Sign in with
Sign up | Sign in
Your question

Can a 16.6mp SLR really outclass ISO 100 color film? -- Po..

Last response: in Digital Camera
Share
May 13, 2005 6:40:26 AM

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

I still prefer film. In all these comparisons, they never address the fact
that film and digital have different looks. The resolution, color accuracy,
and noise ratings don't account for the different results of film and
digital. It's like the difference between the look of movie film and video,
a quality that's readily apparent. Choosing one over the other is a matter
of preference.

http://www.popphoto.com/article.asp?section_id=4&articl...

Digital Directions:
Against The Grain
By Michael J. McNamara
April 2005

Can a 16.6mp SLR really outclass ISO 100 color film?


Ever since we published the test results for the Canon EOS-1Ds Mark II
(February 2005), we've been inundated with letters, e-mails, forum postings,
faxes, and singing telegrams questioning our sanity and objectivity. Here's
a typical one:

"I find it hard to believe that the Canon EOS-1Ds Mark II can produce better
image quality than good ISO 100 film. I've heard that it would take
24MP-30MP to equal the image quality of such film. Since this camera's
sensor is nowhere near that value, I can only believe that this claim is
just a bunch of Canon sales BS. Let's see some side-to-side
comparisons." -James E. Gingrich, via e-mail

Mr. Gingrich is right on one score: Several years ago, we stated that a
digital camera would need between 24 and 30 megapixels to equal the
resolution of ISO 100 film, assuming both were shot on a sturdy tripod using
mirror lockup and a great lens. We based that estimate on the results we
were getting from earlier-model digital cameras in the 3MP-5MP class. But
does our estimate still hold, and if so, how did the EOS-1Ds Mark II get a
higher image-quality rating?

Before we resolve this issue (pun intended), let's clarify the term at the
heart of this controversy: image quality. Pop Photo has always maintained
that the image quality of any photographic system-film, digital, or
otherwise-can't be determined solely by testing resolution. Other important
factors include color accuracy, noise levels (or grain), highlight and
shadow detail (dynamic range), and contrast.

In the Pop Photo Lab, we measure image-quality parameters in a tightly
controlled environment, and weigh them to determine a camera's image-quality
rating. Resolution and color accuracy get the most weight, followed by the
results of our noise, contrast, and highlight/shadow tests. For compact and
EVF cameras, we also determine distortion, chromatic aberration, blur, and
vignetting using our DxO Analyzer software and targets. Then we shoot field
tests to confirm our lab findings and demonstrate a camera's capabilities
visually.

Let's clear up something else, too. Many camera salespeople and
photographers believe that there's a 1:1 relationship between resolution and
megapixels. But that's not the case. Doubling the pixel count of a camera's
CCD or CMOS sensor doesn't double its resolution. (This only happens with
linear sensors found in scanners.) Camera sensors have pixels arrayed in two
dimensions (horizontal and vertical), so it takes four times as many pixels
to double the resolution of a digital camera (with all other things, such as
noise, lens quality, and focus accuracy, being equal). Therefore, the
resolution difference between a full-frame 30MP sensor (if it existed) and a
16.6MP sensor might only be 25-30 percent, not 100 percent.

For our February 2005 camera test, we compared the resolution, color
accuracy, and noise ratings of EOS-1Ds Mark II images to those of Fujicolor
Superia ISO 100 film shot two years ago for our test of the original 11MP
EOS-1Ds (February 2003). But the uproar from our readers convinced us to
re-evaluate film resolution, using a new roll of Kodak Gold 100 loaded in a
Canon EOS 3 SLR. This time around, the Gold 100 film captured 3000 lines in
all directions when shot using our test lab's daylight-balanced HMI lights
(results based on analysis of a standard IT-10 test target shot with a Canon
50mm f/1.4 EF lens at f/8). Using daylight-balanced Elinchrom 1200S flash
units, film delivered 2700 lines in all directions. These results compare to
2400 lines captured by the older Fuji ISO 100 film (illuminated by the
studio flash units) and to the EOS-1Ds Mark II's 2760 Vertical x 2810
Horizontal x 2220 Diagonal lines (HMI or flash).

In real-world shooting, you'd barely notice a resolution difference in
images captured by both systems, especially if you were handholding the
camera. However, color negative film still captures a better image when
overexposed due to its wider exposure range. In its favor, the EOS-1Ds Mark
II gives you several stops of exposure leeway when shooting in RAW mode, and
also lets you adjust ISO, white balance, sharpness, saturation, and contrast
in the camera or afterward when processing RAW files.

The EOS-1Ds Mark II gets a color accuracy rating of excellent based on an
Avg. Delta E of 6.85, while the film gets a normal rating based on a Delta E
of 12.4 (the result was similar when custom-printed on Kodak Endura paper or
when scanned). The EOS also boasts very low noise at ISO 100, with only
moderately low noise at ISO 800. While grain is less obvious in ISO 100 film
compared with ISO 800 film, it's large enough in giant enlargements to
obscure some detail. On the other hand, there's no grain and very low noise
in shots from the EOS-1Ds Mark II, making it the clear winner.

The difference in image quality is apparent when comparing enlargements from
both cameras (see side-by-side photos, above).

I'd say it's settled. ISO 100 color negative film may capture a bit more
detail than the 1Ds Mark II under ideal lighting conditions, with a great
lens, and on a supersteady tripod. But for its better color and lower noise,
the "Color Image Quality" award goes to Canon's $8,000 digital SLR. Seeing
is believing.or is it vice-versa?
May 13, 2005 6:40:27 AM

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

Muffin wrote:

> I still prefer film.

Good thing you're not biased then..
--

Stacey
Anonymous
May 13, 2005 2:06:33 PM

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

Ron Hunter wrote:
> No surprise. Many people still prefer the sound of tube-type audio
> amplifiers... More distortion, but they are USED to that sound, and
> they LIKE it. It doesn't matter that it is not technically as good,
> they just like it better. I like filet mignon, my wife prefers New
York
> Strip. Who's right?
>

You are right, filet mignon is better.

Scott
May 13, 2005 4:09:50 PM

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

>
>> I still prefer film.
>
> Good thing you're not biased then..
> --
>

It's not a bias, just a preference. I prefer the look that I can achieve
using still film, just like some film makers prefer 16mm movie film over
high def video. Higher ratings on certain technical specs are not the
issue.

Muffin
Anonymous
May 13, 2005 4:09:51 PM

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

Muffin wrote:
>>>I still prefer film.
>>
>>Good thing you're not biased then..
>>--
>>
>
>
> It's not a bias, just a preference. I prefer the look that I can achieve
> using still film, just like some film makers prefer 16mm movie film over
> high def video. Higher ratings on certain technical specs are not the
> issue.
>
> Muffin
>
>
No surprise. Many people still prefer the sound of tube-type audio
amplifiers... More distortion, but they are USED to that sound, and
they LIKE it. It doesn't matter that it is not technically as good,
they just like it better. I like filet mignon, my wife prefers New York
Strip. Who's right?


--
Ron Hunter rphunter@charter.net
Anonymous
May 13, 2005 6:47:16 PM

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

"Muffin" <muffin@buttered.net> wrote in
news:iO0he.1213$ce6.233@newssvr30.news.prodigy.com:

>
>>
>>> I still prefer film.
>>
>> Good thing you're not biased then..
>> --
>>
>
> It's not a bias, just a preference.

Then shoot film and be happy.

Why you feel the need to "discuss" this here is completely beyond any
logical explanation.
May 13, 2005 7:18:53 PM

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

"Muffin" <muffin@buttered.net> wrote:

>I still prefer film. In all these comparisons, they never address the fact
>that film and digital have different looks. The resolution, color accuracy,
>and noise ratings don't account for the different results of film and
>digital. It's like the difference between the look of movie film and video,
>a quality that's readily apparent. Choosing one over the other is a matter
>of preference.

I assume you live in a free country so enjoy your freedom. I still
have a bit of nostalgia and break out my old EF and F1 at times but
they are the 'spare' camera's now a days.

Heck, some people prefer b&w. So have fun, be happy and sooner than
later techology will give you another choice.

Wes


--
Reply to:
Whiskey Echo Sierra Sierra AT Alpha Charlie Echo Golf Romeo Oscar Paul dot Charlie Charlie
Lycos address is a spam trap.
Anonymous
May 13, 2005 8:06:22 PM

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

On Fri, 13 May 2005 09:49:58 -0500, Ron Hunter wrote:

> I like filet mignon, my wife prefers New York
> Strip. Who's right?

Both. You don't seem like a Chippendales kinda guy.
Anonymous
May 13, 2005 8:08:47 PM

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

On 13 May 2005 10:06:33 -0700, Scott W wrote:

>> I like filet mignon, my wife prefers New York
>> Strip. Who's right?
>
> You are right, filet mignon is better.

Filet mignon is more tender, but most agree that it's taste
suffers in comparison with many other cuts.
Anonymous
May 13, 2005 10:38:15 PM

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

>Muffin writes ...
>
>When the 6.3mp Digital Rebel came out not long ago, camera store sales

>people were touting it as the equivalent of 35mm film ...
>
>Today, 16.6mp class DSLRs are being hailed by some parties as equal to

>medium format film quality, but Pop Photo measurements put it at the
>35mm ISO 100 level. Interesting.

Here's a quote from the Pop Photo article, snipped from your first post
....
"let's clarify the term at the
heart of this controversy: image quality. Pop Photo has always
maintained
that the image quality of any photographic system-film, digital, or
otherwise-can't be determined solely by testing resolution. Other
important
factors include color accuracy, noise levels (or grain), highlight and
shadow detail (dynamic range), and contrast."

Basically the magazines (including Pop Photo) have said (and measured)
that earlier digital cameras were already better than film with regards
to "color accuracy and noise levels (or grain)", which are key parts of
image quality, but 100 speed neg film resolved more detail. Once you
go to 200-400 speed films it's a different story though. For many of
us this still meant we got better results (ie, prints) from 8 Mpixel
class digital cameras than we did from 35 mm.

Now the new article is simply going a step further and saying the last
advantage of film (except dynamic range) is also gone with 16 Mpix.

I've shot a lot with ISO 50 and 100 slide film and with 6, 8 and 11
Mpixel digital bodies and from those experiences I'd put images (large
prints) from 6 Mpixel bodies below film, 8 Mpixel bodies (1D Mark II)
pretty much equal and 11 Mpixels (1Ds) clearly better than 35 mm film
for prints regardless of resolution, so it's not surprising to hear 16
Mpix is doing an even better job.

Bill
Anonymous
May 14, 2005 12:40:24 AM

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

Scott W wrote:
> Ron Hunter wrote:
> > No surprise. Many people still prefer the sound of tube-type audio
>
>>amplifiers... More distortion, but they are USED to that sound, and
>>they LIKE it. It doesn't matter that it is not technically as good,
>>they just like it better. I like filet mignon, my wife prefers New
>
> York
>
>>Strip. Who's right?
>>
>
>
> You are right, filet mignon is better.
>
> Scott
>
Thanks, I agree. Now I leave it to you to convince HER. Grin.
Hint, it will be like trying to convince some film users that digital is
worth a damn.


--
Ron Hunter rphunter@charter.net
Anonymous
May 14, 2005 12:41:24 AM

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

ASAAR wrote:
> On 13 May 2005 10:06:33 -0700, Scott W wrote:
>
>
>>> I like filet mignon, my wife prefers New York
>>>Strip. Who's right?
>>
>>You are right, filet mignon is better.
>
>
> Filet mignon is more tender, but most agree that it's taste
> suffers in comparison with many other cuts.
>
I like the taste, texture, and the 'all meat' aspect. For sheer best
taste, a nice well-marbled rib-eye is my favorite, but I am trying to
cut down on the fat.


--
Ron Hunter rphunter@charter.net
Anonymous
May 14, 2005 2:04:19 AM

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

On Fri, 13 May 2005 20:41:24 -0500, Ron Hunter wrote:

> I like the taste, texture, and the 'all meat' aspect. For sheer best
> taste, a nice well-marbled rib-eye is my favorite, but I am trying to
> cut down on the fat.

That and porterhouse are very good. Much of the fat can be
rendered out when cooking, but if there isn't much marbling I'd
rather have a decent burger. If I ever have to really cut down I'll
probably cheat by hoarding bacon.
May 14, 2005 3:06:00 AM

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

>>
> No surprise. Many people still prefer the sound of tube-type audio
> amplifiers...

When the 6.3mp Digital Rebel came out not long ago, camera store sales
people were touting it as the equivalent of 35mm film, and many users on
rec.photo.digital believed it. Today, 16.6mp class DSLRs are being hailed
by some parties as equal to medium format film quality, but Pop Photo
measurements put it at the 35mm ISO 100 level. Interesting.

In my opinion, Popular Photography has "gone digital" because most of its
advertisers are selling digital products. Any of their tests comparing film
and digital can hardly be called objective. My subjective preference is for
film.

Muffin
Anonymous
May 14, 2005 5:41:51 AM

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

ASAAR wrote:
> On Fri, 13 May 2005 20:41:24 -0500, Ron Hunter wrote:
>
>
>>I like the taste, texture, and the 'all meat' aspect. For sheer best
>>taste, a nice well-marbled rib-eye is my favorite, but I am trying to
>>cut down on the fat.
>
>
> That and porterhouse are very good. Much of the fat can be
> rendered out when cooking, but if there isn't much marbling I'd
> rather have a decent burger. If I ever have to really cut down I'll
> probably cheat by hoarding bacon.
>
I LOVE bacon, but I don't eat much of it any more. Pork, chicken, fish,
and beef. I like them all.


--
Ron Hunter rphunter@charter.net
May 14, 2005 8:17:52 AM

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

>
> Here's a quote from the Pop Photo article, snipped from your first post
> ...
>> "let's clarify the term at the
>> heart of this controversy: image quality. Pop Photo has always
>> maintained
>> that the image quality of any photographic system-film, digital, or
>> otherwise-can't be determined solely by testing resolution. Other
>> important
>> factors include color accuracy, noise levels (or grain), highlight and
>> shadow detail (dynamic range), and contrast."
>
> Basically the magazines (including Pop Photo) have said (and measured)
> that earlier digital cameras were already better than film with regards
> to "color accuracy and noise levels (or grain)", which are key parts of
> image quality, but 100 speed neg film resolved more detail. Once you
> go to 200-400 speed films it's a different story though. For many of
> us this still meant we got better results (ie, prints) from 8 Mpixel
> class digital cameras than we did from 35 mm.
>

The quoted passage simply says that resolution isn't the only factor in
judging image quality, either from film or digital capture. Later in the
article, the magazine gives this particular digital camera, the EOS-1Ds Mark
II a color rating of excellent, and this particular film, Kodak Gold 100, a
rating of normal on color accuracy. Given that Adorama sells a EOS-1Ds Mark
II kit for $8,000.00, I would hope its color accuracy is freaking excellent.
Is the digital Rebel or the other DSLRs used by poor folks as accurate?
Nope. Gold 100 has good color, but there are other ISO 100 print films to
choose from that have excellent color, and Fuji Superia X-TRA at all speeds
has color accuracy as good as any digital camera. They found that the
EOS-1Ds Mark II had less noise/grain than Gold 100 had grain, but If they
had tested a Fuji 100 film, grain would have been much less noticeable. Pop
Photo extrapolates the results from Gold 100 to all ISO 100 films, which, of
course, isn't tenable.

Bearing in mind that we are only talking about Kodak Gold 100, the film has
a slight advantage in resolution but it's negligible in real world shooting.
The EOS-1Ds Mark II has an edge in color accuracy, but that would likely
disappear if Fuji Superia X-TRA was used for the comparison, as would the
DSLR's lower noise/grain. When you are talking about an $8,000.00 kit
compared to a $5.00 roll of film, that's not saying much.

> I've shot a lot with ISO 50 and 100 slide film and with 6, 8 and 11
> Mpixel digital bodies and from those experiences I'd put images (large
> prints) from 6 Mpixel bodies below film, 8 Mpixel bodies (1D Mark II)
> pretty much equal and 11 Mpixels (1Ds) clearly better than 35 mm film
> for prints regardless of resolution, so it's not surprising to hear 16
> Mpix is doing an even better job.
>

How are you doing the comparison? Do you agree that a fair comparison would
be wet prints from the film and the digital file?

The bottom line, if you believe the article, is that if you shoot Kodak Gold
100 film with a good camera and lens, you're at about the same quality level
as a 16.6mp EOS-1Ds Mark II, with the DSLR doing a little better on color
accuracy and grain/noise. If you shoot Fuji Superia X-TRA, I personally
believe that the film might do a little better, but I don't plan on buying
an EOS-1Ds Mark II and running a test.

Where does that leave the quality of the 6.3mp to 8.3mp DSLRs most people
have? At a level less than ISO 100 print film, obviously.

Muffin

P.S. The EOS-1Ds Mark II is out of stock at Adorama, but you can put your
name down for one and it will ship when they become available. Here's the
link. Go for it.

http://www.adorama.com/ICA1DSM2.html?searchinfo=EOS-1Ds...
May 14, 2005 1:14:37 PM

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

"Muffin" <muffin@buttered.net> wrote in message
news:usUge.1227$aM4.63@newssvr11.news.prodigy.com...
> I still prefer film. In all these comparisons, they never address the
fact
> that film and digital have different looks. The resolution, color
accuracy,
> and noise ratings don't account for the different results of film and
> digital. It's like the difference between the look of movie film and
video,
> a quality that's readily apparent. Choosing one over the other is a
matter
> of preference.
>
>
http://www.popphoto.com/article.asp?section_id=4&articl...
>
> Digital Directions:
> Against The Grain
> By Michael J. McNamara
> April 2005
>
> Can a 16.6mp SLR really outclass ISO 100 color film?
>
>
> Ever since we published the test results for the Canon EOS-1Ds Mark II
> (February 2005), we've been inundated with letters, e-mails, forum
postings,
> faxes, and singing telegrams questioning our sanity and objectivity.
Here's
> a typical one:
>
> "I find it hard to believe that the Canon EOS-1Ds Mark II can produce
better
> image quality than good ISO 100 film. I've heard that it would take
> 24MP-30MP to equal the image quality of such film. Since this camera's
> sensor is nowhere near that value, I can only believe that this claim is
> just a bunch of Canon sales BS. Let's see some side-to-side
> comparisons." -James E. Gingrich, via e-mail
>
> Mr. Gingrich is right on one score: Several years ago, we stated that a
> digital camera would need between 24 and 30 megapixels to equal the
> resolution of ISO 100 film, assuming both were shot on a sturdy tripod
using
> mirror lockup and a great lens. We based that estimate on the results we
> were getting from earlier-model digital cameras in the 3MP-5MP class. But
> does our estimate still hold, and if so, how did the EOS-1Ds Mark II get a
> higher image-quality rating?
>
> Before we resolve this issue (pun intended), let's clarify the term at the
> heart of this controversy: image quality. Pop Photo has always maintained
> that the image quality of any photographic system-film, digital, or
> otherwise-can't be determined solely by testing resolution. Other
important
> factors include color accuracy, noise levels (or grain), highlight and
> shadow detail (dynamic range), and contrast.
>
> In the Pop Photo Lab, we measure image-quality parameters in a tightly
> controlled environment, and weigh them to determine a camera's
image-quality
> rating. Resolution and color accuracy get the most weight, followed by the
> results of our noise, contrast, and highlight/shadow tests. For compact
and
> EVF cameras, we also determine distortion, chromatic aberration, blur, and
> vignetting using our DxO Analyzer software and targets. Then we shoot
field
> tests to confirm our lab findings and demonstrate a camera's capabilities
> visually.
>
> Let's clear up something else, too. Many camera salespeople and
> photographers believe that there's a 1:1 relationship between resolution
and
> megapixels. But that's not the case. Doubling the pixel count of a
camera's
> CCD or CMOS sensor doesn't double its resolution. (This only happens with
> linear sensors found in scanners.) Camera sensors have pixels arrayed in
two
> dimensions (horizontal and vertical), so it takes four times as many
pixels
> to double the resolution of a digital camera (with all other things, such
as
> noise, lens quality, and focus accuracy, being equal). Therefore, the
> resolution difference between a full-frame 30MP sensor (if it existed) and
a
> 16.6MP sensor might only be 25-30 percent, not 100 percent.
>
> For our February 2005 camera test, we compared the resolution, color
> accuracy, and noise ratings of EOS-1Ds Mark II images to those of
Fujicolor
> Superia ISO 100 film shot two years ago for our test of the original 11MP
> EOS-1Ds (February 2003). But the uproar from our readers convinced us to
> re-evaluate film resolution, using a new roll of Kodak Gold 100 loaded in
a
> Canon EOS 3 SLR. This time around, the Gold 100 film captured 3000 lines
in
> all directions when shot using our test lab's daylight-balanced HMI lights
> (results based on analysis of a standard IT-10 test target shot with a
Canon
> 50mm f/1.4 EF lens at f/8). Using daylight-balanced Elinchrom 1200S flash
> units, film delivered 2700 lines in all directions. These results compare
to
> 2400 lines captured by the older Fuji ISO 100 film (illuminated by the
> studio flash units) and to the EOS-1Ds Mark II's 2760 Vertical x 2810
> Horizontal x 2220 Diagonal lines (HMI or flash).
>
> In real-world shooting, you'd barely notice a resolution difference in
> images captured by both systems, especially if you were handholding the
> camera. However, color negative film still captures a better image when
> overexposed due to its wider exposure range. In its favor, the EOS-1Ds
Mark
> II gives you several stops of exposure leeway when shooting in RAW mode,
and
> also lets you adjust ISO, white balance, sharpness, saturation, and
contrast
> in the camera or afterward when processing RAW files.
>
> The EOS-1Ds Mark II gets a color accuracy rating of excellent based on an
> Avg. Delta E of 6.85, while the film gets a normal rating based on a Delta
E
> of 12.4 (the result was similar when custom-printed on Kodak Endura paper
or
> when scanned). The EOS also boasts very low noise at ISO 100, with only
> moderately low noise at ISO 800. While grain is less obvious in ISO 100
film
> compared with ISO 800 film, it's large enough in giant enlargements to
> obscure some detail. On the other hand, there's no grain and very low
noise
> in shots from the EOS-1Ds Mark II, making it the clear winner.
>
> The difference in image quality is apparent when comparing enlargements
from
> both cameras (see side-by-side photos, above).
>
> I'd say it's settled. ISO 100 color negative film may capture a bit more
> detail than the 1Ds Mark II under ideal lighting conditions, with a great
> lens, and on a supersteady tripod. But for its better color and lower
noise,
> the "Color Image Quality" award goes to Canon's $8,000 digital SLR. Seeing
> is believing.or is it vice-versa?
>
>
>

I shoot film & digital. I usually use a transparancy film over negative film
when quality is what I'm after - say Sensia 100 - for me it produces less
grain and a far better image. Also it has the advantage of being a positive
image so is far easier to preview from the film strip. I usually scan them
and then print digitally.

Alan.
Anonymous
May 16, 2005 10:43:04 PM

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

Ron Hunter <rphunter@charter.net> wrote:

> No surprise. Many people still prefer the sound of tube-type audio
> amplifiers... More distortion, but they are USED to that sound, and
> they LIKE it. It doesn't matter that it is not technically as good,
> they just like it better. I like filet mignon, my wife prefers New York
> Strip. Who's right?

Years ago, there was a nice anecdote about an audio mastering lab that
produced really nice and sounding digital recordings. At some point the
owner revealed his secret. He had a "blank" vinyl album made (with
unmodulated grooves) and mixed the sound from that being played back in
with his digital recordings. Listeners were fooled by the almost
subliminal audio information and reported "warmer" sound.

Ton
!