Digital quality (vs 35mm): Any real answers?

Roland Karlsson wrote:
Toralf wrote in news:ce101c$sb4
:


Right... To a certain extent, I meant this particular argument as a
purely principal one, too, though, i.e. the question is not just

whether

it matters for photographs or not, but whether camera producers are
honest when they talk about e.g. 6MP. I just don't like being lied to,
even when it's a white lie, or a lie about something that doesn't
matter, if you know what I mean...



Hmmm ... those that do produce cameras based on Bayer sensors
have a problem here. The Bayer sensor is a sound principle and
it does have a potential for a resolution of 6 Mpixels if there
are 6 Msensels. But ... it is not the same as having 6 Million
full color sensels.
My point, exactly.
So ... what shall they do. Be so honest that
they give the impression that their cameras are worse than they
actually are.
Of course, the actual number doesn't really matter if you're only
comparing different cameras - as long as all producers use the same way
of counting. If you compare e.g. with scans, lower numbers would make
camera image quality sound worse - but that would be fair, really, since
6 million pixels from a scanner is clearly "better" in most cases than 6
million from a camera. But of course, it woudn't be easy to choose the
right number; 1.5M or 2M would probably be unfair to the cameras...

No ... it is not an easy task to choose what to
call their cameras. Furthermore, the only sensible output from
a 6 Msensel Bayer camera is 6 Mpixels (*).
I guess they might simply say that the camera has 6M sensor elements,
though... But of course, most people wouldn't know what they were
talking about... (But then again, do most people know what 6 million
pixels really means?)

If you choose more you
waste space and if you choose less you lose information.

/Roland



(*) ignoring the Fuiji 45 degree tilted sensors where the only
sensible output is twice as many pixels as sensels.

MXP wrote:


The films has not got documentation in english and is quite good.
10 pcs. was EUR 95 including shipping.....not to bad I think.....as each
film has a little bottle of
developer.

Max



Thanks,

Photoimpex is where I buy my Efke film (maybe not as fine-grained, but
certainly cheaper).

Chris

MXP wrote:


The films has not got documentation in english and is quite good.
10 pcs. was EUR 95 including shipping.....not to bad I think.....as each
film has a little bottle of
developer.

Max



Thanks,

Photoimpex is where I buy my Efke film (maybe not as fine-grained, but
certainly cheaper).

Chris

MXP wrote:


The films has not got documentation in english and is quite good.
10 pcs. was EUR 95 including shipping.....not to bad I think.....as each
film has a little bottle of
developer.

Max



Thanks,

Photoimpex is where I buy my Efke film (maybe not as fine-grained, but
certainly cheaper).

Chris

David Dyer-Bennet wrote:
Toralf writes:


David Dyer-Bennet wrote:

[ ... ]


You, and many other people when discussing these matters, seem to
forget that the storage space necessary for the digital cameras is not
free of charge, though (neither are the prints, even if you have your
own printer.)

The storage space costs something like $0.60 per 650 meg, for
archival-grade CDs.

I don't think I could get them *that* cheap around here and/or I
wouldn't trust the ones I do get at that price, but never you mind.


Or something like $100 for 120gig of hard drive.

Yep. Or $100 (2) for 25GB of tape.


I would *never* use tape for archival storage; it has a *very* short
life.
I think you are confusing actual media ("shelf") life with media
durability. Tapes are subject to wear and tear to a much higher degree
than e.g. CD, so their lifetime drops somewhat significantly if they are
used continually. For archival, purposes, however, tapes are usually
expected to last at least 30 years - and that's based on actual
experience, since tape technology is more that 50 years old.

[ ... ]


Slides, too. I hear a lot of professionals actually throw out a lot
of the slides they shoot, after picking out the keepers. (I've seen
articles on their using shredders to make sure nobody steals their
outtakes.)
He, he...


I have gotten some "coasters", immediate burn failures. But I've
never seen a verification error, ever.
Maybe you've been lucky, or perhaps we've been unlucky. Or could it be
that you haven't tested thoroughly enough? We use CDs a lot for data
exchange at work, and while you may say that most CDs written are just
fine, now and then we get a CD where the write operation seems to have
completed successfully, and when we inspect the filesystem everything
all files are there, and everything looks all right. Then if we compare
the contents of the files with the actual data written, we find that
there is a mismatch! Also, there are cases where the CD can be read just
fine on the CD-R(W) unit used to write it, but other readers have
problems with it - watch out for those.

And I've never had an old CD
fail yet, either audio CDs going back to 1983 or CD-Rs. (The failure
model for pressed CDs is actually very different than for CD-R; a
pressed CD can de-laminate, and some early ones did.)
A CD-R is laminated, too, isn't it?


[ ... ]


One CD per 36-picture film? So you're storing 18 megabytes per image?
My fine jpeg (most often used) original is about 2 megabytes. If I've
worked on the image I have a much bigger PSD file, but often the image
can be auto-processed into a screen image of about 40k, with no
intermediate kept. I average a LOT more than 36 pictures per CD.
JPEG destroys the data as a manner of course, and I've been brought up
to believe it's evil... If we assume pictures are kept, we should assume
the *actual pictures* are kept, not some degraded form of them. That's
the only fair way of comparing. But of course, the mere ability to
neither keep nor completely throw away the originals, but do something
in between, is of course in its own right a point in the favour of
digital cameras.


And of course, you'll need to purchase a certain amount of
microdrive/compactflash/SD media, to, and that's still rather
expensive - it's cost may actually be as much as 20 times the one of
film.


So far in 4 years with 2 digital cameras I've purchased 5 CF cards,
yes. And also an Iomega Fotoshow to transfer cards to zip disks in
the field.

Toralf writes:

David Dyer-Bennet wrote:
Toralf writes:

David Dyer-Bennet wrote:

[ ... ]


You, and many other people when discussing these matters, seem to
forget that the storage space necessary for the digital cameras is not
free of charge, though (neither are the prints, even if you have your
own printer.)

The storage space costs something like $0.60 per 650 meg, for
archival-grade CDs.

I don't think I could get them *that* cheap around here and/or I
wouldn't trust the ones I do get at that price, but never you mind.


Or something like $100 for 120gig of hard drive.

Yep. Or $100 (2) for 25GB of tape.
I would *never* use tape for archival storage; it has a *very* short
life.

I think you are confusing actual media ("shelf") life with media
durability. Tapes are subject to wear and tear to a much higher degree
than e.g. CD, so their lifetime drops somewhat significantly if they
are used continually. For archival, purposes, however, tapes are
usually expected to last at least 30 years - and that's based on
actual experience, since tape technology is more that 50 years old.

Last time I looked into tape life, archivists thought 10 years was
optimistic. They may, of course, have improved the oxide and binder
formulations since then.

I have gotten some "coasters", immediate burn failures. But I've
never seen a verification error, ever.

Maybe you've been lucky, or perhaps we've been unlucky. Or could it be
that you haven't tested thoroughly enough? We use CDs a lot for data
exchange at work, and while you may say that most CDs written are just
fine, now and then we get a CD where the write operation seems to have
completed successfully, and when we inspect the filesystem everything
all files are there, and everything looks all right. Then if we
compare the contents of the files with the actual data written, we
find that there is a mismatch! Also, there are cases where the CD can
be read just fine on the CD-R(W) unit used to write it, but other
readers have problems with it - watch out for those.

Always possible that I haven't tested thoroughly enough. I tend to
use data compare verifiction in the writing program, and then
periodically use CDR-diagnostic to review the general state of a CD
(mostly on a different drive from the original writer). I don't
really do this *systematically* enough to be sure I'm on top of
everything, though.

And I've never had an old CD
fail yet, either audio CDs going back to 1983 or CD-Rs. (The failure
model for pressed CDs is actually very different than for CD-R; a
pressed CD can de-laminate, and some early ones did.)

A CD-R is laminated, too, isn't it?

I don't believe so; at least not in anything like the same way.

[ ... ]

One CD per 36-picture film? So you're storing 18 megabytes per
image?
My fine jpeg (most often used) original is about 2 megabytes. If I've
worked on the image I have a much bigger PSD file, but often the image
can be auto-processed into a screen image of about 40k, with no
intermediate kept. I average a LOT more than 36 pictures per CD.
JPEG destroys the data as a manner of course, and I've been brought up
to believe it's evil... If we assume pictures are kept, we should
assume the *actual pictures* are kept, not some degraded form of
them. That's the only fair way of comparing. But of course, the mere
ability to neither keep nor completely throw away the originals, but
do something in between, is of course in its own right a point in the
favour of digital cameras.

The JPEG is the camera original. I'll shoot RAW files for exceptional
circumstances, but at 12MB per shot (and hence long write times, too)
I rarely use it.
--
David Dyer-Bennet, , http://www.dd-b.net/dd-b/
RKBA: http://noguns-nomoney.com/ http://www.dd-b.net/carry/
Pics: http://dd-b.lighthunters.net/ http://www.dd-b.net/dd-b/SnapshotAlbum/
Dragaera/Steven Brust: http://dragaera.info/

Toralf writes:

David Dyer-Bennet wrote:
Toralf writes:

David Dyer-Bennet wrote:

[ ... ]


You, and many other people when discussing these matters, seem to
forget that the storage space necessary for the digital cameras is not
free of charge, though (neither are the prints, even if you have your
own printer.)

The storage space costs something like $0.60 per 650 meg, for
archival-grade CDs.

I don't think I could get them *that* cheap around here and/or I
wouldn't trust the ones I do get at that price, but never you mind.


Or something like $100 for 120gig of hard drive.

Yep. Or $100 (2) for 25GB of tape.
I would *never* use tape for archival storage; it has a *very* short
life.

I think you are confusing actual media ("shelf") life with media
durability. Tapes are subject to wear and tear to a much higher degree
than e.g. CD, so their lifetime drops somewhat significantly if they
are used continually. For archival, purposes, however, tapes are
usually expected to last at least 30 years - and that's based on
actual experience, since tape technology is more that 50 years old.

Last time I looked into tape life, archivists thought 10 years was
optimistic. They may, of course, have improved the oxide and binder
formulations since then.

I have gotten some "coasters", immediate burn failures. But I've
never seen a verification error, ever.

Maybe you've been lucky, or perhaps we've been unlucky. Or could it be
that you haven't tested thoroughly enough? We use CDs a lot for data
exchange at work, and while you may say that most CDs written are just
fine, now and then we get a CD where the write operation seems to have
completed successfully, and when we inspect the filesystem everything
all files are there, and everything looks all right. Then if we
compare the contents of the files with the actual data written, we
find that there is a mismatch! Also, there are cases where the CD can
be read just fine on the CD-R(W) unit used to write it, but other
readers have problems with it - watch out for those.

Always possible that I haven't tested thoroughly enough. I tend to
use data compare verifiction in the writing program, and then
periodically use CDR-diagnostic to review the general state of a CD
(mostly on a different drive from the original writer). I don't
really do this *systematically* enough to be sure I'm on top of
everything, though.

And I've never had an old CD
fail yet, either audio CDs going back to 1983 or CD-Rs. (The failure
model for pressed CDs is actually very different than for CD-R; a
pressed CD can de-laminate, and some early ones did.)

A CD-R is laminated, too, isn't it?

I don't believe so; at least not in anything like the same way.

[ ... ]

One CD per 36-picture film? So you're storing 18 megabytes per
image?
My fine jpeg (most often used) original is about 2 megabytes. If I've
worked on the image I have a much bigger PSD file, but often the image
can be auto-processed into a screen image of about 40k, with no
intermediate kept. I average a LOT more than 36 pictures per CD.
JPEG destroys the data as a manner of course, and I've been brought up
to believe it's evil... If we assume pictures are kept, we should
assume the *actual pictures* are kept, not some degraded form of
them. That's the only fair way of comparing. But of course, the mere
ability to neither keep nor completely throw away the originals, but
do something in between, is of course in its own right a point in the
favour of digital cameras.

The JPEG is the camera original. I'll shoot RAW files for exceptional
circumstances, but at 12MB per shot (and hence long write times, too)
I rarely use it.
--
David Dyer-Bennet, , http://www.dd-b.net/dd-b/
RKBA: http://noguns-nomoney.com/ http://www.dd-b.net/carry/
Pics: http://dd-b.lighthunters.net/ http://www.dd-b.net/dd-b/SnapshotAlbum/
Dragaera/Steven Brust: http://dragaera.info/

Toralf writes:

David Dyer-Bennet wrote:
Toralf writes:

David Dyer-Bennet wrote:

[ ... ]


You, and many other people when discussing these matters, seem to
forget that the storage space necessary for the digital cameras is not
free of charge, though (neither are the prints, even if you have your
own printer.)

The storage space costs something like $0.60 per 650 meg, for
archival-grade CDs.

I don't think I could get them *that* cheap around here and/or I
wouldn't trust the ones I do get at that price, but never you mind.


Or something like $100 for 120gig of hard drive.

Yep. Or $100 (2) for 25GB of tape.
I would *never* use tape for archival storage; it has a *very* short
life.

I think you are confusing actual media ("shelf") life with media
durability. Tapes are subject to wear and tear to a much higher degree
than e.g. CD, so their lifetime drops somewhat significantly if they
are used continually. For archival, purposes, however, tapes are
usually expected to last at least 30 years - and that's based on
actual experience, since tape technology is more that 50 years old.

Last time I looked into tape life, archivists thought 10 years was
optimistic. They may, of course, have improved the oxide and binder
formulations since then.

I have gotten some "coasters", immediate burn failures. But I've
never seen a verification error, ever.

Maybe you've been lucky, or perhaps we've been unlucky. Or could it be
that you haven't tested thoroughly enough? We use CDs a lot for data
exchange at work, and while you may say that most CDs written are just
fine, now and then we get a CD where the write operation seems to have
completed successfully, and when we inspect the filesystem everything
all files are there, and everything looks all right. Then if we
compare the contents of the files with the actual data written, we
find that there is a mismatch! Also, there are cases where the CD can
be read just fine on the CD-R(W) unit used to write it, but other
readers have problems with it - watch out for those.

Always possible that I haven't tested thoroughly enough. I tend to
use data compare verifiction in the writing program, and then
periodically use CDR-diagnostic to review the general state of a CD
(mostly on a different drive from the original writer). I don't
really do this *systematically* enough to be sure I'm on top of
everything, though.

And I've never had an old CD
fail yet, either audio CDs going back to 1983 or CD-Rs. (The failure
model for pressed CDs is actually very different than for CD-R; a
pressed CD can de-laminate, and some early ones did.)

A CD-R is laminated, too, isn't it?

I don't believe so; at least not in anything like the same way.

[ ... ]

One CD per 36-picture film? So you're storing 18 megabytes per
image?
My fine jpeg (most often used) original is about 2 megabytes. If I've
worked on the image I have a much bigger PSD file, but often the image
can be auto-processed into a screen image of about 40k, with no
intermediate kept. I average a LOT more than 36 pictures per CD.
JPEG destroys the data as a manner of course, and I've been brought up
to believe it's evil... If we assume pictures are kept, we should
assume the *actual pictures* are kept, not some degraded form of
them. That's the only fair way of comparing. But of course, the mere
ability to neither keep nor completely throw away the originals, but
do something in between, is of course in its own right a point in the
favour of digital cameras.

The JPEG is the camera original. I'll shoot RAW files for exceptional
circumstances, but at 12MB per shot (and hence long write times, too)
I rarely use it.
--
David Dyer-Bennet, , http://www.dd-b.net/dd-b/
RKBA: http://noguns-nomoney.com/ http://www.dd-b.net/carry/
Pics: http://dd-b.lighthunters.net/ http://www.dd-b.net/dd-b/SnapshotAlbum/
Dragaera/Steven Brust: http://dragaera.info/

Toralf writes:

Of course, the actual number doesn't really matter if you're only
comparing different cameras - as long as all producers use the same
way of counting. If you compare e.g. with scans, lower numbers would
make camera image quality sound worse - but that would be fair,
really, since 6 million pixels from a scanner is clearly "better" in
most cases than 6 million from a camera. But of course, it woudn't be
easy to choose the right number; 1.5M or 2M would probably be unfair
to the cameras...

You've got that backwards; digital original pixels are *better* than
scanned pixels, at a ratio of somewhere around 1.2:1 to 2:1 (varies
with image content, camera, phase of the moon, etc.).
--
David Dyer-Bennet, , http://www.dd-b.net/dd-b/
RKBA: http://noguns-nomoney.com/ http://www.dd-b.net/carry/
Pics: http://dd-b.lighthunters.net/ http://www.dd-b.net/dd-b/SnapshotAlbum/
Dragaera/Steven Brust: http://dragaera.info/

Toralf writes:

Of course, the actual number doesn't really matter if you're only
comparing different cameras - as long as all producers use the same
way of counting. If you compare e.g. with scans, lower numbers would
make camera image quality sound worse - but that would be fair,
really, since 6 million pixels from a scanner is clearly "better" in
most cases than 6 million from a camera. But of course, it woudn't be
easy to choose the right number; 1.5M or 2M would probably be unfair
to the cameras...

You've got that backwards; digital original pixels are *better* than
scanned pixels, at a ratio of somewhere around 1.2:1 to 2:1 (varies
with image content, camera, phase of the moon, etc.).
--
David Dyer-Bennet, , http://www.dd-b.net/dd-b/
RKBA: http://noguns-nomoney.com/ http://www.dd-b.net/carry/
Pics: http://dd-b.lighthunters.net/ http://www.dd-b.net/dd-b/SnapshotAlbum/
Dragaera/Steven Brust: http://dragaera.info/