Focal plane vs. leaf shutters in MF SLRs

On Thu, 17 Jun 2004 02:20:45 GMT, Fil Ament
wrote:

In article ,
"David J. Littleboy" wrote:

I've seen a lot of 35mm at 20x24 to 20x30; 35mm types seem to like it. It
looks gross even from across the room.

Yeah well I would say the same, I don't know exactly how he does it
but I have seen the pictures, theyr'e good. I've also seen plenty of other photographers
work done this size from small negatives that look good. You need to look up
from that microscope on occasion.


Fil, you haven't been around long enough to know about Dave.
The very model of a modern grain sniffer, and I mean that with
all due respect to Dave. It's his thing.

Whatever. I've recently had the experience of selling 20x30"
and 24x36" prints of a number of my photos. Of these, nine
were from 35 mm, four from MF, and one from the 10D.

Needless to say it was a challenge with the 35 mm stuff.
I used Neatimage to process the grain. Nice tool!

Now get this. In one room with three 20x30" prints, two are
from MF and the third is from the 10D. Guess which one
my wife prefers?

FWIW, the prints were made on a Durst Epsilon; shot
mostly on Reala or Portra, and scanned on an LS-8000.


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

On Thu, 17 Jun 2004 02:37:27 GMT, "Neil Gould"
wrote:


I don't hang any prints from my Epson on the wall. I use it to proof
edits, period. It's great for that.

Well, la-de-da, Neil. I'm glad my tastes aren't as
refined as yours, or I'd not be able to indulge in
this hobby/business.

Honestly, this comment would be laughed at
by dozens of fine artists I know and admire.

I'm prepared to accept that others can make
sharper optical prints from 35 or MF than I made
in my own darkroom thirty years ago. I just
haven't seen the evidence yet.

I'm not saying that the prints are sharper than I can do myself. I'm
saying that color wet printing is not something that I find practical in
the volumes that I need. None the less, there isn't an inkjet made that
can match the quality of a wet print.

Ahhh. Repeat that tired old claim again.

In any case you're not inclined to defend
the claim with a print of your own, because
by your own estimate, your optical print
would cost about ten times what mine does,
at 20x20".

Do you get what I mean by irrelevant?
I couldn't afford this hobby (or conduct
my business) if big prints cost $100 apiece.


Has it been 30 years since you last
looked at a good one? ;-)


It may well have been... I haven't seen
photos in museums for a while, and on
the last visit... they were Piezo prints, as
I recall. Most of the galleries I visit are
selling Epson prints.

My interest in "fine" photography and
printmaking was dormant for a good long
time. It was rekindled in 1998 upon seeing
a print from an Epson 600.

Have enlargers improved appreciably
since 1970? I mean, the optics or basic
mechanics?

Given a 35 mm original, how much can
one improve on an Omega B22, EL-
Nikkor 50 mm lens, and reasonably
good technique?


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

yes, however, the stats on replacement value was from a survey limited to
the members of the Advertising Photographers of America with average sales
of $378k (median just under $250k), and represented the top 1% of
income/sales for photography profession as a whole - i.e., the guys doing
shots for advertising agencies and ads etc. ;-)

So think of a guy billing $1,500-$3,000/day - how much gear would s/he
have? ;-) Heck, lots of us amateurs on here have $10k in lenses and 35mm
bodies, or the same in digital gear or in MF (esp. in MF ;-).

If the pros at the top of the profession $$-wise have such moderate kits,
I agree with Gordon that lots of other pros with less variable demands can
have modest kits. Our local campus white-haired contract pro uses a Leica
kit with 3 lenses and flash for campus events and portraits of big shots
and big donors (redundant, I know ;-)

I recall a fun shot of nearly forty amateur photographers lined up and
shooting some birds in the Everglades, nearly every one of them had at
least a 500mm f/4 or 600mm f/4 lens, a few "poor folks" had 300mm f/2.8
and TC ;-) My example use was more to highlight that a single pro 35mm
lens in the fast end of the gene pool and high end brand name (zeiss,
leica..) can easily eat up less than $10k budget for 35mm gear.

Similarly, an SWC is pricey, but again, $10k for MF gear would not go far
if buying rollei or hasselblad, and even bronica or mamiya kit would not
be extensive. And you would still have to buy all those accessories like
tripods and studio strobes and all that

(see http://medfmt.8k.com/mf/gadgets.html for sample camera bag contents
and stuff you probably don't need that you will still want to buy once you
read this list ;-)

grins bobm
--
************************************************** *********************
* Robert Monaghan POB 752182 Southern Methodist Univ. Dallas Tx 75275 *
********************Standard Disclaimers Apply*************************

Hi again Mike ;-)

No, I'm going the reverse way. I am saying that 6 MP is interpolated to
make a decent 8x10" print, and that the underlying data is circa 40 lpmm
resolution equiv. on a 35mm film SLR. Partly, this is due to the
anti-aliasing filter in DSLRs/EVFs, partly other system losses etc.

As for prints, I am NOT claiming that 24 MP means a 16x20" prints at 6MP
8x10" print quality. I am saying that 24 MP (Kodak's estimate for
mid-speed films, "at least") is equal to about 84 lpmm on film. That is a
high but not impossible resolution for slow speed 25-100 ISO films and
very good lenses and technique.

To get a Leica standard 8x10" print, held at 10", you need 8 lpmm on the
paper. Using 24x36mm film, you need an 8 times enlargement factor to do
an 8x10" print. So 8 lpmm * 8X = 64 lpmm needed on film. Add in losses in
enlarging lens or scanning lens and you need well over 64 lpmm on film to
make a leica standard print. So our 24MP equiv. is easily able, at 84
lpmm, to produce an 8x10" Leica standard print. But it can't do so at
16x20", as you infer, as the math here shows. So that's why I'm NOT saying
that ;-)

The implications of the popphoto rating of 100 ISO film at 40 megapixels
is that you can put about 108 lpmm onto film, with the best lenses and
technique. Again, this is in good agreement with a lot of test reports
etc. This would permit an enlargement factor of 13X, meaning a 13"x 19"
print to the leica standard of 8 lpmm on paper (no losses in lens etc.).

Again, neither of these claims is remarkable to most darkroom types or
those reading lens test reports and film resolution values of mfgers etc.
When you make the equiv. calculations for how many Megapixels this
represents, you find that most of the information on film is being lost in
conversions and scanning to digital files. That info is the 60% or so
above 42 lpmm (to 107 lpmm per popphoto tests).

That's why we film "luddites" are screaming foul! ;-) The scanning process
is losing 60% of the data. Yes, the images may still look okay from a 6 MP
camera. But you _don't_ have the fine resolution data. David L. seems to
believe this data is not important to the image - his subjective opinion
and valid for him.

For many of us, we have seen the difference between 40 lpmm lenses and 70
lpmm and even 100 lpmm lenses, and this is why we go for the higher
contrast and higher resolution lenses, even at considerably higher expense
over the trash cheapy lenses which can only deliver 40 lpmm (i.e., digital
camera quality, again, which is limited by the low pass filter for
anti-aliasing in the camera).

So I am saying that 6MP makes a nice image because it is interpolated to
fill in and smooth out the missing data, but close examination should show
that the image lacks the really fine detail and high contrast that
typifies high end lenses and film with proper technique (i.e., beyond 40
lpmm and into 70-100+ lpmm ranges). This is why it takes 24 MP for film to
produce a Leica standard 8x10" print with this kind of high resolution and
fine contrast detail. The 6 MP can't do so, because it lacks this high
frequency data, again, due to the anti-aliasing filter. The digital image
is effectively one taken thru a softening filter of moderate strength.

And so that is the digital vs. film quality difference. The digital image
is essentially what you would get if you shot film with a softening filter
in place - i.e., the anti-aliasing, low pass, softening filter used in
most DSLRs to prevent aliasing. When you take off the softening filter
with a film camera, you get noticeably better contrast and fine detail.
That is the essential difference between digital and analog imagery today.

my $.02

;-) bobm
--
************************************************** *********************
* Robert Monaghan POB 752182 Southern Methodist Univ. Dallas Tx 75275 *
********************Standard Disclaimers Apply*************************

Hi David!

yes, you are right; 35mm enlarged past 12X is hard to get a "leica
standard" print quality (8 lpmm on print). That said, a lot of people are
happy with less than 8 lpmm on the print, some with older eyes (ahem) find
6 lpmm okay, others are happy at even 4 lpmm.

In a related post, I do the math for what you can expect for 35mm enlarged
8x (about what 24 MP equiv. kodak claims suggest "at least") to 13X
(maximum of 107 lpmm (which is popphoto's test result for 40 MP equiv. for
film).

The real question is why does a 6MP print look okay at 8x10" when it takes
a 24 MP film print to reach the "leica standard". First, the
smoothed/spread dots of ink jet printers don't reach the 8 lpmm leica
standard (8 lpmm * 2 dots/line * 24.5 mm/" = 392 dpi (there's about 71%
more info on the leica standard print, ignoring dot spreading issues).

The bigger issue is digital vs film print quality. 6 million pixels
spread over 80 square inches (8x10" print) would be 75,000 pixels per sq.
inch, or 273 pixels per inch or 11.2 pixels per mm, and at 2 pixels per
line, that's only 5.6 lpmm. Again, 5.6 lpmm * 2 dots/line * 24.5mm per
inch = 275 dpi. That's close enough to 300 dpi with a border.

But 392 dpi (film) vs. 275 dpi (digital) on an 8x10" print is still
comparing 392*392 (153k) against 275*275 (75.6k), which is 203% more dots
in the film based prints. So the short answer is that the film print made
from a 24 MP film equiv should be twice as good as a 6 MP (again, obvious
if you think about the area issues, 6*4=24, and 2*2=4 for area ratio).

But what isn't obvious is that the 6MP is just below (or with a border,
right at) what the 300 dpi printers can print (275 dpi full frame in our
analysis above). Conversely, what isn't obvious is that half of the film
image data is being thrown away in the digital printer.

If the 275 dpi of the 6MP camera is just sufficient to make an 8x10" print
with a border, than the 392 dpi of the 24 MP equiv. film data would have
to be reduced in HALF when printed as an equivalent 8x10".

On the other hand, we can also say that the digital print at 5.6 lpmm
falls below the 8 lpmm of the leica standard, and that "photo-realistic"
printers at 300 dpi also fall below the leica standard, and that a good
wet print at 8 lpmm will have roughly four times the info (area, or 2.04
times the linear resolution) of the 6 MP print (for 24 MP starting equiv.)

This is why wet prints look more details than so-called "photo realistic
prints" at 300 dpi, because they have higher 8 lpmm potential printing
quality. On the other hand, older eyes (ahem) often limit out at 6 lpmm or
so, so if you can't see the difference today, younger eyes or your own
eyes some years ago might have seen the difference between 6 lpmm and 8
lpmm. ;-) If you relax the printing standard to 5.6 lpmm (as in 300 dpi)
then you can get larger enlargements by a modest 2x area or so with a film
print (at 24 MP).

If you use really superb lenses, good technique, tripod, and slow 100 ISO
film, you can put circa 107 lpmm on film (as per popphoto tests, which
equates to 40 MP). In such cases, you can probably squeeze another paper
size out of your enlargement over the 80 lpmm case for 24 MP equiv. for
100 ISO films ("at least" per Kodak), i.e., 11x14" or so ;-)

Hopefully, none of this will contradict either David's or rafe's or anyone
elses considerable experience in printing and testing lenses and all that.

Digital has picked 300 dpi as photo realistic for the reasons noted; aging
eyes can't see much difference between it and the previous "leica
standard", but it means a 6 MP camera can do a nice 8x10" as all have
noted. But a 100 iso film shot can do an even nicer 8x10", with twice as
much data on the wet print, including much higher resolution and contrast
which is lost on the digital print due to the anti-aliasing filter in
DSLRs which typically limit you to around 40-50+ lpmm (42 lpmm in the 6 MP
case here).

again, my $.02 worth of electrons

bobm
--
************************************************** *********************
* Robert Monaghan POB 752182 Southern Methodist Univ. Dallas Tx 75275 *
********************Standard Disclaimers Apply*************************

hi Mike ;p-)

the canon 10d is a 6.3 MP with 22.7x15.1mm (1.6X) sensor if sources are
right ;-) You can expect a maximum of circa 68 lpmm resolution. So I have
to assume values over 68 lpmm being reported (e.g., 75 lpmm) are results
of post-sensor processing or subjective interpretations of test charts?

Yes, you are right that a smaller sensor size than full format (24x36mm)
means a higher resolution value, which puts more demands on lenses used
with the DSLR etc. It does NOT however mean more pixels, since you still
have 6 MP (actually 6.3 MP for this specific case).

So regardless of the sensor physical size, you have 6 megapixels. When
printed at 8x10" at 300 dpi you still have 5.6 lpmm and all that in my
previous analysis. And the film camera still has a 2X advantage, not
because of its sensor size, but because it is effectively a 24 MP sensor
equiv. (for the full 24x36mm frame). So comparing a 24 MP (film) sensor
and a 6.3 MP sensor, the on print factors remain as I analysized, yes?
;'-)

Now if canon could get a full frame sensor of 24x36mm at that same
density, they would have a circa 16 MP sensor (!). Again, this is
something I have been saying all along too. That 16 Megapixels is the
"sweet spot" for DSLRs, and when cheap 16 MP full frame sensors hit, the
results will be "good enough" for nearly all 35mm film users.

The other thing your post points out, with the above analysis, is that we
evidently agree that it is going to take some rather good lenses, at these
high densities, whether just the central lens area with the current 6.3MP
smaller sensors at these densities (1.6X factor), or with the future 16 MP
version at full 24x36mm frame size.

And to make it relevant to the MF group, you have to have 4X again for
MF's advantage in film, which is why a 64 MP sensor will be needed to
equal mid-speed films in MF (6x6cm). Right?

This is why I object to the claims that the latest 8.2 MP DSLRs are
"medium format" killers. Now 8.2 MP is nowhere close to 64 MP in my book
;-) Are 1.2 MP webcams good enough to challenge today's 8.2 MP DSLRs? No,
for the same reasons ;-) But the 8.2 MP will turn out pretty good 8x10"
and even 11x14" for most users, so they may be "good enough" ;-)

my $.02 again

bobm



--
************************************************** *********************
* Robert Monaghan POB 752182 Southern Methodist Univ. Dallas Tx 75275 *
********************Standard Disclaimers Apply*************************

"Bob Monaghan" wrote in message
...

Hi again Mike ;-)

No, I'm going the reverse way. I am saying that 6 MP is interpolated to
make a decent 8x10" print, and that the underlying data is circa 40 lpmm
....

Hi, Bob. Yes, I'm good with all that, even PopPhoto's 100+ lpmm. No problems
with that here. Farther in the thread, I think we individually satisfied
ourselves that these are all reasonable numbers, and matches our personal
experiences.

How can that be? The total system resolution is given by:

1/R = 1/R1 + 1/R2 + ...

Reasonable ranges for the combined resolution for the enlarger + paper --
about 60 lpmm to 80 lpmm -- gives results that are in line with what I've
seen. It also matches very well with what I see in film scans: about 3400
dpi is required to match that same total system resolution.

In short, all three systems -- 6 MP dSLR, scanned 35mm, and optical prints
from 35mm -- produce roughly equal quality results when printed at 8x10.
Optical prints on a well aligned enlarger will do slightly better, and
direct digital trails slightly behind very carefully scanned film. I find
the convergence of accepted industry data, information theory, and personal
experience quite satisfying.

What do you think? Is this horse well and truly flogged at last?

Hi David,

yes, from a parallel posting I noted:
the canon 10d is a 6.3 MP with 22.7x15.1mm (1.6X) sensor if sources are
right ;-) You can expect a maximum of circa 68 lpmm resolution. endquote

so 48 lpmm allowing for anti-aliasing filter and lens systems resolution
calculations (and that we are using only the center of high end canon
lenses ;-) makes getting 48 lpmm starting with a sensor spaced to get 68
lpmm equiv. seem reasonable. As I noted, claims of 75 lpmm may reflect
post-image processing by in-camera software, or subjective chart reading
or ?? I generally state 40-50+ lpmm due to these sensor anti-aliasing
issues as the effective DSLR system resolutions, so this is reasonable ;-)

Some context - I have noted how hard it is to break the 50 lpmm "barrier"
Erwin Puts, the noted leica lens testing guru, has suggested that 40 lpmm
is even more likely a barrier past which few amateur photographers reach,
unless they master good technique (tripod etc.) and careful (slow) film
selection etc.

So it isn't revolutionary to me that lots of folks find digicams and even
pricey 6 to 8 MP DSLRs are delivering - at 42 to 48 lpmm - pretty decent
photos comparable to what they achieved with film cameras in the past.
They weren't beating 40 or 50 lpmm in the past, and they aren't doing much
worse with digital photography, so why not be happy? ;-)

With good technique, tripods, and slow 100 ISO film and fixed lenses, they
could probably have beaten 60+ lpmm (at least centrally). And with better
lenses, then 80 lpmm and beyond are possible with slow films ;-) But maybe
5% of amateurs that I have seen are using a tripod, so the vast majority
are understandably happy with 40-50 lpmm equiv. achieved with digicams ;-)

But that also doesn't help those of us shooting MF for more quality ;-( We
aren't satisfied with 40-50 lpmm handheld shots on 35mm SLRs with fast
400-800+ ISO films and slow zone AF etc. ;-) So for us, it should take a
64 MP sensor to make us as happy as the 35mm full format 16MP DSLR crowd
at 60+ lpmm (on film equiv.).

again, my $.02 ;-)

bobm
--
************************************************** *********************
* Robert Monaghan POB 752182 Southern Methodist Univ. Dallas Tx 75275 *
********************Standard Disclaimers Apply*************************

quoting:
Son how much 35mm film have you shot ? Maybe if you shot the
10,000+ pictures I have on 35mm film you would see it differently.
endquote:

careful, lots of the pros and many of the serious amateurs on here have
shot 10,000 photos on 35mm this last year (only what, one 36 exp roll per
day), some in this last month, and perhaps a few in the last week or so
;-)

I have finally started to wear out some of my favorite lenses and replace
them. But then, I know of pros like Fred Whitlock wearing out three Nikkor
AF lenses. Since they are pro nikkors, they ain't cheapies, and since they
are AF lenses, they aren't ancient. And he specifically noted that they
weren't abused but just wore out ;-) So how many photos do you think he's
taken? ;-0) ;-)

grins bobm

--
************************************************** *********************
* Robert Monaghan POB 752182 Southern Methodist Univ. Dallas Tx 75275 *
********************Standard Disclaimers Apply*************************

Hi Jack,

yes, a very good point. However, sellers are not necessarily MF users,
just someone who has an MF camera to sell. In many cases, it may have been
inherited from an older relative, who wasn't actively using it. In other
cases, it was discovered abandoned in an attic (cf. folders, TLRs etc.).
Some of it is military surplus (e.g., koni-omega) no longer used there.

And you also point out, correctly I think, that some MF users selling MF
gear on EBAY are also going to continue to use MF, just use the dollars
from selling their extra MF kits and lenses to raise cash to go digital
etc.? ;-)

And some of us just want to sell to try the other guy's stuff, always
greener on the other pasture? ;-)

But the net effect, I think, is that a lot of "dormant" MF cameras and
items which were NOT in use, in closets or attics or inherited etc. end up
on EBAY.

Also many stores going out of business end up putting some stuff on EBAY.
An odd example is I bought one of my 45mm super komura very wide lenses
for bronica (for stereo work on 2 S2A bodies) from a Hallmark Card Shop
for circa $100. I asked the lady how the heck she had such odd-ball lenses
to list on Ebay in her card shop stock. Turns out the camera store closed
and left behind what they couldn't sell (!) and she moved in and had to
sell it to make space :-0) So there is also a source of gear from used
shops and pawn shops which haven't been in use either for years, yet which
when sold on Ebay can go to current users, including potentially new users
;-)

So yes, I think a lot of this used gear at low prices is attracting buyers
who would otherwise buy new gear at local camera shops in the pre-EBAY
days. And yes, some business schools are studying the impacts, as they are
major, of EBAY on marketing and pricing models and all that ;-) But all I
have to do is compare prices today against what I recorded on my website
for the past (see http://medfmt.8k.com/bronpg.html and mf/kowapg.html for
examples). It is quickly clear that current MF prices for used gear are
often half of what they were four or five years ago.

regards bobm
--
************************************************** *********************
* Robert Monaghan POB 752182 Southern Methodist Univ. Dallas Tx 75275 *
********************Standard Disclaimers Apply*************************