through any transformation of mere RGB values.

The reality is that a 35 mm Ektachrome slide has approx
the same image resolution as a Canon Digital SLR,
some degree of approximation is inevitable in any
recording method, digital cameras are now as good as
high end 35 mm film and everything I used to do with
filters and in the darkroom I van now do digitally.

None of this has anything to do with the point I'm discussing. The
limitations I describe apply to _any_ type of image recording, not just
digital or film recording specifically.

FWIW, most 35mm slide films have better resolution than 35mm DSLRs, if
properly exposed, processed, and scanned.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

Miguel Cruz writes:

Not in any meaningful way if the granularity is finer than the human ability
to discern differences, in the expected viewing mode.

With some types of filters, the effect is extremely obvious, and it is
impossible to duplicate through any transformation of RGB values.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

JohnT writes:

Understood. What you really need is a course which will help you to grasp simple
concepts.

Nobody needs a course to grasp _simple_ concepts, by definition.

But this concept is not simple, which is why so many people have trouble
with it.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.