8Mp Digital The Theoretical 35mm Quality Equivelant

You haven't made any compelling case questioning the legitimacy, and of
course you have not yet posted a single link or reference that supports
your case.

Just like you seem to now be down to zero supporters.

I think it is now getting perilously close to the point where you just
get ignored.

(Stephen H. Westin) wrote in
:

Jon Pike writes:

"David J. Littleboy" wrote in
:


"Mike Kohary" wrote:
David J. Littleboy wrote:

Quite the contrary, science is about quantifying objective
reality. The objective reality in photography is which picture
looks better to the viewers.

You are exactly correct.

Thanksg!

The problem here is that

The real problem here is that Jon has an agenda, namely that 35mm
film is better than dSLR digital. When a test show otherwise, he
attacks the test. That's what he's doing with his "The scans don't
resolve the grain" rant.

I'll 'attack' (QUESTION) any test that's conducted in such a
haphazard and unmethodical way.

Your claim is that his results *may* be wrong. The best way to prove
that is with results of your own.

His 'results' -are- useless until he gets -all- the bugs out of his
methodology. It's that simple. And judging from his attitude ("Show me
the IMAGES!"), that will never happen because he's approaching it from a
purely subjective standpoint. That means he'll never engage in
reproducable, standardized tests.

He's not alone in his claim that scanners don't do justice to film
and that you have to make projection prints using wet chemistry.
That this claim doesn't hold up in real life doesn't prevent the
film agenda types from making it over and over and over again.

Actually, we've seen pages that quantifiably show a 10% reduction of
resolution when film is scanned.

Sorry, but what two numbers did you divide to get 10%? I've lost
track.

Then go back through the thread.

It seems like you're chosing to ignore
those pages. Why is that?

Because, as I recall, you were complaining about any reduction in
limiting spatial resolution as a serious loss of image detail. And
that's not clear at all.

You all seem to assume that the exact opposite is the case, and there's
no foundation to make that assumption whatsoever.
What we do know is there is a measurable loss of resolution, and that is
-not- "noise," and it is -not- grain. Yet you choose to ignore it anyway,
because it goes against what you think.

Is it because you claim you can make a subjective judgement about
image quality and how it's not affected by lack of scanning
precision? You may be willing to accept a 10% margin of error across
the board, but I'm not.

That is to say, ignorance is better than any measurement that may
contain error. Which is all of them, by the way...

That's not what I'm saying at all. What I'm saying is you should not be
going out of your way to produce error, when there is clearly no need to.

Jon is attempting to
apply severe scientific scrutiny to a subject that isn't hard
science at all. There is a heavy artistic aspect here to
consider, and any attempt
to
quantify such is doomed to fail before we even begin.

Ah, but I disagreeg. I think that much of photography very much
is quantifiable, and that by quantifying it we can better select
our tools for our imaging needs. But you do have to ask the right
question to get the right answer. The ability of scanners to
resolve 120 lp/mm (bobm) or the grain (Jon) is completely
irrelevant.

there you go again with the subjective judgement. that's what roger
was trying to do with this page:
http://clarkvision.com/tmp/testchart...ntlowermtf.jpg
when in actuality, neither of them are perfectly clear. Why are they
not clear? He's 'c'd himself, so that implies he's generated this
image from scratch. If not, then he's stolen it from someone else and
not given them credit.

You're surely not suggested that that image was scanned, are you?
If so, you've obviously never dealt with scanned images.

.....
if it was scanned, he must have generated it on his computer in the first
place, and then printed it. and then scanned it.
How stupid would that be?
If he scanned it from somewhere else, again, he's stealing the image
because he's c'd it to himself.
Most likely, he generated it on his computer, and didn't take into
account the effects that jpeg compression has on images. That's just
carelessness and bad methodology.

If he's the one who generated it, the question stands, why is one
not perfectly clear, and the other (the 'distorted' one) not clear?
Chances are, the lack of clarity results from jpeg compression,

Bingo.

but he
should have taken that into account before-hand.

Why?

Because it skews an already screwed up test.

more bad methodology from roger.

Why don't you quantify the difference in results using a printed copy
of his image vs. those using a printed copy of a "perfect" version?
Are there any? I bet not.

If the printer's of any quality, of course there are. But you're trying
to change the issue again.

The fact of the matter is he didn't (and continues not to, as we see from
this latest 'test') plan out any sort of procedure, didn't take care to
eliminate (and in this case, yes, some errors could have been -
eliminated-) or reduce any of the "unforseen" (which he really should
have seen coming) errors that arise from generally sloppy procedures.

This test is not testing whether there is a quantifiable difference
between two sets of lines. Rather, it's attempting to test your eyes.
It's attempting to test your monitor. It's attempting to test your
"skills/knowledge/whateveryouwannacallit" about how you set up your
monitor resolution for your particular screen size. Some people (and i'm
sure he's one of them) will argue that there is an "ideal" resolution to
make things appear "right" on your screen. But if I want things, in
general, to appear bigger than you do, or smaller, that doesn't make
either person "right" or "wrong."

Again, a seriously flawed test from the conception ("i'll see if they can
tell the difference! that'll show them!") to the completion (saving in
jpg!).

What's going on though is a ****-poor comparison. The question
originally asked about theoretical resolution,

Actually, two questions were asked: "Does this mean they have the
theoretical equivalent resolution?" and "Are they the equivalent to
35mm?"

I suppose we have been discussing the latter question, or in fact
disputing whether those were the right questions to ask.

"Does this mean they have the theoretical equivalent resolution?"

I think we can all agree that the answer to this one is "no". But
that doesn't answer the question of which might be better, or whether,
though measurably different, the results might be similar in quality.

As for "equivalent to 35mm", I suppose that no digital camera will
consistently give identical results to any 35mm still film camera.
But if we broaden that to "meeting the same requirement as 35mm",
it's clear that various digital cameras have done this for various
photographers, and not just those who accept a sacrifice in quality.

If you want to reduce things to purely qualitative 'measures' and say
"well, it works for some people, and that's good enough for them" then
i'll agree with that 100%. My mother has a 3mp p&s canon camera, one of
those horrid little rectangles, and it works good enough for her.

However, when discussing things like "theoretical equivalen(cies)," I
prefer to be a little more precise than that.

--
http://www.neopets.com/refer.phtml?username=moosespet

(Stephen H. Westin) wrote in
:

Jon Pike writes:

"David J. Littleboy" wrote in
:


"Mike Kohary" wrote:
David J. Littleboy wrote:

Quite the contrary, science is about quantifying objective
reality. The objective reality in photography is which picture
looks better to the viewers.

You are exactly correct.

Thanksg!

The problem here is that

The real problem here is that Jon has an agenda, namely that 35mm
film is better than dSLR digital. When a test show otherwise, he
attacks the test. That's what he's doing with his "The scans don't
resolve the grain" rant.

I'll 'attack' (QUESTION) any test that's conducted in such a
haphazard and unmethodical way.

Your claim is that his results *may* be wrong. The best way to prove
that is with results of your own.

His 'results' -are- useless until he gets -all- the bugs out of his
methodology. It's that simple. And judging from his attitude ("Show me
the IMAGES!"), that will never happen because he's approaching it from a
purely subjective standpoint. That means he'll never engage in
reproducable, standardized tests.

He's not alone in his claim that scanners don't do justice to film
and that you have to make projection prints using wet chemistry.
That this claim doesn't hold up in real life doesn't prevent the
film agenda types from making it over and over and over again.

Actually, we've seen pages that quantifiably show a 10% reduction of
resolution when film is scanned.

Sorry, but what two numbers did you divide to get 10%? I've lost
track.

Then go back through the thread.

It seems like you're chosing to ignore
those pages. Why is that?

Because, as I recall, you were complaining about any reduction in
limiting spatial resolution as a serious loss of image detail. And
that's not clear at all.

You all seem to assume that the exact opposite is the case, and there's
no foundation to make that assumption whatsoever.
What we do know is there is a measurable loss of resolution, and that is
-not- "noise," and it is -not- grain. Yet you choose to ignore it anyway,
because it goes against what you think.

Is it because you claim you can make a subjective judgement about
image quality and how it's not affected by lack of scanning
precision? You may be willing to accept a 10% margin of error across
the board, but I'm not.

That is to say, ignorance is better than any measurement that may
contain error. Which is all of them, by the way...

That's not what I'm saying at all. What I'm saying is you should not be
going out of your way to produce error, when there is clearly no need to.

Jon is attempting to
apply severe scientific scrutiny to a subject that isn't hard
science at all. There is a heavy artistic aspect here to
consider, and any attempt
to
quantify such is doomed to fail before we even begin.

Ah, but I disagreeg. I think that much of photography very much
is quantifiable, and that by quantifying it we can better select
our tools for our imaging needs. But you do have to ask the right
question to get the right answer. The ability of scanners to
resolve 120 lp/mm (bobm) or the grain (Jon) is completely
irrelevant.

there you go again with the subjective judgement. that's what roger
was trying to do with this page:
http://clarkvision.com/tmp/testchart...ntlowermtf.jpg
when in actuality, neither of them are perfectly clear. Why are they
not clear? He's 'c'd himself, so that implies he's generated this
image from scratch. If not, then he's stolen it from someone else and
not given them credit.

You're surely not suggested that that image was scanned, are you?
If so, you've obviously never dealt with scanned images.

.....
if it was scanned, he must have generated it on his computer in the first
place, and then printed it. and then scanned it.
How stupid would that be?
If he scanned it from somewhere else, again, he's stealing the image
because he's c'd it to himself.
Most likely, he generated it on his computer, and didn't take into
account the effects that jpeg compression has on images. That's just
carelessness and bad methodology.

If he's the one who generated it, the question stands, why is one
not perfectly clear, and the other (the 'distorted' one) not clear?
Chances are, the lack of clarity results from jpeg compression,

Bingo.

but he
should have taken that into account before-hand.

Why?

Because it skews an already screwed up test.

more bad methodology from roger.

Why don't you quantify the difference in results using a printed copy
of his image vs. those using a printed copy of a "perfect" version?
Are there any? I bet not.

If the printer's of any quality, of course there are. But you're trying
to change the issue again.

The fact of the matter is he didn't (and continues not to, as we see from
this latest 'test') plan out any sort of procedure, didn't take care to
eliminate (and in this case, yes, some errors could have been -
eliminated-) or reduce any of the "unforseen" (which he really should
have seen coming) errors that arise from generally sloppy procedures.

This test is not testing whether there is a quantifiable difference
between two sets of lines. Rather, it's attempting to test your eyes.
It's attempting to test your monitor. It's attempting to test your
"skills/knowledge/whateveryouwannacallit" about how you set up your
monitor resolution for your particular screen size. Some people (and i'm
sure he's one of them) will argue that there is an "ideal" resolution to
make things appear "right" on your screen. But if I want things, in
general, to appear bigger than you do, or smaller, that doesn't make
either person "right" or "wrong."

Again, a seriously flawed test from the conception ("i'll see if they can
tell the difference! that'll show them!") to the completion (saving in
jpg!).

What's going on though is a ****-poor comparison. The question
originally asked about theoretical resolution,

Actually, two questions were asked: "Does this mean they have the
theoretical equivalent resolution?" and "Are they the equivalent to
35mm?"

I suppose we have been discussing the latter question, or in fact
disputing whether those were the right questions to ask.

"Does this mean they have the theoretical equivalent resolution?"

I think we can all agree that the answer to this one is "no". But
that doesn't answer the question of which might be better, or whether,
though measurably different, the results might be similar in quality.

As for "equivalent to 35mm", I suppose that no digital camera will
consistently give identical results to any 35mm still film camera.
But if we broaden that to "meeting the same requirement as 35mm",
it's clear that various digital cameras have done this for various
photographers, and not just those who accept a sacrifice in quality.

If you want to reduce things to purely qualitative 'measures' and say
"well, it works for some people, and that's good enough for them" then
i'll agree with that 100%. My mother has a 3mp p&s canon camera, one of
those horrid little rectangles, and it works good enough for her.

However, when discussing things like "theoretical equivalen(cies)," I
prefer to be a little more precise than that.

--
http://www.neopets.com/refer.phtml?username=moosespet

You haven't made any compelling case questioning the legitimacy, and of
course you have not yet posted a single link or reference that supports
your case.

Just like you seem to now be down to zero supporters.

I think it is now getting perilously close to the point where you just
get ignored.

You haven't made any compelling case questioning the legitimacy, and of
course you have not yet posted a single link or reference that supports
your case.

Just like you seem to now be down to zero supporters.

I think it is now getting perilously close to the point where you just
get ignored.

(Stephen H. Westin) wrote in
:

The question was NOT: "which looks better, film or digital?"

But the ultimate question is whether the final image looks better.
Not who can brag about test results.

If you want to bring up that question in a different thread, go ahead!
That was -not- the question asked in this thread though.

--
http://www.neopets.com/refer.phtml?username=moosespet

(Stephen H. Westin) wrote in
:

The question was NOT: "which looks better, film or digital?"

But the ultimate question is whether the final image looks better.
Not who can brag about test results.

If you want to bring up that question in a different thread, go ahead!
That was -not- the question asked in this thread though.

--
http://www.neopets.com/refer.phtml?username=moosespet

Roger's example does that pretty well, if you look at it properly. But
(see about 30 posts down) it appears you have screwed up THERE as well.

And besides, do you not understand the concept of burden of proof? -
you are the one claiming the overall importance of the loss, so simply
show us an example.
No, of course you won't. You *can't* - it's all just talk.

Roger's example does that pretty well, if you look at it properly. But
(see about 30 posts down) it appears you have screwed up THERE as well.

And besides, do you not understand the concept of burden of proof? -
you are the one claiming the overall importance of the loss, so simply
show us an example.
No, of course you won't. You *can't* - it's all just talk.

(Dave Martindale) wrote in news:cp2ijj$o10$1
@mughi.cs.ubc.ca:

Jon Pike writes:

Except that's not what the question was asking.

The question was, is 8mp about equal to 35mm film?
The question was NOT: "which looks better, film or digital?"

You're being awfully fuzzy yourself. What do you think that first
question means? Be as scientific as you want.

I will suggest that someone who asks "is 8 MP about equal to 35 mm film"
is really interested in the answer to "Which subjectively looks better
overall, an 8 MP digital camera image or a 35 mm film image?", while
they mostly do not care about the answer to "Which has the greater
limiting resolution, a 8 MP digital camera image or a 35 mm film image?"

....
except that he clearly stated "theoretical equivalent resolution."

Which of these two very different questions do you think Kodak's 25 MP
number is addressing?

"I heard someone say that 8Mp digital cameras were the equivalent to 35mm
film quality?

Does this mean they have the theoretical equivalent resolution? Are they
the equivalent to 35mm?"

Since it seems you've come into the thread rather "late," there's the
question that started it.

He asked very clearly about theoretical equivalent resolution. He didn't
ask "which one do you think looks better."

--
http://www.neopets.com/refer.phtml?username=moosespet