shooting digital vs film
The Dave©
shooting film? Is there any aspect that the thinking or approach would
be different? It seems aperture and DOF would be the same, for
example, but white balance vs film type would be approached
differently. In film, not all 100 ISO films are the same, and white
balance seems to be the digital variable of what type of film one is
using, if that makes sense.
--
"Through pride we are ever deceiving ourselves. But deep down below the
surface of the average conscience a still, small voice says to us,
'Something is out of tune.'" - - - Carl Gustav Jung
k
digital = 4 stops
"The Dave©" <n...@no.com> wrote in message
news:Ksb%b.968$Cf3...@news01.roc.ny...
Joseph Meehan
> What would be the most drastic difference in shooting digital vs.
> shooting film? Is there any aspect that the thinking or approach
> would be different? It seems aperture and DOF would be the same,
Not exactly ..
> for
> example, but white balance vs film type would be approached
> differently. In film, not all 100 ISO films are the same, and white
> balance seems to be the digital variable of what type of film one is
> using, if that makes sense.
It's a new tool. You need to learn how to use it. Some differences are
more aparent than others. There is a great deal of difference over all.
I know that is not the answer you were looking for, but what difference
is most drastic is going to depend greatly on you.
--
Joseph E. Meehan
26 + 6 = 1 It's Irish Math
Mxsmanic
> What would be the most drastic difference in shooting digital vs.
> shooting film?
Speed and convenience.
> Is there any aspect that the thinking or approach would
> be different?
Not really. The effect of digital vs. film has been tremendously
exaggerated. About 99.99% of photography does not change.
> It seems aperture and DOF would be the same, for
> example, but white balance vs film type would be approached
> differently.
The differences still are not that great. With film you do your white
balance in Photoshop; with digital the camera tries to do it (not always
successfully, so you may still have to do it in Photoshop).
> In film, not all 100 ISO films are the same, and white
> balance seems to be the digital variable of what type of film one is
> using, if that makes sense.
Except that digital white balance doesn't actually change the image
sensor, whereas changing films does. This has consequences for things
like image noise.
--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.
Mxsmanic
> film = 7-9 stop range
Depends on the film.
> digital = 4 stops
Depends on the sensor and hardware/software behind it. This figure is
probably accurate for most digicams on the market right now. It's
higher for pro digicams, and much higher for specially built and cooled
image sensors (not practical for use in still cameras, though).
Flycaster
news:Lzb%b.3820$Yf....@twister01.bloor.is.net.cable.rogers.com...
> film = 7-9 stop range
> digital = 4 stops
Oh boy, here we go...
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----
bmoag
Don't try to judge if focus is correct by the teensy LCD image.
With electronic viewfinder cameras you really have to depend on autofocus.
Images will look "flatter" than film: I call it the Jan van Eyck effect. It
is not visible with all subjects and possibly occurs because all the pixels
are in the same plane whereas with film, especially color, the imaging
surface has depth and irregularity.
David J. Littleboy
"k" <as...@asdf.com> wrote:
> film = 7-9 stop range
That's _negative_ films. Color slide films are 5 stops. And most negative
films are grainy enough that they look a lot worse than dSLR digital. It
takes slide film and good technique to complete with dSLR digital.
> digital = 4 stops
The dSLRs are all well over 5 stops, and the 10D at ISO 100 in RAW mode
should be well over 6.
David J. Littleboy
Tokyo, Japan
David J. Littleboy
"The DaveŠ" <n...@no.com> wrote in message
news:Ksb%b.968$Cf3...@news01.roc.ny...
> What would be the most drastic difference in shooting digital vs.
> shooting film?
Film is a gainy disaster at ISO 400 and over, and dSLR digital isn't. You
can make grain-free A4 prints from ISO 1600 shots right out of the camera
with the 10D or 300D (if you turn off in-camera sharpening and don't mind
soft prints).
GLC1173
>What would be the most drastic >difference in shooting digital vs.
>shooting film?
You can do a lot more to "clean up" imperfect digital photos than color film
ones.
~~~~~~~~~~~~~~~~~~~~~
<B>Dissident news - plus immigration, gun rights, nationwide weather
<I><A HREF="http://www.alamanceind.com">ALAMANCE INDEPENDENT:
official newspaper of the Vast Right Wing Conspiracy</A></b></i>
Lisa Horton
The Dave© wrote:
>
> What would be the most drastic difference in shooting digital vs.
> shooting film? Is there any aspect that the thinking or approach would
> be different? It seems aperture and DOF would be the same, for
> example, but white balance vs film type would be approached
> differently. In film, not all 100 ISO films are the same, and white
> balance seems to be the digital variable of what type of film one is
> using, if that makes sense.
Compared to shooting slide film, it's fairly similar. If anything, I
think digital may be slightly more unforgiving to overexposed
highlights.
While shutter speed and aperture would be the same, DOF is usually
different because of the APS sized sensor.
White balance functionally exactly like film color temp, except that
it's infinitely variable after the fact. It is as though you have a
roll of film in your camera that is all different color temps at once,
and can be any of a wide range of ISO values frame to frame.
Lisa
Mike Kohary
news:403D4FF9...@lisahorton.net...
>
> White balance functionally exactly like film color temp, except that
> it's infinitely variable after the fact. It is as though you have a
> roll of film in your camera that is all different color temps at once,
> and can be any of a wide range of ISO values frame to frame.
Which is really one of the most awesome features of digital versus film. :)
Mike
Jim Waggener
"k" <as...@asdf.com> wrote in message
news:Lzb%b.3820$Yf....@twister01.bloor.is.net.cable.rogers.com...
> digital = 4 stops
>
Mxsmanic
> That's _negative_ films. Color slide films are 5 stops.
No, slide films manage about the same as negative films, depending on
how you deal with them.
> And most negative films are grainy enough that they look
> a lot worse than dSLR digital.
That depends on how much you care about grain, how much you enlarge, and
how much you look for it.
Mxsmanic
> There is a great deal of difference over all.
There is virtually no difference.
Mxsmanic
> Film is a gainy disaster at ISO 400 and over, and dSLR digital isn't.
Portra 400BW looks fine at ISO 400; so does Portra 400UC.
Unfortunately, Kodak, in another of its neverending series of stupid
mistakes, has just discontinued both of these films.
> You can make grain-free A4 prints from ISO 1600 shots
> right out of the camera with the 10D or 300D (if you turn
> off in-camera sharpening and don't mind soft prints).
Now if only the colors were in register. Much of the image noise is
hidden by interpolation, at the expense of resolution. But if you like
silky smooth pastel colors, no problem.
Mxsmanic
> You can do a lot more to "clean up" imperfect digital photos than color film
> ones.
It's actually the other way around, since film captures more
information. With digital, what you see is what you get. With film,
there is often a lot more there than you can see.
Mxsmanic
> White balance functionally exactly like film color temp, except that
> it's infinitely variable after the fact.
Not infinitely.
> It is as though you have a
> roll of film in your camera that is all different color temps at once,
> and can be any of a wide range of ISO values frame to frame.
The results are inferior, though, because in digital, you are playing
with signal levels from a single sensor, whereas with film, you are
dealing with a different, custom image sensor for every purpose. In
other words, tungsten film really is balanced for tungsten light,
whereas a digicam's white balance will simply boost signal in the blue,
increasing image noise.
The best way to compensate for color casts with both digital and film is
to use filters at the time the photo is shot, but that does reduce
sensitivity to light.
J...@no.komm
"David J. Littleboy" <dav...@gol.com> wrote:
>The dSLRs are all well over 5 stops, and the 10D at ISO 100 in RAW mode
>should be well over 6.
What are we talking about here, the range of exposure compensation
possible for an average contrast scene, or the difference between the
darkest usable shadows and the brightest highlights in a single
exposure?
--
<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <J...@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Mxsmanic
> Which is really one of the most awesome features of
> digital versus film.
It's very deceptive, so beware.
Jim Waggener
"> The best way to compensate for color casts with both digital and film is
> to use filters at the time the photo is shot, but that does reduce
> sensitivity to light.
>
why bother with filters. Photoshop can easily replicate any filter.
Lisa Horton
Mxsmanic wrote:
>
> Lisa Horton writes:
>
> > White balance functionally exactly like film color temp, except that
> > it's infinitely variable after the fact.
>
> Not infinitely.
True, but functionally so close as to be a moot point, IMO.
>
> > It is as though you have a
> > roll of film in your camera that is all different color temps at once,
> > and can be any of a wide range of ISO values frame to frame.
>
> The results are inferior, though, because in digital, you are playing
> with signal levels from a single sensor, whereas with film, you are
> dealing with a different, custom image sensor for every purpose. In
> other words, tungsten film really is balanced for tungsten light,
> whereas a digicam's white balance will simply boost signal in the blue,
> increasing image noise.
True in theory to be sure, but in actual practice, I believe that for
all but the most demanding users/applications, the difference in this
area is negligible.
In both these cases, the probable small degradation of quality must be
balanced against the utility of having a wide choice at all times. Each
user must make their own determination of what is most important.
>
> The best way to compensate for color casts with both digital and film is
> to use filters at the time the photo is shot, but that does reduce
> sensitivity to light.
Agreed.
Which makes me wonder if CCD or CMOS sensors have a "native" white
balance...
Lisa
Mike Kohary
news:kjmq301tqo53n6fgb...@4ax.com...
> Mike Kohary writes:
>
> > Which is really one of the most awesome features of
> > digital versus film.
>
> It's very deceptive, so beware.
(On having multiple ISOs and color temperatures in "one roll of film", so to
speak...)
In what way?
Mike
John Navas
In <colq301jhagc56n6s...@4ax.com> on Thu, 26 Feb 2004 03:17:40
+0100, Mxsmanic <mxsm...@hotmail.com> wrote:
>David J. Littleboy writes:
>
>> That's _negative_ films. Color slide films are 5 stops.
>
>No, slide films manage about the same as negative films, depending on
>how you deal with them.
False, as I demonstrated to you in the past:
<http://www.fujifilm.com/JSP/fuji/epartners/bin/Provia100f.pdf>
Exposure latitude/range is generally taken as 0.1 above the minimum to 90
percent of maximum, which is -0.3 to -2.0 for this film, or a range of
1.7 ÷ 0.3 = 5 2/3 stops. That's actually generous, since in practice many
photographers use a much lower upper limit because of the high densities.
(Eight stops would be a range of 8 * 0.3 = 2.4) Negative film is more like
7-9 stops.
--
Best regards,
John Navas
[PLEASE NOTE: Ads belong *only* in rec.photo.marketplace.digital, as per
<http://bobatkins.photo.net/info/charter.htm> <http://rpdfaq.50megs.com/>]
John Navas
In <fqlq30hf0b7qtmg31...@4ax.com> on Thu, 26 Feb 2004 03:18:06
+0100, Mxsmanic <mxsm...@hotmail.com> wrote:
>Joseph Meehan writes:
>
>> There is a great deal of difference over all.
>
>There is virtually no difference.
Back to trolling?
Jim Waggener
> > The best way to compensate for color casts with both digital and film is
> > to use filters at the time the photo is shot, but that does reduce
> > sensitivity to light.
>
> Agreed.
>
> Which makes me wonder if CCD or CMOS sensors have a "native" white
> balance...
>
> Lisa
using a filter on a Dslr is absurd. A post process with PS will easily
mimic any filter.
John Navas
In <lrlq309ttunfue7rl...@4ax.com> on Thu, 26 Feb 2004 03:19:59
+0100, Mxsmanic <mxsm...@hotmail.com> wrote:
>Portra 400BW looks fine at ISO 400; so does Portra 400UC.
>
>Unfortunately, Kodak, in another of its neverending series of stupid
>mistakes, has just discontinued both of these films.
These "mistakes" might actually save the company.
J...@no.komm
Mxsmanic <mxsm...@hotmail.com> wrote:
>The best way to compensate for color casts with both digital and film is
>to use filters at the time the photo is shot, but that does reduce
>sensitivity to light.
You'd be better off ignoring white balance at the time of exposure, and
use light or filters that would get the red, green, and blue channels
all just below the clipping point for RAW data, and then white-balance
the RAW data after the shot. This would give the most detailed shadows
and the least noise in every channel.
John Navas
In <h0mq30h7p6le325rk...@4ax.com> on Thu, 26 Feb 2004 03:22:45
+0100, Mxsmanic <mxsm...@hotmail.com> wrote:
>Lisa Horton writes:
>> It is as though you have a
>> roll of film in your camera that is all different color temps at once,
>> and can be any of a wide range of ISO values frame to frame.
>
>The results are inferior, though, because in digital, you are playing
>with signal levels from a single sensor, whereas with film, you are
>dealing with a different, custom image sensor for every purpose. In
>other words, tungsten film really is balanced for tungsten light,
>whereas a digicam's white balance will simply boost signal in the blue,
>increasing image noise.
No, film is simply biased in order to compensate for lighting differences.
Digital is more flexible and accurate, because you're not stuck with one
imperfect color profile.
>The best way to compensate for color casts with both digital and film is
>to use filters at the time the photo is shot, but that does reduce
>sensitivity to light.
Film, yes; digital, no. Since a filter serves only to *reduce* light, it's at
best no better than digital post-processing, and worse in less than optimum
lighting.
J...@no.komm
Lisa Horton <Lisa...@lisahorton.net> wrote:
>Which makes me wonder if CCD or CMOS sensors have a "native" white
>balance...
The Canon 10D will record a grey card shot in RAW mode under 5500K light
as cyan, if the raw data is taken literally. It is about 0.6 stops less
sensitive to red than green, and about 0.1 stops less sensitive to green
than blue.
Mike Kohary
news:403d5...@corp.newsgroups.com...
>
> > > The best way to compensate for color casts with both digital and film
is
> > > to use filters at the time the photo is shot, but that does reduce
> > > sensitivity to light.
> >
> > Agreed.
> >
> > Which makes me wonder if CCD or CMOS sensors have a "native" white
> > balance...
>
> mimic any filter.
That's true, but it doesn't make use of a filter "absurd". Some people like
to pre-process, so to speak, as much as possible, and go for a "natural"
shot with a minimum of post processing. It just depends on your personal
taste...
Mike
Mxsmanic
> False, as I demonstrated to you in the past:
I've demonstrated it in my own photographs.
Mxsmanic
> These "mistakes" might actually save the company.
Since film is a major cash cow with fat margins, I don't really see how
it could save the company, although I can easily see how it might hurt
it.
David J. Littleboy
<J...@no.komm> wrote in message
news:d8mq30lds3q9d679v...@4ax.com...
> In message <c1ji66$6l2$1...@nnrp.gol.com>,
> "David J. Littleboy" <dav...@gol.com> wrote:
>
> >The dSLRs are all well over 5 stops, and the 10D at ISO 100 in RAW mode
> >should be well over 6.
>
> What are we talking about here, the range of exposure compensation
> possible for an average contrast scene, or the difference between the
> darkest usable shadows and the brightest highlights in a single
> exposure?
The latter.
David J. Littleboy
> David J. Littleboy writes:
>
> > That's _negative_ films. Color slide films are 5 stops.
>
> No, slide films manage about the same as negative films, depending on
> how you deal with them.
You are all by yourself in that opinion.
> > And most negative films are grainy enough that they look
> > a lot worse than dSLR digital.
>
> That depends on how much you care about grain, how much you enlarge, and
> how much you look for it.
Decent photographs don't show grain.
David J. Littleboy
"Mxsmanic" <mxsm...@hotmail.com> wrote:
> David J. Littleboy writes:
>
> > Film is a gainy disaster at ISO 400 and over, and dSLR digital isn't.
>
> Portra 400BW looks fine at ISO 400; so does Portra 400UC.
Your standards are abysmally low. The chromogenic films in particular are
amazingly bad. Barely adequate for 8x10s from 645, and then only if you fill
the frame. Seriously pitiful.
> > You can make grain-free A4 prints from ISO 1600 shots
> > right out of the camera with the 10D or 300D (if you turn
> > off in-camera sharpening and don't mind soft prints).
>
> Now if only the colors were in register. Much of the image noise is
> hidden by interpolation, at the expense of resolution. But if you like
> silky smooth pastel colors, no problem.
dSLR colors show _less_ hue shift than slide films. But that's at ISO 100.
Shooting at ISO 1600 involves compromises, of course. Still, the sample
shots in the dpreview Canon 300D gallery at ISO 1600 print as is at A4 with
no visible noise.
Mxsmanic
> In what way?
When you adjust color temperature or ISO on a digital camera, you're not
really changing color balance or sensitivity at all; you're just
instructing the camera to amplify one or more channels in a different
way.
The sensor does not become four times more sensitive to light when you
switch from 100 to 400; instead, software and hardware try to amplify
the signal from the sensor four times more than they did at 100. But
since it's the same sensor, this increases noise in the image. It's the
equivalent of shooting with ISO 100 film and then pushing two stops in
development, or trying to boost the shadows in a scan of the film.
This is quite different from using a selection of films. ISO 400 film
_really is_ four times more sensitive to light than ISO 100 film; no
extra development or scanning tricks are required.
The same situation obtains with color balance. Tungsten film _really
is_ more sensitive to blue light, so it records balanced colors in
tungsten lighting without any increase in noise. But a digital image
sensor "balanced" for tungsten light is just adding a lot of
amplification to the blue channel, which makes the light look more white
but also greatly increases noise in the image.
If digicams had interchangeable sensors, you could get the same results
you get by changing films. But with just one sensor, things like
changing ISO and changing white balance are just simulations, not the
real thing.
Mxsmanic
> why bother with filters. Photoshop can easily replicate any filter.
No, it cannot, for reasons I've explained in the past at length.
If you don't believe me, try reproducing the results from an infrared
filter in Photoshop.
Mxsmanic
> True, but functionally so close as to be a moot point, IMO.
Not when you are trying to compensate for low-temperature incandescent
lighting, low-pressure sodium vapor, or some other light sources.
> True in theory to be sure, but in actual practice, I believe that for
> all but the most demanding users/applications, the difference in this
> area is negligible.
I can see the extra noise in the blue channel. It's not negligeable,
I'm sorry to say. And the only cure would be interchangeable digital
sensors, but I'm not holding my breath over that.
> In both these cases, the probable small degradation of quality must be
> balanced against the utility of having a wide choice at all times. Each
> user must make their own determination of what is most important.
You have the same choice with film in post-production, and you'll get
the same results.
> Which makes me wonder if CCD or CMOS sensors have a "native" white
> balance...
They do. CCDs have a surprisingly linear response through the spectrum,
as I recall, but they are a bit weak at the blue end, and their response
extends too far into the infrared (which usually requires a filter to
block infrared in consequence). This native balance favors daylight,
because daylight is also fairly even across the spectrum. Incandescent
sources such as tungsten lamps are extremely red by comparison.
Mxsmanic
> using a filter on a Dslr is absurd. A post process with PS will easily
> mimic any filter.
Show me your Photoshop replication of polarizing filters.
Mxsmanic
> You'd be better off ignoring white balance at the time of exposure, and
> use light or filters that would get the red, green, and blue channels
> all just below the clipping point for RAW data, and then white-balance
> the RAW data after the shot. This would give the most detailed shadows
> and the least noise in every channel.
All you need to do is find a filter that can actually do that.
Mxsmanic
> No, film is simply biased in order to compensate for lighting
> differences.
The bias is built into the sensor, so it does not produce any additional
noise.
> Digital is more flexible and accurate, because you're not stuck with one
> imperfect color profile.
I'm afraid that's not true. Digital simulates, whereas film records.
A CCD, like a piece of a given film emulsion, has only one response
curve. Any other response curve can only be a simulation. Digital
sensors, like most films, work best in daylight. In unusual conditions,
weird adjustments must be made (such as cranking up the blue channel),
and noise and degradation result. A filter can avoid this, but it
reduces overall sensitivity substantially.
Note that professional video cameras typically use filters for white
balance, instead of just trying to tweak the output from the CCD. This
reduces sensitivity but produces a much cleaner image, with far less
noise.
> Film, yes; digital, no.
See above. Apparently video camera manufacturers disagree with you.
> Since a filter serves only to *reduce* light, it's at
> best no better than digital post-processing, and worse
> in less than optimum lighting.
Not so. A filter reduces light, but it produces a balance of light
reaching the image sensor that more closely matches the sensor's
sensitivity curve, and thus the output image requires less boosting of
weak channels, and contains less noise in consequence.
Mike Kohary
> Mike Kohary writes:
>
> > In what way?
>
> When you adjust color temperature or ISO on a digital camera, you're not
> really changing color balance or sensitivity at all; you're just
> instructing the camera to amplify one or more channels in a different
> way.
>
> The sensor does not become four times more sensitive to light when you
> switch from 100 to 400; instead, software and hardware try to amplify
> the signal from the sensor four times more than they did at 100. But
> since it's the same sensor, this increases noise in the image. It's the
> equivalent of shooting with ISO 100 film and then pushing two stops in
> development, or trying to boost the shadows in a scan of the film.
>
> This is quite different from using a selection of films. ISO 400 film
> _really is_ four times more sensitive to light than ISO 100 film; no
> extra development or scanning tricks are required.
Perhaps, but that's all background information, and it's the same result
either way to the user: the higher the ISO, the more noise is introduced.
I don't know about film advances, but in the digital world, there are
solutions to reduce or eliminate said noise.
How it all actually works is interesting, but kind of beside the point. A
digital camera effectively has the convenience of allowing all ISOs and
color balances (within the camera's feature set) from frame to frame, which
you can't do with film. That's a huge factor that pushed me over to digital
and off film altogether.
> If digicams had interchangeable sensors, you could get the same results
> you get by changing films. But with just one sensor, things like
> changing ISO and changing white balance are just simulations, not the
> real thing.
The results are what matter, and for all intents and purposes, those things
are the real thing when it comes to real-world application. That it
achieves those things differently than a film camera does isn't of any real
consequence, IMO, as long as those things work well.
Mike
Mxsmanic
> You are all by yourself in that opinion.
I'm not sure how you polled the rest of the population, but I tend to
disagree.
> Decent photographs don't show grain.
Why not? Some of them have pretty candy-pastel colors.
Mxsmanic
> Your standards are abysmally low. The chromogenic films in particular are
> amazingly bad.
They are amazingly good, at least Portra 400BW is. I use the latter
specifically because it has virtually no grain and yet is ISO 400.
Of course, in formats larger than 35mm, the results are even better.
> dSLR colors show _less_ hue shift than slide films.
Not a shift; just general inaccuracy, resulting from heavy interpolation
and matrix filters.
Mxsmanic
> Perhaps, but that's all background information, and it's the same result
> either way to the user: the higher the ISO, the more noise is introduced.
It depends on the film or sensor architecture, although that is true as
a general rule.
> I don't know about film advances, but in the digital world, there are
> solutions to reduce or eliminate said noise.
They aren't affordable or available for current digicams.
> How it all actually works is interesting, but kind of beside the point.
It's important to understand how it actually works, in order to
understand just what digital gives--and more importantly, what it does
_not_ give you.
> A digital camera effectively has the convenience of allowing all ISOs and
> color balances (within the camera's feature set) from frame to frame, which
> you can't do with film. That's a huge factor that pushed me over to digital
> and off film altogether.
It's not much of an issue for me. Most of my shots are in daylight,
with fairly consistent lighting. I can make any corrections required in
Photoshop. And for artificial light, very often serious corrections are
required, even for digital shots.
> The results are what matter, and for all intents and purposes, those things
> are the real thing when it comes to real-world application.
Perhaps for you, but not for me. I do see the difference.
Doug
pics as I feel like, to shoot pics of silly things( my big toe), to let my
kids shoot as many pics as they feel like. It is worry free, no film to buy,
to process, and processing fees, no changing film, no worrying about how
many shots are left on the roll of film ....in one word....Freedom.
Doug
"The DaveŠ" <n...@no.com> wrote in message
news:Ksb%b.968$Cf3...@news01.roc.ny...
> What would be the most drastic difference in shooting digital vs.
> shooting film? Is there any aspect that the thinking or approach would
> be different? It seems aperture and DOF would be the same, for
> example, but white balance vs film type would be approached
> differently. In film, not all 100 ISO films are the same, and white
> balance seems to be the digital variable of what type of film one is
> using, if that makes sense.
>
> --
> "Through pride we are ever deceiving ourselves. But deep down below the
> surface of the average conscience a still, small voice says to us,
> 'Something is out of tune.'" - - - Carl Gustav Jung
David J. Littleboy
> Mike Kohary writes:
>
> > In what way?
>
> When you adjust color temperature or ISO on a digital camera, you're not
> really changing color balance or sensitivity at all; you're just
> instructing the camera to amplify one or more channels in a different
> way.
>
> The sensor does not become four times more sensitive to light when you
> switch from 100 to 400;
Neither does film. When you switch film, you get grains that are four (or
whatever) times larger. 35mm ISO 400 films simply can't make a clean 8x10,
but the dSLRs have no trouble with A4s at ISO 400.
> instead, software and hardware try to amplify
> the signal from the sensor four times more than they did at 100. But
> since it's the same sensor, this increases noise in the image. It's the
> equivalent of shooting with ISO 100 film and then pushing two stops in
> development, or trying to boost the shadows in a scan of the film.
>
> This is quite different from using a selection of films. ISO 400 film
> _really is_ four times more sensitive to light than ISO 100 film; no
> extra development or scanning tricks are required.
But the grain noise is four (or whatever) times worse. In practical terms,
the 10D at ISO 400 is lower noise than the consumer digital cameras at their
lowest ISO, and produces image of essentially the same quality (for A4
prints) as at ISO 100. It's only at ISO 800 that noise begins to degrade
shadow detail significantly. Meanwhile, ISO 400 films in 35mm are a major
disaster at 8x10.
> The same situation obtains with color balance. Tungsten film _really
> is_ more sensitive to blue light, so it records balanced colors in
> tungsten lighting without any increase in noise. But a digital image
> sensor "balanced" for tungsten light is just adding a lot of
> amplification to the blue channel, which makes the light look more white
> but also greatly increases noise in the image.
Again, the dSLR noise at ISO 100 is so low as to be irrelevant, whatever you
do to the white balance. At ISO 800 and above, you don't have as much room
to play with.
> If digicams had interchangeable sensors, you could get the same results
> you get by changing films. But with just one sensor, things like
> changing ISO and changing white balance are just simulations, not the
> real thing.
Again, this ignores the far cleaner images that dSLRs produce, and so is
quite wrong in practice.
Mike Kohary
> Mike Kohary writes:
>
> > I don't know about film advances, but in the digital world, there are
> > solutions to reduce or eliminate said noise.
>
> They aren't affordable or available for current digicams.
My point being that they will be. That's the great thing about digital
photography - it's still in its infancy. The advances to come will be
mind-blowing.
> > How it all actually works is interesting, but kind of beside the point.
>
> It's important to understand how it actually works, in order to
> understand just what digital gives--and more importantly, what it does
> _not_ give you.
Point taken as a general rule, but to this specific example, it's academic.
> > A digital camera effectively has the convenience of allowing all ISOs
and
> > color balances (within the camera's feature set) from frame to frame,
which
> > you can't do with film. That's a huge factor that pushed me over to
digital
> > and off film altogether.
>
> It's not much of an issue for me. Most of my shots are in daylight,
> with fairly consistent lighting. I can make any corrections required in
> Photoshop. And for artificial light, very often serious corrections are
> required, even for digital shots.
Sure, everyone will have different requirements and desires. I shoot in a
wide range of lighting conditions, so interchangeable ISO is a big deal for
me.
> > The results are what matter, and for all intents and purposes, those
things
> > are the real thing when it comes to real-world application.
>
> Perhaps for you, but not for me. I do see the difference.
See the difference between what? Film grain and digital noise are easy to
tell apart, if that's what you mean, but that also seems beside the point.
Both are noise as a result of trying to capture more light from a
diminishing source. I certainly don't think one is more or less desireable
than the other - they're both flaws, about equally so. In a perfect world,
we'll eventually be able to take noiseless pictures in any light (and
introduce noise at our discretion rather than due to the limitations of
trying to capture low light).
Mike
GLC1173
>In what way?
Try doing some night photos with a digicam set on ISO 400. Then take it
home - and you'll find you can make the scene (within limits) look much lighter
or showing much more detail with Photoshop.
~~~~~~~~~~~~~~~~~~~~~
<B>Dissident news - plus immigration, gun rights, nationwide weather
<I><A HREF="http://www.alamanceind.com">ALAMANCE INDEPENDENT:
official newspaper of the Vast Right Wing Conspiracy</A></b></i>
PhotoMan
> Lisa Horton writes:
>
>> White balance functionally exactly like film color temp, except that
>> it's infinitely variable after the fact.
>
> Not infinitely.
>
>> It is as though you have a
>> roll of film in your camera that is all different color temps at
>> once, and can be any of a wide range of ISO values frame to frame.
>
> The results are inferior, though, because in digital, you are playing
> with signal levels from a single sensor, whereas with film, you are
> dealing with a different, custom image sensor for every purpose. In
> other words, tungsten film really is balanced for tungsten light,
> whereas a digicam's white balance will simply boost signal in the
> blue, increasing image noise.
>
> is to use filters at the time the photo is shot, but that does reduce
> sensitivity to light.
Actually, the light is reduced, not the film or sensor's sensitivity.
gsum
Film images do look more '3D' than digital images. I've done a lot
of computer graphics work in the past and have noticed that 3D (or at
least 2 1/2D effects) can be obtained by careful usage of background
colour. The most obvious of these is red on blue which causes red
objects to appear to float above the blue background. It might
just be the case that in landscapes, film tends to oversaturate the blues
(unless you use a filter). Desaturating blue has the effect of flattening
the image.
Graham
"bmoag" <ae...@yahoo.com> wrote in message
news:rUb%b.30372$9e1....@newssvr25.news.prodigy.com...
> Expose as you would for slide film.
> Don't try to judge if focus is correct by the teensy LCD image.
> With electronic viewfinder cameras you really have to depend on autofocus.
> Images will look "flatter" than film: I call it the Jan van Eyck effect.
It
> is not visible with all subjects and possibly occurs because all the
pixels
> are in the same plane whereas with film, especially color, the imaging
> surface has depth and irregularity.
>
>
gsum
Graham
"Mike Kohary" <sp...@be.gone> wrote in message
news:c1jqr0$eia$0...@pita.alt.net...
gsum
patronising and rude. Please think about the impression you're
giving before pressing the 'send' key.
Graham
"Mxsmanic" <mxsm...@hotmail.com> wrote in message
news:hkpq30lvm0vg9a3fs...@4ax.com...
Mxsmanic
> It might just be the case that in landscapes, film tends
> to oversaturate the blues ...
It is more likely that digital undersaturates it. Remember, current
digital cameras have very low color resolution compared to film (because
the digicams use only a single sensor with a matrix filter, giving them
only 1/3 of the required color information). I note that this often
gives digital photos a pastel-color look, with lots of creamy but
unsaturated tones. The best digicams compensate for this pretty well,
but there's no way to eliminate it completely as long as a single sensor
with a matrix filter and interpolation is being used.
Blue and red are the main losers in this arrangement. Only 1/4 of the
blue and red information in the original scene is captured by a digicam.
Mxsmanic
> My point being that they will be.
I don't expect to see actively-cooled sensors any time soon. Bigger
sensors help, but I don't expect to see those on any but the most
high-end cameras any time soon, either.
> That's the great thing about digital photography - it's
> still in its infancy. The advances to come will be
> mind-blowing.
Well, I'll wait for the mind-blowing advances before I switch. The
current state of digital is unsatisfactory for my purposes, and worse
yet, you pay a lot more at the same time--more money for less.
> Sure, everyone will have different requirements and desires. I shoot in a
> wide range of lighting conditions, so interchangeable ISO is a big deal for
> me.
You're really just skipping a step in Photoshop for the most part,
though. The camera boosts the curves instead of you, but there isn't
any less noise.
> See the difference between what?
Artificially boosted sensitivity and real differences in sensitivity.
When it is boosted artificially, the noise goes up. This is especially
obvious when you are trying to compensate for artificial lighting.
> In a perfect world, we'll eventually be able to take noiseless
> pictures in any light (and introduce noise at our discretion
> rather than due to the limitations of trying to capture low light).
If companies weren't so hellbent on killing off their film cash cows,
we'd have that for film right now (at least ten times more sensitivity
for equivalent noise--that's over three stops, and is pretty much the
theoretical limit).
Mxsmanic
> Neither does film. When you switch film, you get grains that are four (or
> whatever) times larger.
Today's ISO 800 films are comparable to yesterday's ISO 100 films.
Grain has been reduced enormously with no loss of sensitivity. Things
like tabular grains make a big difference.
> 35mm ISO 400 films simply can't make a clean 8x10 ...
Sure they can. In fact, I do 8x12 prints from ISO 800 all the time.
> But the grain noise is four (or whatever) times worse.
No. It's worse, but not that much worse. Even ISO 800 film is
surprisingly sharp and clear.
> Again, the dSLR noise at ISO 100 is so low as to be irrelevant, whatever you
> do to the white balance.
Show me some examples of noiseless, balanced DSLR shots under LPS light.
> At ISO 800 and above, you don't have as much room
> to play with.
You've already used it up to simulate ISO 800.
> Again, this ignores the far cleaner images that dSLRs produce, and so is
> quite wrong in practice.
Cleaner or duller? Digital has a pastel look that can be irritating at
times. Of course, having three sensors instead of one would fix that,
but nobody seems to care.
Mxsmanic
> Starting most of your senetences with the word 'no' is extremely
> patronising and rude. Please think about the impression you're
> giving before pressing the 'send' key.
Correction: You interpret it as patronizing and rude. Please consider
the potential inaccuracy of your inferences when reading a message.
Mxsmanic
> Actually, the light is reduced, not the film or sensor's sensitivity.
Sure, but the usable sensitivity is reduced--a major problem in
precisely the situations that often call for filters. Still, if you
have a tripod or lots of light, it's the best way to correct color casts
with no loss of image quality.
Mxsmanic
> To me , the biggest difference is freedom....the freedom to shoot as many
> pics as I feel like, to shoot pics of silly things( my big toe), to let my
> kids shoot as many pics as they feel like.
I presume that this excludes the hours required to sort and archive all
those hundreds or thousands of digital pictures, right?
I have digital pictures that are several months old and still unedited
because it is such a pain to go through such large numbers of shots in
the hope of finding a good one.
> It is worry free, no film to buy,
> to process, and processing fees, no changing film, no worrying about how
> many shots are left on the roll of film ....in one word....Freedom.
It depends on your purposes. I've never considered the things you
mention as restrictions. For me, the most annoying restriction in
photography is a lack of light, and digital does nothing at all to help
in that domain.
David J. Littleboy
> Doug writes:
>
> > To me , the biggest difference is freedom....the freedom to shoot as
many
> > pics as I feel like, to shoot pics of silly things( my big toe), to let
my
> > kids shoot as many pics as they feel like.
>
> I presume that this excludes the hours required to sort and archive all
> those hundreds or thousands of digital pictures, right?
It's a lot faster than archiving scans, since most folks who scan think that
they need to save tiffs.
> I have digital pictures that are several months old and still unedited
> because it is such a pain to go through such large numbers of shots in
> the hope of finding a good one.
But at least you can see them if you wanted to. I'm still only half way
through scanning my Jan 1 to Jan 5 trip shots. I really don't feel like
shooting any more until I've caught up with the scanning, so my cameras are
gathering dust.
> > It is worry free, no film to buy,
> > to process, and processing fees, no changing film, no worrying about how
> > many shots are left on the roll of film ....in one word....Freedom.
>
> It depends on your purposes. I've never considered the things you
> mention as restrictions. For me, the most annoying restriction in
> photography is a lack of light, and digital does nothing at all to help
> in that domain.
Wrong again. Digital is much better than film at ISO 400 and above, making
low-light work possible where it wasn't before.
David J. Littleboy
> David J. Littleboy writes:
>
> > Neither does film. When you switch film, you get grains that are four
(or
> > whatever) times larger.
>
> Today's ISO 800 films are comparable to yesterday's ISO 100 films.
You are dreaming. Today's ISO 100 films are quite acceptable, but ISO 400
and higher films are not. dSLRs are a lot better than modern films at ISO
400 and higher.
> Grain has been reduced enormously with no loss of sensitivity. Things
> like tabular grains make a big difference.
>
> > 35mm ISO 400 films simply can't make a clean 8x10 ...
>
> Sure they can. In fact, I do 8x12 prints from ISO 800 all the time.
You have much lower standards than I, then. From 6x6, ISO 400 and 800 films
make decent 8x10s, but not from 35mm.
> > But the grain noise is four (or whatever) times worse.
>
> No. It's worse, but not that much worse. Even ISO 800 film is
> surprisingly sharp and clear.
You must be shooting 6x9. The ISO 800 stuff is a major disaster compared to
Reala, and even Reala is too grainy for my preferences.
> > Again, this ignores the far cleaner images that dSLRs produce, and so is
> > quite wrong in practice.
>
> Cleaner or duller? Digital has a pastel look that can be irritating at
> times. Of course, having three sensors instead of one would fix that,
> but nobody seems to care.
No, as I mentioned in my other note, one or three sensors per site doesn't
have anything whatsoever to do with color rendition.
David J. Littleboy
"Mxsmanic" <mxsm...@hotmail.com> wrote:
> gsum writes:
>
> > It might just be the case that in landscapes, film tends
> > to oversaturate the blues ...
>
> It is more likely that digital undersaturates it. Remember, current
> digital cameras have very low color resolution compared to film (because
> the digicams use only a single sensor with a matrix filter, giving them
> only 1/3 of the required color information).
You are confusing resolution with color reproduction. Those are completely
different things. The rendition of color is completely independent of the
resolution of detail in color patterns. Taking 3 measurements per site
doesn't change the way colors are rendered, only how fast you can change
color.
> I note that this often
> gives digital photos a pastel-color look, with lots of creamy but
> unsaturated tones. The best digicams compensate for this pretty well,
It's not the camera, but the users. Most people don't understand color.
All the dSLRs do color very well in the hands of a competent user. (Well,
there is one exception<g>.)
> but there's no way to eliminate it completely as long as a single sensor
> with a matrix filter and interpolation is being used.
>
> Blue and red are the main losers in this arrangement. Only 1/4 of the
> blue and red information in the original scene is captured by a digicam.
Again, you are simply wrong here. Color rendition is independent of
resolution.
Mxsmanic
> You are confusing resolution with color reproduction.
They are not distinct in practice. Poor color resolution leads to poor
color reproduction.
> Those are completely different things.
Only if you are shooting a blank wall. If there is any detail at all in
the image, they interact.
> The rendition of color is completely independent of the
> resolution of detail in color patterns.
No, it's not. Each pixel with a single sensor and matrix filter is
receiving information for only one primary color; the other two colors
are missing. The interpolation required to simulate full color
introduces a multitude of rendering errors that damage color
reproduction and resolution.
> Taking 3 measurements per site doesn't change the way colors
> are rendered, only how fast you can change color.
Same thing. There's a reason why video cameras use three CCDs, instead
of one. I suspect that once camera manufacturers figure out how to use
three CCDs in a single still camera, all of a sudden they'll be crowing
about how important it is; in the meantime, with no camera that has
this, they take care to create the impression that it results in no
improvement. But in fact, three sensors produce a _huge_ improvement.
> It's not the camera, but the users.
No, it's the camera. Users have no control over this.
> Most people don't understand color.
I agree.
> All the dSLRs do color very well in the hands of a competent user.
None of the DSLRs that I've seen are especially good at it. Certainly
no better than film. They _can't_ be, because they don't capture enough
information.
> Again, you are simply wrong here. Color rendition is independent of
> resolution.
Look back at your posts ten years from now, when 3-CCD still cameras are
common. I'll be sure to ask why you said that one sensor was sufficient
in the olden days.
Mxsmanic
> You are dreaming. Today's ISO 100 films are quite acceptable, but ISO 400
> and higher films are not.
Then why is so much film above ISO 100 being sold? So people can throw
it away because it is "unacceptable"?
> You have much lower standards than I, then.
I don't share your preoccupation with grain, apparently.
> From 6x6, ISO 400 and 800 films make decent 8x10s, but not from 35mm.
I've had no complaints.
> You must be shooting 6x9.
No, this is 35mm.
> The ISO 800 stuff is a major disaster compared to
> Reala, and even Reala is too grainy for my preferences.
Your preferences sound unusual.
> No, as I mentioned in my other note, one or three sensors per site doesn't
> have anything whatsoever to do with color rendition.
It has everything to do with it, and the use of a single sensor in
digital cameras is one of the things that makes them inferior to film.
I don't expect that any digital photography advocates will admit this
until cameras become available with three sensors, though.
Mxsmanic
> It's a lot faster than archiving scans, since most folks who scan think that
> they need to save tiffs.
I save JPEGs. More importantly, though, if I shoot slides, I can pick
the ones I want to scan in a few seconds. With digital photos, I have
to go through them all on the screen of my computer. Indeed, slides are
already in archived form; I don't have to scan them to archive them.
> But at least you can see them if you wanted to.
Ninety-nine out of a hundred aren't worth seeing, which is why I'm so
slow to trudge through them. Like most people, I tend to shoot a lot
more images with digital, but the number of keepers doesn't really
change.
> Wrong again. Digital is much better than film at ISO 400 and above, making
> low-light work possible where it wasn't before.
It has less grain, but poorer resolution and color. I want resolution
and color.
gsum
Graham
"Mxsmanic" <mxsm...@hotmail.com> wrote in message
news:gscr305ls5h5vp8el...@4ax.com...
Chris Brown
>In message <403D5C8B...@lisahorton.net>,
>Lisa Horton <Lisa...@lisahorton.net> wrote:
>
>>Which makes me wonder if CCD or CMOS sensors have a "native" white
>>balance...
>
>The Canon 10D will record a grey card shot in RAW mode under 5500K light
>as cyan, if the raw data is taken literally. It is about 0.6 stops less
>sensitive to red than green, and about 0.1 stops less sensitive to green
>than blue.
This is very handy if you wnat to rescue blown highlights from night
cityscape shots from that nasty, omnipresent sodium light.
Chris Brown
Mxsmanic <mxsm...@hotmail.com> wrote:
>I'm afraid that's not true. Digital simulates, whereas film records.
Yeah, but make sure you use that single-crystal, oxygen free film, and make
sure your DX coding is done using gold leaf.
David J. Littleboy
"Mxsmanic" <mxsm...@hotmail.com> wrote:
> David J. Littleboy writes:
>
> > It's a lot faster than archiving scans, since most folks who scan think
that
> > they need to save tiffs.
>
> I save JPEGs.
Me too<g>. One of the few things we agree on.
> More importantly, though, if I shoot slides, I can pick
> the ones I want to scan in a few seconds.
Not me. If I went to the effort of exposing a medium format frame, I want it
scanned so I can learn from it. I don't shoot MF willy-nilly.
> With digital photos, I have
> to go through them all on the screen of my computer. Indeed, slides are
> already in archived form; I don't have to scan them to archive them.
Find a program like Qimage. It creates thumbnails and displays them
instantly as soon as you switch to the folder. Much easier than looking
through a binder. Besides, the slides aren't "archived" unless they're in a
fireproof safe, and getting slides out of a fireproof safe is a lot slower
than clicking through a directory tree.
> > But at least you can see them if you wanted to.
>
> Ninety-nine out of a hundred aren't worth seeing, which is why I'm so
> slow to trudge through them. Like most people, I tend to shoot a lot
> more images with digital, but the number of keepers doesn't really
> change.
The delete key is pretty quick. Learn to use it. It's your friend.
> > Wrong again. Digital is much better than film at ISO 400 and above,
making
> > low-light work possible where it wasn't before.
>
> It has less grain, but poorer resolution and color. I want resolution
> and color.
If you're using a recent dSLR, it's got better resolution and color than
fast films. It's only at ISO 100 that 35mm is competitive with dSLR digital
in image quality. If you are using a consumer camera, then you are being
silly if you are making generalizations from it. Consumer cameras are barely
acceptable at their lowest ISO.
David J. Littleboy
dav...@gol.com
Tokyo, Japan
Mxsmanic
> No, It is patronising and rude. See what I mean?
No.
Mxsmanic
> Not me. If I went to the effort of exposing a medium format frame, I want it
> scanned so I can learn from it. I don't shoot MF willy-nilly.
I was talking about 35mm. When I shoot MF slides, I still sort them,
though, and I don't scan them all (there are far fewer of them, but they
take a lot longer to scan).
> Find a program like Qimage. It creates thumbnails and displays them
> instantly as soon as you switch to the folder. Much easier than looking
> through a binder.
I just slide the slides around on a light table. I haven't found any
manipulation on a computer that's faster than that.
> The delete key is pretty quick. Learn to use it. It's your friend.
Not when the images are large and have to load. Plus they have to be
transferred from a CF card. It's all very cumbersome.
> If you're using a recent dSLR, it's got better resolution and color than
> fast films.
No, it just has less grain, which isn't the same thing.
Joseph Meehan
> Joseph Meehan writes:
>
>> There is a great deal of difference over all.
>
> There is virtually no difference.
For some that would be true. The more precise the photographer them
greater the difference. For someone who takes just family snapshots are
birthdays, almost none. For someone who is carefully chooses their film and
exposure for each exposure, it would be great.
--
Joseph E. Meehan
26 + 6 = 1 It's Irish Math
gsum
of both worlds. With digital I'm very experimental and occasionally
get some great shots by that method. With film I'm very careful
to go through process of composing the shots carefully as I don't want
to waste film due to the cost. The result is that I miss quite a lot
when using film.
Graham
"Doug" <dougt...@invalid.com> wrote in message
news:vVd%b.4291$Yf....@twister01.bloor.is.net.cable.rogers.com...
> To me , the biggest difference is freedom....the freedom to shoot as many
> pics as I feel like, to shoot pics of silly things( my big toe), to let my
> kids shoot as many pics as they feel like. It is worry free, no film to
buy,
> to process, and processing fees, no changing film, no worrying about how
> many shots are left on the roll of film ....in one word....Freedom.
>
> Doug
David J. Littleboy
> David J. Littleboy writes:
>
> > Find a program like Qimage. It creates thumbnails and displays them
> > instantly as soon as you switch to the folder. Much easier than looking
> > through a binder.
>
> I just slide the slides around on a light table. I haven't found any
> manipulation on a computer that's faster than that.
That only works for 20 or so slides. I can get to anything I've shot in the
last two years in 3 seconds.
(Athough it's getting to be time to get a new hard disk. Still, I could put
in 6 times the disk space I now have easily.)
> > The delete key is pretty quick. Learn to use it. It's your friend.
>
> Not when the images are large and have to load. Plus they have to be
> transferred from a CF card. It's all very cumbersome.
Compared to scanning film? You're being quite dizzy. dSLR files are still
tiny compared to scans.
> > If you're using a recent dSLR, it's got better resolution and color than
> > fast films.
>
> No, it just has less grain, which isn't the same thing.
No, the faster color films really don't capture the detail. And especially
the chromogenics are a mess. The 10D, at least, doesn't start dropping
significant detail until ISO 1600. The "3200" ISO films are barely ISO 1200,
and are a serious mess at 5x7.
gsum
The use off the word 'no' at the start of a sentence makes you seem
arrogant, pompous and unable or unwilling to consider the other person's
point of view. In addition, you are usually expressing a contridictory
opinion rather than stating a hard fact so the use of the black and
white 'no' is illogical.
Far more important though, is the way in which your posts generally
diminish the opinion that you are contradicting. That is why they your posts
are often patronising and rude and is the reason why many people get so
irritated with you.
By the way, don't try to shift the blame me for misunderstanding your
posts.You should try to express yourself in a clearer manner.
Graham
"Mxsmanic" <mxsm...@hotmail.com> wrote in message
news:t9jr30tnm86f9lv3n...@4ax.com...
k
unless its tungsten or some odd type.
"Mxsmanic" <mxsm...@hotmail.com> wrote in message
news:t7iq309le974gej0j...@4ax.com...
> The Dave© writes:
>
> > What would be the most drastic difference in shooting digital vs.
> > shooting film?
>
> Speed and convenience.
>
> > Is there any aspect that the thinking or approach would
> > be different?
>
> Not really. The effect of digital vs. film has been tremendously
> exaggerated. About 99.99% of photography does not change.
>
> > It seems aperture and DOF would be the same, for
> > example, but white balance vs film type would be approached
> > differently.
>
> The differences still are not that great. With film you do your white
> balance in Photoshop; with digital the camera tries to do it (not always
> successfully, so you may still have to do it in Photoshop).
>
> > In film, not all 100 ISO films are the same, and white
> > balance seems to be the digital variable of what type of film one is
> > using, if that makes sense.
>
> Except that digital white balance doesn't actually change the image
> sensor, whereas changing films does. This has consequences for things
> like image noise.
John Navas
wrote:
>John Navas writes:
>
>> These "mistakes" might actually save the company.
>
>Since film is a major cash cow with fat margins, ...
If those films were "a major cash cow with fat margins" then they
undoubtedly would not have been discontinued. The likely problem is
that demand was too low to make it economic to continue production.
The unit cost of producing film is highly sensitive to volume.
--
Best regards,
John Navas
[PLEASE NOTE: Ads belong *only* in rec.photo.marketplace.digital, as per
<http://bobatkins.photo.net/info/charter.htm> <http://rpdfaq.50megs.com/>]
Chris Brown
gsum <gs...@baesyst.com> wrote:
>
>Far more important though, is the way in which your posts generally
>diminish the opinion that you are contradicting. That is why they your posts
>are often patronising and rude and is the reason why many people get so
>irritated with you.
FWIW, I find that someone who has, shall we say, a somewhat prickly
communication style is probably worth listening to as long as they're
providing useful information and/or imparting the benefits of a large and
detailed knowledge.
After all, some of the worst dicators, murderers and other assorted scumbags
throughout history have almost certainly posessed impeccable manners and
decorum, and would never stoop to being *rude*. They may have been nasty
pieces of work, but they were *polite* nasty pieces of work. ;-)
On the other hand, someone who has an abrasive communication style, but who
isn't half as clever or knowledgable as they like to think they are is just
a bloody nuisance, and best ignored.
I wouldn't dream of suggesting which category any particular poster might be
placed in, of course. Killfiles are handy things, however. ;->
Larry Caldwell
The_Dave=A9?=) says...
> What would be the most drastic difference in shooting digital vs.
> shooting film? Is there any aspect that the thinking or approach would
> be different? It seems aperture and DOF would be the same, for
> example, but white balance vs film type would be approached
> differently. In film, not all 100 ISO films are the same, and white
> balance seems to be the digital variable of what type of film one is
> using, if that makes sense.
As long as the lighting is flat, digital sensors work fine. If there is
much contrast in the scene, you will lose major portions of the image.
Shooting print film you can just expose for the shadows and let the
highlights burn in. You will still pick up image detail even when
severely over-exposed, though you might have to do a lot of dodging to
get a usable print. Shooting digital is more like shooting slide film.
You have to expose for the highlights to avoid saturating the sensor,
then try to pick detail out of the noise in the shadows.
Use fill flash whenever your subject is backlit, and if you are in an
outdoor situation with a lot of light and shadow, switch to print film.
As with slide film, bias toward underexposure to minimize burning out
the highlights.
John Navas
In <20pq30damvgmse9qe...@4ax.com> on Thu, 26 Feb 2004 04:12:19
+0100, Mxsmanic <mxsm...@hotmail.com> wrote:
>John Navas writes:
>
>> False, as I demonstrated to you in the past:
>
>I've demonstrated it in my own photographs.
Whatever you say. The rest of us, however, are still subject to the
limitations of the films.
drhowarddrfinedrhoward
drhowarddrfinedrhoward
if all photos taken have a lot of grain. A properly exposed and developed
photo taken with the correct film will not show grain. Just like a properly
lit digital image will not show noise in the blacks.
John Navas
In <hkpq30lvm0vg9a3fs...@4ax.com> on Thu, 26 Feb 2004 04:26:18
+0100, Mxsmanic <mxsm...@hotmail.com> wrote:
>John Navas writes:
>
>> No, film is simply biased in order to compensate for lighting
>> differences.
>
>The bias is built into the sensor, so it does not produce any additional
>noise.
Meaningless -- film is still subject to sensitivity limitations, and the bias
is comparable to white balance adjustment in the digital realm.
>> Digital is more flexible and accurate, because you're not stuck with one
>> imperfect color profile.
>
>I'm afraid that's not true. Digital simulates, whereas film records.
No, both measure photon intensity, just in different ways.
>... Digital
>sensors, like most films, work best in daylight. In unusual conditions,
>weird adjustments must be made (such as cranking up the blue channel),
>and noise and degradation result. A filter can avoid this, but it
>reduces overall sensitivity substantially.
Indeed, but that's true of color temperature biasing in general -- whether
film or digital, sensitivity to certain colors is *reduced* to achieve proper
color balance. Sensitivity is only 'cranked up' to increase ISO.
>Note that professional video cameras typically use filters for white
>balance, instead of just trying to tweak the output from the CCD.
Which cameras? The (solid state) professional cameras I've used don't adjust
color temperature that way.
>This
>reduces sensitivity but produces a much cleaner image, with far less
>noise.
Other way around, since it reduces light to the sensor, which generally
decreases signal-to-noise ratio. Adjusting color temperature in the digital
domain is thus better, since it starts with a cleaner signal.
>> Film, yes; digital, no.
>
>See above. Apparently video camera manufacturers disagree with you.
Citation?
>> Since a filter serves only to *reduce* light, it's at
>> best no better than digital post-processing, and worse
>> in less than optimum lighting.
>
>Not so. A filter reduces light,
True, and that's the problem.
>but it produces a balance of light
>reaching the image sensor that more closely matches the sensor's
>sensitivity curve,
Not true.
>and thus the output image requires less boosting of
>weak channels, and contains less noise in consequence.
Also not true. Color temperature is balanced by reducing, not boosting,
certain colors. Boosting is a different matter, that of adjusting ISO, if
necessary.
k
lol
one of the features of film is grain.
"drhowarddrfinedrhoward" <drhowar...@sbcglobal.net> wrote in message
news:Hdp%b.2732$JI4...@newssvr22.news.prodigy.com...
John Navas
In <403da8fc$1...@baen1673807.greenlnk.net> on Thu, 26 Feb 2004 08:11:35 -0000,
"gsum" <gs...@baesyst.com> wrote:
>"Mxsmanic" <mxsm...@hotmail.com> wrote in message
>news:hkpq30lvm0vg9a3fs...@4ax.com...
>>[SNIP]
>Starting most of your senetences with the word 'no' is extremely
>patronising and rude. Please think about the impression you're
>giving before pressing the 'send' key.
From long experience with him, I'd say he is thinking about it -- he seems to
delight is stirring up flame wars by taking extreme, contrary positions. It's
a form of trolling, but at least it's on-topic, unlike the recent offensive
troll.
Alan Browne
> What would be the most drastic difference in shooting digital vs.
> shooting film? Is there any aspect that the thinking or approach would
> be different? It seems aperture and DOF would be the same, for
> example, but white balance vs film type would be approached
> differently. In film, not all 100 ISO films are the same, and white
> balance seems to be the digital variable of what type of film one is
> using, if that makes sense.
>
Drastic?
Digital most resembles slide film in exposure behavior, so attention to
highlights will get the shot in the exposure range of the sensor.
Digital has one stop or more to the dark end, so going all out on the
highlights will still get good shaddow detail, and allowing some burning
on the highlights will alllow even more shaddow detail.
As has been discussed, evaluation of the histogram and moving the image
to the 'high end' of the histo will permit finer tonal gradations in
that part of the image, eg: use all of the dynamic range of the image.
I'm not sure if this is 'drastic' but it is a huge change in controlling
the palcement of the scene in the exp latitude of the 'film'.
One thing a digital shooter does not have is the variety of palettes
that the film shooter has in film types... this, in any case is taken
care of in PS and is largely a non-issue.
Digital's low dynamic noise (at least as the eye can see it) with
increasing ISO is a huge benefit, as the ISO can go high without any
change to the grain size ... a pixel remains a pixel regardless of its
gain. ( a pj who presented at our club last year showed photos of
Masters Tennis he shot available light late in the evening that simply
would not have been possible with a film camera ...eg: he was shooting
at ISO 1000, 1600 and pulling magazine quality photos... )
Digital shooters are less likely to hesitate for a shot. The recurring
cost is 0, and the instant feedback on the shot puts the shooter in the
'quality' zone very quickly. We film purists thumb our noses at this
but the fact is that digital will either get people up the proficiency
curve very rapidly and/or alow them to shotgun their way to glory.
Anyway, nothing new or drastic in the stuff above....
Cheers,
Alan
--
e-meil: there's no such thing as a FreeLunch.
Alan Browne
> film = 7-9 stop range
> digital = 4 stops
Slide film covers 5 stops, (sometimes just 4 and bit, depending on the
film...)
Color Negative about 7 stops.
B&W negative about 8 stops.
For all of the above there is variance of about a stop depending on film
version.
digital covers 6 stops reasonably, and another 2 at the low end for
desperate recovery in PS.
If digital sensors go up to 16 bits/color on the A/D, then another stop
or so of useful information might get 'picked' out of the scene. This
will always be to the low side, and will be relatively noisy...eg: not
worth the expense...
Cheers,
Alan.
>
>
> "The Dave©" <n...@no.com> wrote in message
> news:Ksb%b.968$Cf3...@news01.roc.ny...
>
>>What would be the most drastic difference in shooting digital vs.
>>shooting film? Is there any aspect that the thinking or approach would
>>be different? It seems aperture and DOF would be the same, for
>>example, but white balance vs film type would be approached
>>differently. In film, not all 100 ISO films are the same, and white
>>balance seems to be the digital variable of what type of film one is
>>using, if that makes sense.
>>
>>--
>>"Through pride we are ever deceiving ourselves. But deep down below the
>>surface of the average conscience a still, small voice says to us,
>>'Something is out of tune.'" - - - Carl Gustav Jung
Alan Browne
> "k" <as...@asdf.com> wrote in message
> news:Lzb%b.3820$Yf....@twister01.bloor.is.net.cable.rogers.com...
>
>>film = 7-9 stop range
>>digital = 4 stops
>>
>
> With the D100 it is -5 to +5
Doubt it, but I'm willing to be wrong, so quote source please.
Reasonably, the D100 (and 10D, and Drebel, etc., cover:
-3.5 to +2.5 ...6 stops (eg: all of zone II to all of zone VII)
might need to save RAW to conserve that lower stop
Desperately:
-5.5 to 2.5 ...8 stops, save RAW, and the image in the lower two
stops will be noisy and details will be hardly
discernable.
Cheers,
Alan
Alan Browne
> David J. Littleboy writes:
>
>
>>That's _negative_ films. Color slide films are 5 stops.
>
>
> No, slide films manage about the same as negative films, depending on
> how you deal with them.
Hogwash. Most negative film handilly covers 7 stops; slide barely
squeezes in 5 stops.
Alan Browne
> David J. Littleboy writes:
>
>
>>You are all by yourself in that opinion.
>
>
> I'm not sure how you polled the rest of the population, but I tend to
> disagree.
This is not about polling for opinion, it is about objective fact. See
my other reply for the numbers; see others reply if yor really learn by
"polling".
Cheers,
Alan
John Navas
In <7oer30lf1fi4vgpai...@4ax.com> on Thu, 26 Feb 2004 10:28:11
+0100, Mxsmanic <mxsm...@hotmail.com> wrote:
>David J. Littleboy writes:
>> Taking 3 measurements per site doesn't change the way colors
>> are rendered, only how fast you can change color.
>
>Same thing. There's a reason why video cameras use three CCDs, instead
>of one. I suspect that once camera manufacturers figure out how to use
>three CCDs in a single still camera, all of a sudden they'll be crowing
>about how important it is;
I doubt it, since single-ship Bayer sensors are already very good indeed, and
since digital video seems to be moving in the other direction (i.e., toward
single sensor). For example, see the new JVC JY-HD10U, a "pro" HD DV
camcorder based on a single chip. ("... gorgeous images with stunning detail.
Concerns about an HD camera with only a single CCD vanish when you see the
HD10's color reproduction. ... [Video Systems review])
>in the meantime, with no camera that has
>this, they take care to create the impression that it results in no
>improvement.
There are 3-chip cameras (digital camcorders) that can produce stills, but
they are blown away by better current Bayer-based digital cameras.
>But in fact, three sensors produce a _huge_ improvement.
With older technology, there was a significant difference, but technology
marches on.
John Navas
In <d2qq30l9epdh7em74...@4ax.com> on Thu, 26 Feb 2004 04:32:34
+0100, Mxsmanic <mxsm...@hotmail.com> wrote:
>David J. Littleboy writes:
>> dSLR colors show _less_ hue shift than slide films.
>
>Not a shift; just general inaccuracy, resulting from heavy interpolation
>and matrix filters.
Actually quite accurate, as shown by objective testing.
John Navas
In <42pq3097tdkh2ufes...@4ax.com> on Thu, 26 Feb 2004 04:17:14
+0100, Mxsmanic <mxsm...@hotmail.com> wrote:
>Mike Kohary writes:
>
>> In what way?
>
>When you adjust color temperature or ISO on a digital camera, you're not
>really changing color balance or sensitivity at all; you're just
>instructing the camera to amplify one or more channels in a different
>way.
You're actually instructing the camera to *reduce* certain colors to balance
the image. Amplification is only needed to increase ISO.
>The sensor does not become four times more sensitive to light when you
>switch from 100 to 400; instead, software and hardware try to amplify
>the signal from the sensor four times more than they did at 100. But
>since it's the same sensor, this increases noise in the image. It's the
>equivalent of shooting with ISO 100 film and then pushing two stops in
>development, ...
Comparable only in terms of objective; totally different in terms of results.
>This is quite different from using a selection of films. ISO 400 film
>_really is_ four times more sensitive to light than ISO 100 film; no
>extra development or scanning tricks are required.
The tricks are in the film itself, where such things as grain size are
sacrificed for the additional speed. TANSTAAFL. It's no different in the
digital domain, except that digital can actually minimize the downside by such
things as digital noise reduction techniques.
>The same situation obtains with color balance. Tungsten film _really
>is_ more sensitive to blue light, so it records balanced colors in
>tungsten lighting without any increase in noise.
Wrong. TANSTAAFL.
>But a digital image
>sensor "balanced" for tungsten light is just adding a lot of
>amplification to the blue channel, which makes the light look more white
>but also greatly increases noise in the image.
Also wrong. See above.
>If digicams had interchangeable sensors, you could get the same results
>you get by changing films.
Wrong again. The sensitivity of one color could only be increased by reducing
the sensitivity to other colors. TANSTAAFL.
>But with just one sensor, things like
>changing ISO and changing white balance are just simulations, not the
>real thing.
They are every bit as much the "real thing" as the comparable film methods.
John Navas
In <o6qq30l347f5vlt3k...@4ax.com> on Thu, 26 Feb 2004 04:35:17
+0100, Mxsmanic <mxsm...@hotmail.com> wrote:
>Mike Kohary writes:
>> I don't know about film advances, but in the digital world, there are
>> solutions to reduce or eliminate said noise.
>
>They aren't affordable or available for current digicams.
They are not only affordable but also widely available; e.g., Canon's noise
reduction on its CMOS sensors.
>> A digital camera effectively has the convenience of allowing all ISOs and
>> color balances (within the camera's feature set) from frame to frame, which
>> you can't do with film. That's a huge factor that pushed me over to digital
>> and off film altogether.
>
>It's not much of an issue for me. Most of my shots are in daylight,
>with fairly consistent lighting. I can make any corrections required in
>Photoshop. And for artificial light, very often serious corrections are
>required, even for digital shots.
Then you must not be taking to the time to set white balance correctly
(assuming you are using a decent camera).
John Navas
In <403da711$1...@baen1673807.greenlnk.net> on Thu, 26 Feb 2004 08:03:24 -0000,
"gsum" <gs...@baesyst.com> wrote:
>You can't win - the Maniac is *always* 100% correct.
Indeed. :)
Mxsmanic
> one of the features of film is grain.
For many slow films, the grain is so fine that it isn't visible.
Mxsmanic
> Hogwash. Most negative film handilly covers 7 stops; slide barely
> squeezes in 5 stops.
Slides hold more than that, but they concentrate most of their density
span in the midtones, for greater realism and more precise tonal
resolution in projection. So shadow and highlight detail are often
there, but noisier than with negative film.
Mxsmanic
> film is already white balanced.
Only for one color temperature, typically daylight.
Mxsmanic
> If those films were "a major cash cow with fat margins" then they
> undoubtedly would not have been discontinued.
True ... at any company except Kodak.
> The likely problem is that demand was too low to make
> it economic to continue production.
Then why were Portra 400BW and Portra 400UC replaced with new, different
emulsions?