Unnatural looking digital photos

Archived from groups: rec.photo.digital (More info?)

I have read a lot from film shooters that they don't like the
"unnatural look" that digital photos have and prefer the look of
film photos. After looking at a large number of digital and film
photos I have to assume that these people have gotten use to the way
film looks and have come to accept it as being the norm for what looks
natural.

So now I have to wonder, do these film shooters complain when they go
out in the real world that it looks unnatural. Does the real world
have that "unnatural digital look" to them. I have to wonder
because looking at digital photos and comparing them to film the
digital seem much closer to the real world then film.

Does the fact that the real world is missing that nice grain look
bother them, do the colors of the real world seem off to them, perhaps
flat and uninteresting?

Scott

Archived from groups: rec.photo.digital (More info?)

"Scott W" <biphoto@hotmail.com> wrote in message
news:1108933286.265301.81330@o13g2000cwo.googlegroups.com...
>I have read a lot from film shooters that they don't like the
> "unnatural look" that digital photos have and prefer the look of
> film photos. After looking at a large number of digital and film
> photos I have to assume that these people have gotten use to the way
> film looks and have come to accept it as being the norm for what looks
> natural.
>
> So now I have to wonder, do these film shooters complain when they go
> out in the real world that it looks unnatural. Does the real world
> have that "unnatural digital look" to them. I have to wonder
> because looking at digital photos and comparing them to film the
> digital seem much closer to the real world then film.
>
> Does the fact that the real world is missing that nice grain look
> bother them, do the colors of the real world seem off to them, perhaps
> flat and uninteresting?

One only captures a facsimile of the real world with any camera. Actually,
a photograph can look better than reality (even without severe retouching).
Then, there are scenes that photographs can only struggle with.

We are at the point where digital is better than film for most
photographers. Those who still use film are certainly justified in their
decision, however. I went to an art show last month with the works of five
photographers on display and every one of them captured in medium format
film and then went digital with a scanner. Their work was stunning (some of
it) and I'd guess they will stick with film for at least several more years.

Archived from groups: rec.photo.digital (More info?)

> Agree. Digital is linear, while film is not entirely linear. Therefore
> digital more accurately records reality

Better to say digital is quantized and film is continuous. A fine point,
perhaps, but worth mentioning.

Archived from groups: rec.photo.digital (More info?)

Thus spake Alfred Molon:
<Snipped>
> Agree. Digital is linear, while film is not entirely linear. Therefore
> digital more accurately records reality

Records reality? Are you sure any member of the human race is capable of
recognising reality even if bitten on the bum by it? Maybe after evolving
another million years Maybe by that time, we will also recognise that
"reality" is not a great deal more than what the majority thinks it is.

What do you think of the 8080 Alfred? I haven't handled one but the full
size pictures I've seen on the web were gobsmackingly good IMO: free from
CA/bloom artefacts, natural looking & sharp. I'm just growing older with my
C4040!

--
Thank people in advance? Thanking or cursing them afterwards at least
gives some feedback!

Archived from groups: rec.photo.digital (More info?)

"Scott W" <biphoto@hotmail.com> wrote in message
news:1108933286.265301.81330@o13g2000cwo.googlegroups.com...
> I have read a lot from film shooters that they don't like the
> "unnatural look" that digital photos have and prefer the look of
> film photos.

I suspect that in double blind tests that the vast majority of people saying
things like that could not actually tell the difference between digital and
silver photography. After all, their film is digitally processed, too. It is
all digital photography, now.

Archived from groups: rec.photo.digital (More info?)

"Scott W" <biphoto@hotmail.com> wrote:

> I have read a lot from film shooters that they don't like the
> "unnatural look" that digital photos have and prefer the look of
> film photos.

It's only _35mm_ shooters you hear that from. MF and LF photographers don't
make grainy prints, so we're used to clean prints. We complain about the
lack of resolution in digital, but other than that are quite happy with it.

David J. Littleboy
Tokyo, Japan

Archived from groups: rec.photo.digital (More info?)

Well, there are people swear by the difference between Leica and
Konica, and some can't tell the difference. There are also people swear
by the difference between the Midium format and the 35mm, and some
can't tell the difference. You might also have heard same story between
the Ilfochrome and the RA prints, and, blah blah blah.

It's certainly nothing to excited about if it happens between the
digital and film.



C J Campbell wrote:
> "Scott W" <biphoto@hotmail.com> wrote in message
> news:1108933286.265301.81330@o13g2000cwo.googlegroups.com...
> > I have read a lot from film shooters that they don't like the
> > "unnatural look" that digital photos have and prefer the look of
> > film photos.
>
> I suspect that in double blind tests that the vast majority of people
saying
> things like that could not actually tell the difference between
digital and
> silver photography. After all, their film is digitally processed,
too. It is
> all digital photography, now.

Archived from groups: rec.photo.digital (More info?)

Film has never been regarded as a continuous process. It is the original
digital process. A grain is 'turned on' by the incidence of light. Its
chemistry is altered such that it will decompose when washed with
developer. Grains not exposed to light do not decompose and are
dissolved by fixer. A grain is therefore the perfect example of a
bistate i.e. a bit. Film appears to bo continuous due to the large
number of grains.

Its always struck me as ironic that the film crowd who decry digital
don't realise that film is more digital than digital.

Graham


"Charles Schuler" <charleschuler@comcast.net> wrote in message
news:M8Kdnf6B6OIFgYTfRVn-sw@comcast.com...
>
> "Roland Karlsson" <roland_dot_karlsson@bonetmail.com> wrote in message
> news:Xns9603F2D622269klotjohan@130.133.1.4...
> > "Charles Schuler" <charleschuler@comcast.net> wrote in
> > news:r8KdnbitxuxzjYTfRVn-rQ@comcast.com:
> >
> >> Better to say digital is quantized and film is continuous. A fine
> >> point, perhaps, but worth mentioning.
> >
> > Both are quantizised. Grains are not continous and
> > so are not photons.
>
> Photons are indeed quantized units of electromagnetic radiation. Film
> chemistry and how it reacts to bombardment by light is widely/universally
> regarded as a continuous process. You might not realize that photons are
> merely a collapse of the wave function of light, and thus have little to
do
> with this discussion.
>
>

Archived from groups: rec.photo.digital (More info?)

Scott W Feb 20, 8:12 pm show options

Newsgroups: rec.photo.digital
From: "Scott W" <biph...@hotmail.com>
David J. Littleboy wrote:
> It's only _35mm_ shooters you hear that from. MF and LF photographers

don't
> make grainy prints, so we're used to clean prints. We complain about
the
> lack of resolution in digital, but other than that are quite happy
with it.

> David J. Littleboy
> Tokyo, Japan


Well what you said does tend to be true, the vast majority of film
photographer complaining about digital do seem to be shooting 35mm.

Scott

Yabbut the vast majority of film shooters shoot 35mm, so it only seems
logical they would have more complaints about the changeover.

In any case, it sounds like pretentious twaddle to me.

Archived from groups: rec.photo.digital (More info?)

"Scott W" <biphoto@hotmail.com> wrote in message
news:1108933286.265301.81330@o13g2000cwo.googlegroups.com...
> I have read a lot from film shooters that they don't like the
> "unnatural look" that digital photos have and prefer the look of
> film photos. After looking at a large number of digital and film
> photos I have to assume that these people have gotten use to the way
> film looks and have come to accept it as being the norm for what looks
> natural.
>
> So now I have to wonder, do these film shooters complain when they go
> out in the real world that it looks unnatural. Does the real world
> have that "unnatural digital look" to them. I have to wonder
> because looking at digital photos and comparing them to film the
> digital seem much closer to the real world then film.
>
> Does the fact that the real world is missing that nice grain look
> bother them, do the colors of the real world seem off to them, perhaps
> flat and uninteresting?
>
> Scott
>
Some people always resist change. Digital really poured on over the last two
or three years thanks to affordable dSLRs, and 3-5 MP cameras in the
$150-$300 range.

Some compact cameras are not the best way to judge digital images. Blooming
of highlights (limited dynamic range), over processed and sharpened look to
some images jpeg artifacts if compression is to high. dSLRs with their
larger sensors produce much more natural looking images. Dynamic range is
still limited, but not as much.
bg

Archived from groups: rec.photo.digital (More info?)

In article <1108985723.623439.248490@l41g2000cwc.googlegroups.com>,
charliediy@aol.com says...
> Well what you said does tend to be true, the vast majority of film
> photographer complaining about digital do seem to be shooting 35mm.
>
> Scott
>
> Yabbut the vast majority of film shooters shoot 35mm, so it only seems
> logical they would have more complaints about the changeover.
>
> In any case, it sounds like pretentious twaddle to me.
>
>


"Pretentious Twaddle"


I LOVE IT!

CAN I USE IT???

Seriously, I've been shooting film since 1959 (started with a 120 box
camera).

This "film is better" argument comes mostly from pixel peepers who think that
anything they can find in a photo under a magnifier is pertinant.

The issue SHOULD be the final product, but they cant get past their
predjudices to the print (or slide).

For years and years I would go through the ritual of changing film and/or
filters every time the light changed. Now its a simple matter of setting a
couple of buttons on the camera, or shooting RAW and making the decision
LATER.

Those advantages alone should drive a photographer to use digital for at
least SOME of his ((at least) outside the studio) work.

I'm not saying ABANDON FILM, I'm saying "If you dont familiarize yourself
with, and LEARN digital you will lose in the long run".





--
Larry Lynch
Mystic, Ct.

Archived from groups: rec.photo.digital (More info?)

"gsum" <gsum@bst.com> wrote in message
news:421992bf$1_1@baen1673807.greenlnk.net...
> Film has never been regarded as a continuous process. It is the original
> digital process. A grain is 'turned on' by the incidence of light. Its
> chemistry is altered such that it will decompose when washed with
> developer. Grains not exposed to light do not decompose and are
> dissolved by fixer. A grain is therefore the perfect example of a
> bistate i.e. a bit. Film appears to bo continuous due to the large
> number of grains.

I was afraid that this discussion was headed toward the metaphysical. A
process can be binary at its lowest level but function as a continuous
process at a practical level. The macro behavior of film is a continuous
process.

In a digital camera, the A to D converters "force" a digital constraint on
the information. Also, the fixed sensor site pitch produces spatial
aliasing, which is not to be found with film processes.

Archived from groups: rec.photo.digital (More info?)

Roland Karlsson wrote:

> "Charles Schuler" <charleschuler@comcast.net> wrote in
> news:r8KdnbitxuxzjYTfRVn-rQ@comcast.com:
>
>
>>Better to say digital is quantized and film is continuous. A fine
>>point, perhaps, but worth mentioning.
>
>
> Both are quantizised. Grains are not continous and
> so are not photons.
>
>
> /Roland
I would say it a little differently. Spatially, both are sampled data.

However, in a digital the data is truly quantized in a constant
increment step quantization.

Film density is a little different. The density has noise due to the
grain. So the math is a little different. In a sense, film at the
limit is somewhat like a one bit quantization (line drawing).

Archived from groups: rec.photo.digital (More info?)

Chris Brown <cpbrown@ntlworld.no_uce_please.com> wrote:
> Seriously - all changing the white balance in-camera is dooing is
> selective boosting of some colour channels. If you're shooting in
> tungsten light, for example, you will get a boost applied to the
> blue channel, which will be under-exposed compared to the red and
> green channels. This isn't really any different from shooting
> daylight film and boosting the blue yourself before you print.

> The problem with doing this is that, in the above scenario, the
> boosted channel will essentially be getting push-processed quite a
> bit, and that increases noise. For the best results with digital,
> you want to be using coloured filters, just like the "old-fashioned"
> way with film. That way, you expose the channels more evenly, and
> get a cleaner end result.

I can see the reasoning behind this but, but... use a filter, and you
lose light. So, if you happen to be working in a controlled
environment with plenty of light this might make sense, but otherwise
you'll just have to increase exposure, with all the usual problems.
For many people, most of the time, the problem of tungsten light is
accompained by the problem of low light. So, use a filter and you may
need to boost the ISO. In such circumstances you're better off
without the filter.

Andrew.

Archived from groups: rec.photo.digital (More info?)

In article <111k16ok6ti8770@news.supernews.com>,
<andrew29@littlepinkcloud.invalid> wrote:

[tungsten filters]

>I can see the reasoning behind this but, but... use a filter, and you
>lose light. So, if you happen to be working in a controlled
>environment with plenty of light this might make sense, but otherwise
>you'll just have to increase exposure,

Well, that depends on what you're doing - if you're using a nice, fast lens,
on a decent DSLR at 400 ISO, you would expect to still be in the range of
decent shutter speeds for handholding even under artificial light. You might
also be using a tripod, or an IS lens.

Of course, if using a filter does put you in a situation where the shutter
speed is pushed over the line, or you need an ISO boost, you may indeed be
better off taking the noise-hit from using white balance. The best course of
action depends on the conditions.

>with all the usual problems.
>For many people, most of the time, the problem of tungsten light is
>accompained by the problem of low light. So, use a filter and you may
>need to boost the ISO. In such circumstances you're better off
>without the filter.

Indeed - setting the white balance to tungsten is essentially boosting the
ISO for the blue channel, using a filter may require you to boost all 3.

However, if you do have sufficient light, or are using a tripod, and can
live with the slow shutter speed, using a filter will always give better
results.

Archived from groups: rec.photo.digital (More info?)

In message <4m7pe2-tvk.ln1@narcissus.dyndns.org>,
Chris Brown <cpbrown@ntlworld.no_uce_please.com> wrote:

>The problem with doing this is that, in th eabove scenario, the boosted
>channel will essentially be getting push-processed quite a bit, and that
>increases noise. For the best results with digital, you want to be using
>coloured filters, just like the "old-fashioned" way with film. That way, you
>expose the channels more evenly, and get a cleaner end result.

On my 20D, under incandescent light, the ratio of R:G:B in RAW, linear
values for a white card are about 90:100:25. An 80B filter changes that
to about 90:100:50.

The problem with incandescents and digitals is generally not too much
red, as many people think; it is too little blue. The incandescent does
have more red than green, but the bayer filters are generally less
sensitive for the red channel, evening the green and red out pretty
well.

What would be really nice would be a light source that allowed seperate
control of R, G, and B levels and thoroughly mixed them to avoid colored
shadows.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><

Archived from groups: rec.photo.digital (More info?)

In message <f8fpe2-nal.ln1@narcissus.dyndns.org>,
Chris Brown <cpbrown@ntlworld.no_uce_please.com> wrote:

>Indeed - setting the white balance to tungsten is essentially boosting the
>ISO for the blue channel, using a filter may require you to boost all 3.

The results of such a compromise depends on which end of the ISO range
you're at. Going from 100 to 200 or 400 is not as big a compromise as
going from 800 to 1600, or 400 to 1600. ISO 100 is generally very
compromised already (compared to its analog potential), as the
posterization floor is more significant than the sensor and readout
noise floor. The 12-bit ADC circuit wastes much of the analog level
detail captured in the sensor at ISO 100.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><

Archived from groups: rec.photo.digital (More info?)

In article <37sjhuF5d9k3oU1@individual.net>, Paul Busby says...

> What do you think of the 8080 Alfred? I haven't handled one but the full
> size pictures I've seen on the web were gobsmackingly good IMO: free from
> CA/bloom artefacts, natural looking & sharp. I'm just growing older with my
> C4040!

The 8080 has a very good lens, much better than the one of the 4040 or
5050 (which both have the same lens). Chromatic aberrations are low, and
the lens is sharp. The lens of the 8080 is more capable of supporting
8MP than the lens of the 5050 is capable of supporting 5MP (this is the
result of a test I made - I currently have both cameras). By the way,
the China photo gallery is entirely shot with an 8080:
http://www.molon.de/galleries/China/
--

Alfred Molon
------------------------------
Olympus 4040, 5050, 5060, 7070, 8080, E300 forum at
http://groups.yahoo.com/group/MyOlympus/
Olympus 8080 resource - http://myolympus.org/8080/

Archived from groups: rec.photo.digital (More info?)

In message <Xns9604E3B7FA72Fklotjohan@130.133.1.4>,
Roland Karlsson <roland_dot_karlsson@bonetmail.com> wrote:

>JPS@no.komm wrote in news:rh9k11h304j2b09ltrp172quebikgvu2ro@4ax.com:

>> This will be true until the ADCs used go higher than 12-bit. Most DSLRs
>> at ISO 100 are limited by quantization, not noise. If they were limited
>> by noise, ISO 1600 on these cameras would be unusable.

>Yepp, sounds reasonable. It could be though that a pre amplification,
>before doing the A/D conversion, might limit the noise. But ... you
>are probably right. More bits and under exposure would then be the
>same as pushing to a higher ISO.

That's where all the dynamic range and exposure lattitude that many
people are complaining about lies; it doesn't matter one bit how
noiseless a sensor is, if it is only digitizing at 12 bits.


--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><

Archived from groups: rec.photo.digital (More info?)

"Charles Schuler" <charleschuler@comcast.net> wrote in message
newsaidnXbZwIUbbITfRVn-vg@comcast.com...
>
> "gsum" <gsum@bst.com> wrote in message
> news:421992bf$1_1@baen1673807.greenlnk.net...
> > Film has never been regarded as a continuous process. It is the original
> > digital process. A grain is 'turned on' by the incidence of light. Its
> > chemistry is altered such that it will decompose when washed with
> > developer. Grains not exposed to light do not decompose and are
> > dissolved by fixer. A grain is therefore the perfect example of a
> > bistate i.e. a bit. Film appears to bo continuous due to the large
> > number of grains.
>
> I was afraid that this discussion was headed toward the metaphysical. A
> process can be binary at its lowest level but function as a continuous
> process at a practical level. The macro behavior of film is a continuous
> process.

And to add insult to injury, all continuous processes with noise (i.e., all
continuous processes) are only capable of representing a discrete number of
levels, i.e. the range of values divided by the noise. (Look up "dynamic
range" in an engineering handbook.)

So all analog processes are actually discrete, i.e. digital.

> In a digital camera, the A to D converters "force" a digital constraint on
> the information.

As long as the resolution of the A/D converter exceeds what is required by
the dynamic range of the signal, nothing is imposed at all. The apparently
digital result contains all the information of the nominally analog source
signal. (As long as you interpret the digital value to be specifying a
range, not an exact value.)

> Also, the fixed sensor site pitch produces spatial
> aliasing, which is not to be found with film processes.

Well, not if there's an antialiasing filter. In which case, the signal is
merely bandlimited. Note that spatial aliasing is _artifacting_. That is,
the snap-to-grid eye candy the Foveon fans are so fond of are in reality
incorrect, artifact-ridden images.

David J. Littleboy
Tokyo, Japan

Archived from groups: rec.photo.digital (More info?)

On Mon, 21 Feb 2005 17:45:23 -0500, "Charles Schuler"
<charleschuler@comcast.net> wrote:

>
>
>> At the quantum physical level all matter
>> is "digital". Physical chemistry and
>> semiconductors are based on that, so it's
>> hardly an abstract theory.
>
>Richard Feynman said, in essence, that "Nobody understands quantum
>mechanics."


And who would I be to argue with Richard Feynman?


rafe b.
http://www.terrapinphoto.com

Archived from groups: rec.photo.digital (More info?)

<JPS@no.komm> wrote:
>
> The problem with incandescents and digitals is generally not too much
> red, as many people think; it is too little blue. The incandescent does
> have more red than green, but the bayer filters are generally less
> sensitive for the red channel, evening the green and red out pretty
> well.
>
> What would be really nice would be a light source that allowed seperate
> control of R, G, and B levels and thoroughly mixed them to avoid colored
> shadows.

What would be nice would be a camera with interchangeable sensors and a
monochrome sensor option. That would almost double the sensitivity (collects
all the R+G at every sensor). Also, you could do B&W landscapes with a deep
red filter with no loss of resolution, and you probably wouldn't need the IR
cut filter, so you could use the camera for IR as well.

What might be neat would be an 8MP monochrome sensor in a full-frame
switchable with the regular 16.7MP Bayer sensor.

David J. Littleboy
djl@Dreaming.on.com
Tokyo, Japan

Archived from groups: rec.photo.digital (More info?)

J...@no.komm wrote:

> That's where all the dynamic range and exposure lattitude that many
> people are complaining about lies; it doesn't matter one bit how
> noiseless a sensor is, if it is only digitizing at 12 bits.

Well 12 bits does give you 12 stops of dynamic range, if the noise is
low enough. And you can go down 7 stops and the level to level change
is only on the order of 3% not bad at all. So just how is this so
limiting?

Scott

Archived from groups: rec.photo.digital (More info?)

In message <1109035246.989000.201380@o13g2000cwo.googlegroups.com>,
"Scott W" <biphoto@hotmail.com> wrote:

>J...@no.komm wrote:
>
>> That's where all the dynamic range and exposure lattitude that many
>> people are complaining about lies; it doesn't matter one bit how
>> noiseless a sensor is, if it is only digitizing at 12 bits.
>
>Well 12 bits does give you 12 stops of dynamic range, if the noise is
>low enough. And you can go down 7 stops and the level to level change
>is only on the order of 3% not bad at all.

Go down 7 stops from where?

You need to leave some headroom, and you have a range used in the
subject(s). Going down 7 stops with those things considered is all the
way down in posterization city.

>So just how is this so limiting?

It is limiting in the sense that 12 bits is not enough to clearly
resolve what the sensor records at ISO 100. Here's an example; I put a
black guitar bag on the floor; and laid my "Canon Digital" neck strap on
top of it. I mounted the 20D on a tripod, pointed down at the strap,
covered it with a grey card, and used manual exposure at ISO 1600 to
center the exposure. I took away the grey card, and took an exposure.
I then left the aperture and shutter speed where they were, and changed
the ISO to 100 (this means the EXACT same exposure on the sensor, other
than noise issues). I took another shot. I put the same crop of the
two RAW (but interpolated within each channel) exposures together, after
using Levels in Photoshop to give them the same tonal curve:

http://www.pbase.com/jps_photo/image/40038800

The strap and bag are black; the green cast is because the color balance
is the native RAW one, under incandescent light.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><

Archived from groups: rec.photo.digital (More info?)

In article <UfudndcdXvOYFoffRVn-pw@comcast.com>,
John McWilliams <jpmcw@comcast.net> wrote:
>Chris Brown wrote:
>
>> The problem with doing this is that, in th eabove scenario, the boosted
>> channel will essentially be getting push-processed quite a bit, and that
>> increases noise. For the best results with digital, you want to be using
>> coloured filters, just like the "old-fashioned" way with film. That way, you
>> expose the channels more evenly, and get a cleaner end result.
>
>Do you have any published research that you can share in the NG that
>corroborates the above? It's intuitively correct to me, but I wonder if
>the difference is measurable.

I don't, but it shouldn't be difficult to try for anyone who wants to see if
it does make a useful difference.

Archived from groups: rec.photo.digital (More info?)

"David J. Littleboy" <davidjl@gol.com> writes:

>What might be neat would be an 8MP monochrome sensor in a full-frame
>switchable with the regular 16.7MP Bayer sensor.

Why two different resolutions? That means two different basic sensors,
and thus likely different drive voltages and clock frequencies, a pain
for the camera manufacturer to deal with. Why not use Bayer-filtered
and unfiltered versions of the same sensor, where almost everything will
remain the same?

Dave