Musing on Medium Format vs. Digital

With due apologies for asking a question that no doubt has been asked

many times:

 

I am a 35mm film shooter, using mainly Contax manual focus lenses and

the Contax G2 rangefinder system. Occasionally I shoot with a

Rolleiflex and love the results, but as a parent of an 18-month old

girl who has the misfortune of being my principal photographic

subject these days, I find that the Rollei, for all its charms, isn't

the most convenient tool, and for a long time I've thought that it

would be fun to use a hand-holdable, metered medium format

rangefinder like the Mamiya 6 or the Bronica RF. Even though my

photography has lately been all about documenting my daughter, I do

occcasionally manage to get out and shoot other things, and I'd like

to be able to enlarge my prints to sizes 8x10 and up. Great though

the Contax/Zeiss stuff I use is, it's still 35mm and I usually don't

find prints bigger than 8x10 to be all that satisfying.

 

For a while I have been wondering whether a DSLR, with all the

attendant advantages and conveniences like fast autofocus, dead-on

metering and compact zoom lenses, would be a better option than a MF

rangefinder (which is about as good as it gets for metered, hand-held

convenience in medium format).

 

I have never embraced digital photography, probably out of

ignorance. I was always deterred by the work-flow part of it -- I

don't have the time or maybe even the inclination to spend hours

Photoshopping, worrying about RAW or JPEG, calibrating monitors,

buying printer inks, struggling to make a decent B&W print and

pulling my hair out because of software installation glitches and

crashes -- there just aren't enough hours in the day for me. I would

rather drop a roll of NPH or Provia or Portra UC off at the lab.

Maybe even more importantly -- and this is where I confess to total

ignorance and fully accept getting flamed -- I have never fully been

persuaded by the "look" of digital, even shots from a good DSLR

properly printed. They have just looked off to me -- either

surrealistically sharp or weirdly colored but seldom appealing. I

know, though, that this can't be right, because so many photographers

who've forgotten more than I will ever know have dumped their film

cameras and enthusiastically embraced digital.

 

So my questions boil down to these:

 

1. Am I wrong to think that it's tough to get a "natural" and sharp

looking print -- color or B&W -- from a good DSLR without Photoshop?

I have read a lot about the new-ish Nikon D70, and a recurrent theme

seems to be that unless you do a fair amount of post-processing

having to do with things I don't understand like unsharp masking,

prints are going to be disappointing and maybe even soft; simply

printing straight from the camera won't do it justice, and you will

need to invest time and resources in learning to process your digital

images.

 

2. Assuming I am wrong and that it's not a huge challenge to get

nice, clean, natural prints from a good DSLR, how do the results

compare to medium format prints? I'm thinking of getting something

like a Bronica RF 645 rangefinder, but for about the same price I

could get myself a new Nikon DSLR and a nice kit zoom lens. The

metering, autofocus and zoomability of the Nikon are all advantages,

but can the finished result compete with a good MF print? Which

stands up better to enlargement? What about B&W -- can you get a

Leica/Tri-X look with digital? Qualitatively, how do they compare?

For me the holy grail has been to find a medium format camera with

all the conveniences of 35mm, but no such beast exists. Maybe the

DSLR really is the functional equivalent.

 

3. In the world of 35mm film photography, many people obsess over

whether one brand of glass is better than another and spend lots of

energy worrying about MTF curves, sharpness, optical characteristics,

you name it. From what I can tell, there seems to be less of this in

the digital context. Why is it (if it is indeed the case) that

people seem to worry less about marginal differences in optical

quality between, say, Zeiss and Canon L glass, when it comes to

digital photography? Is there some technical reason why differences

in lens quality are less significant in the digital context?

 

Look forward to your insights.

 

Thanks.

Have you considered medium format SLR?

 

I have a medium format SLR (Bronica) and my husband has a Roleiflex (rangefinder).

Rangefinders are not fast enough for me either (basically, i'm not fast enough).

 

I wont go into the "digital v/s film" war. I've owned both and choose film for various

reasons. Perhaps an SLR is what you're looking for?

I shoot MF 6x7 along with a Canon 10D, and am willing to brag I've been able to push each system to it's theoretical limits. I've also found many of the controversial web-site comparisons from both side of the fence to be basically irrelevant. Most of them dwell on essoteric and juvenile resolution comparisons that have nothing to do with using either system that then get debated here by a lot of individuals who don't shoot either MF or own a dSLR.<P>

 

Unless you are doing you own darkroom printing from your MF film the main advantages I find with my 6x7 are (a) easier wide angle applications, and (b) slide film. In terms of (a), nobody will dispute that dSLRs, with the exception of possibly 1Ds, have a significant problem with wide angle applications because of the small sensor. Depending on how much landscape work you do, this may/may not be a problem. As far as I'm concerned, the 50mm equivelant on my RB *does not exist* in the dSLR world with the same level of quality, and this includes Canon's full size sensor cameras.<P>

 

(b) is a lot more controversial. The biggest comlaint I have about dSLR capture quality is it's poor handling under low contrast situations. In this respect a dSLR has a similiar tonal response to low contrast print films, which also have lousy performance under low contrast (overcast skies). For instance, an overcast day with perhaps a bit of drizzle use to be my favorite conditions for outdoor macro work because I'd simply load up with Provia or Velvia in my RB, and the resulting trannies would just glow when given a quality scan. However, I've found these conditions to be horrible for landscape work with my 10D, just like using print film. When the sun comes out though, it's a whole different story. <P>

 

If you primarily shoot print film in 120, you'll be happy with the quality of dSLR capture and the slide film scenario is a bit irrelevant. It's also irrelevant if you aren't doing your own slide scanning in the first place since dropping off a roll of Provia at the lab amounts to nothing except getting back a roll of 120 slide film you still have to scan to do anything with.<P>

 

Portraiture or studio or casual shooting is where I feel my 10D puts my RB in it's place. Once I figured out the right preset combinations, <b>I can produce superior portraits with dSLR capture than film - period.</b> Dropping a roll of NPH off at a lab isn't a comparison in quality - it's an excercise in chaos theory because no lab can deliver consistent proofs as well as submiting your own digital files. Many/Most of the anti-digital posters here need the lab to think for them because they can't make their own color corrections, yet they won't admit it. I have to go a lot bigger than 8x10 with portraits to show any advantage with my RB, and even with the quirky ergonomics of dSLRs they are still easier to handle than any MF system I've used, and I don't have to scan film. Scanned Portra UC 400 in 645 is no match for my 10D.<P>

 

Your 35mm/Leica question doesn't make any sense. If you actually prefer the 'look' of smaller formats to larger ones, I can't help you there.

<i>Am I wrong to think that it's tough to get a "natural" and sharp looking print -- color or B&W -- from a good DSLR without Photoshop?...unless you do a fair amount of post-processing having to do with things I don't understand...</i><p>

 

You can't get a good print from film without a fair amount of post-processing either, it just comes down to who is going to do the work. If you really want to control the end images and produce the best prints, you either have to learn to use Photoshop or the darkroom, or pay a lot of money for someone who has learned those skills.

Nobody can tell you that the POTENTIAL print quality from a DSLR in the price range you are talking about is going to hold a candle to the quality of a MF camera. But, at 8x10, I seriously doubt you will tell a difference between a good image from a DSLR and a MF camera. I just don't think our eyes are sharp enough to detect improvements over about 200-300 dpi images (what you will get from your DSLR prints at 8x10). 20x30 prints will be a different story, though.

 

One of my good friends used to tell me that he didn't like the look of digital portraits until I shot his family with both film and digital. Then, he decided digital was okay. Do you know anybody with a DSLR that you could borrow?

 

About the workflow - to get the best from your DSLR camera, you will probably need to do a little bit or work in the computer. But, if the camera you buy comes with good software (the Digital Rebel does, I don't know about Nikon cameras), it doesn't have to be that bad. You can make across-the-board (batch) adjustments to all the files in a folder. So, for example, if, for a particular shoot, you wanted to emulate portrait film, you might tell the software to keep saturation neutral, and bring contrast ans sharpness down a notch. Then, just say "go" for all the pictures you just took and the computer runs off and does its thing. You then burn the folder to a CD and bring it to your favorite photo shop for prints.

Cliff,

 

Re #3: Digital resolution is lot smaller than film so the "superior"quality of lens doe not matter much. However, the sensor always lays flat as opposed to film, so you do not have the worries of not getting the maximum out of the sensor.

 

If you are interested, I have a complete set of Bronica ETRs (very used)for sale. :)

 

Vivek.

<I>Am I wrong to think that it's tough to get a "natural" and sharp looking print -- color

or B&W -- from a good DSLR without Photoshop?</I> No you are not and yes you are.

<P>First the "no" part and then the "Yes" part. it does take work to undlearn and

understand what you are doing and how to do it and to set up your system. Once you have

learned the skills and crafts involved it gets much easier. therea re various plateaus you

can stop at. the big thing with digital is that you now take on all ofthe responsibilities &

skills you formerly relied on a lab for. I naddition isf you are shooting NEF or other flavors

of "raw' to a large degree you also become the "film manufacturer" as well and you can do

this on a frame by frame basis if you choose. That is both the bad and the good news.<P>

I like the Nikon D70 9and other 6mp class cameras, but in terms of sheer image quality it

is no match for a well crafted medium format image. But we are comparing apples and

oranges here. Currently what is the bottom line in a DSLR that will take you close or into

medium format class? Maybe a Canon EOS 1D mark II. Either a Nikon D2X, Kodak Pro SLR/

n or SLR/c, Canon EOS 1Ds or EOS 1Ds mark II will certainly get you in to the m"medium

format' quality ballpark, but any of those will cost you around 3x the price of a D70 kit.

Since this is one of the well-known 'hot-button' issues with a lot of emotional investment among the adherents, it is all the more important that we have some objective information on what kind of quality can be expected from the various formats.</p>

A quick search of the Internet can yield several sites that can shed some light on the truth of the matter.</p>

<a href="http://truckgenerator.com/subdomain/sueandneal/digital_vs_film_essay.htm">Digital compared to film resolution Essay</a><P>

<a href="http://medfmt.8k.com/mf/filmwins.html">Monaghan on Film and digital</a><P>

<a href="http://www.medfmt.8k.com/mf/lenslpm.html">Monaghan on Film Resolution in lpmm</a><P>

<a href="http://www.clarkvision.com/roger.photo/scandetail.htm">How much detail can be captured?</a><P>

It totally depends on what you're doing. For me, both medium format and digital produce less satisfying pictures than 35mm film. That's only because my photography involves being able to shoot freely and frequently with an easily portable camera (not the best role for MF) and it also requires insane exposure latitude (not the advantage of digital).

 

But those are just my priorities, and a lot of people do great with other priorities and other gear. Go ahead and try everything, and see what works best for the pictures you like taking. That's really the only way you're going to know the answers to your questions.

"Since this is one of the well-known 'hot-button' issues with a lot of emotional investment among the adherents, it is all the more important that we have some objective information on what kind of quality can be expected from the various formats."

 

Meryl, I'm a bit confused after looking at your links. I thought you were going to post *objective* information?

 

The first link was full of the author's misconceptions and ignorance about film, digital, resolution, and photography in general. And it had absolutely no comparison photos, the mark of a site created by a film fanatic who substitutes for a lack of photography skills by writing anti-digital essays.

 

The second link was pretty much more of the same crap. Aren't any of these anti-digital zealots *photographers*??? Can't any of them take pictures?

 

The third link (still no pictures) has lots of 1000:1 contrast lpmm numbers. I have to admit that if you shoot monochrome subject matter at 1000:1 contrast, film is your best bet. If you shoot color subjects in real world contrast, the numbers get cut by half or more.

 

And the fourth link: we finally have pictures!!! Only the digital crops are "simulations" using a low rez digicam at "different distances". Sigh...

 

Here are some links with pictures by photographers who actually have enough confidence in their skills to throw a couple cameras on a couple tripods, shoot a scene, and compare the real world results.

 

10D vs. Film:

 

http://www.sphoto.com/techinfo/dslrvsfilm.htm

 

http://194.100.88.243/petteri/pont/Pontification/m_Aesthetics_Shootout/_Is_slide_better.html

 

http://www.normankoren.com/Tutorials/MTF7.html

 

http://www.jamesphotography.ca/bakeoff2004/scanner_test_results.html

 

http://www.jamesphotography.ca/bakeoff2004/10d.html

 

D100 vs. Film:

 

http://www.borutfurlan.com/test_results.html

 

1Ds vs. Film:

 

http://www.luminous-landscape.com/reviews/shootout.shtml

 

http://www.photographical.net/canon_1ds_mf.html

 

http://www.photographical.net/canon_1ds_35mm.html

 

http://www.photographical.net/canon_1ds_prints.html

Scott's post is the most useful here. A real photographer telling you the pros/cons from real world experience.

 

"1. Am I wrong to think that it's tough to get a "natural" and sharp looking print -- color or B&W -- from a good DSLR without Photoshop?"

 

Way too much is made of the post processing involved in digital. I'm perfectly comfortable shooting JPEG's and using my 10D's built-in sharpness/saturation options when I want a fast workflow. Working this way I merely have to plug my camera into a printer and print.

 

For more serious work I do shoot RAW and use Capture One and/or Photoshop. But even working this way I can pick a shot, process it, and have a perfected 8x10 off my printer in <5 minutes. I set levels, run a quick USM, maybe play a bit with saturation or color balance. Coming from a film scanning background I'm often amazed at how quickly I can prepare and print an image. The focus of my work has really shifted from the computer to the field. If I captured the scene correctly in the field, the work at home is cake, and I'll take it over film scanning or a darkroom any day of the week.

 

"2. Assuming I am wrong and that it's not a huge challenge to get nice, clean, natural prints from a good DSLR, how do the results compare to medium format prints?"

 

Re-read Scott's post.

 

"3. In the world of 35mm film photography, many people obsess over whether one brand of glass is better than another and spend lots of energy worrying about MTF curves, sharpness, optical characteristics, you name it. From what I can tell, there seems to be less of this in the digital context. Why is it (if it is indeed the case) that people seem to worry less about marginal differences in optical quality between, say, Zeiss and Canon L glass, when it comes to digital photography?"

 

I'm not sure how to look at this because I'm not sure who you're referring to. On one hand it seems pros are more concerned with the glass in front of their DSLR's. Reichmann has sold some pretty good lenses on the basis that they couldn't cut it on his 1Ds.

 

On the other hand there's a subculture of photographers obsessed over minor differences in lenses and equipment not because those differences add up to better images, but because they can't make decent images to begin with. These are often the same types bashing digital over absurd numbers they got with a calculator and a half understood theory (see any of Meryl's links). The working pros producing good images never obsessed to this level to begin with. They will certainly choose a better lens, but they're not going to spend hours photographing newsprint to prove to some kids on photo.net that X Leica lens is "vastly superior" to Y Canon lens.

 

So I guess it just depends on who you're referring to when you say "many people".

sorry about that, a Roleiflex is a TLR

Why do digital SLR users care so much about what I shoot? I'm happy with the results from my film cameras. And I'm glad that you are happy with your DSLR's. Just don't try to convert me. You sound like religious zealots. I know. There are plenty on the film side, too. Just take a deep breath and try to separate your identity from the brand or technology of your camera. That comment is for the zealots on both sides of this "debate."

I have the D70 and use prime lenses with it. I always convert the raw with the auto sharpening option and the prints are excellent throughout. This camera produces files which only require a levels (contrast) adjustment to get a great print. No sharpness problems to speak of, except with 50/1.4 wide open and that sort of things.

 

Film requires about 10 times longer post-scan processing to get a decent print in many cases. Your worry is imho completely off the truth wrt. the D70.

<i>Why do digital SLR users care so much about what I shoot?</i><p>

 

I don't think anyone cares. From what I can tell, everyone here was answering the question that was asked. What does that have to do with what you shoot?

Keep in mind that most comercial processing nowadays is from Frontier machines. These machines take a digital picture of negatives and use lasers to write to triditional photo paper at 300 dpi.

 

Film will reproduce detail at 80 to 200 line PAIRS per mm. It takes two rows of pixels to reproduce a line pair and usually at least 3.

 

Therefore 300 dpi is about 4 line pairs per mm.

 

Most people consider 6 line pairs per mm "sharp" and beyond 30 lp/mm human vision isn't capiable of telling the difference.

 

Long story short:

 

If you use most modern comercial labs to process medium format film to 8x10 prints you won't be able to tell the difference between what you shoot with medium format and what you shoot with 35mm or most modern digital cameras.

 

P.S. believing scientific data over opinions is not "digital bashing".

 

http://www.fbi.gov/hq/lab/fsc/backissu/april2002/swgitfield1.htm

"Film will reproduce detail at 80 to 200 line PAIRS per mm."

 

Color film will do no such thing under real world contrast. Sorry, but telling people they can get 200 lpmm on film is a flat out lie, and I'm going to call you on it.

 

The best you are going to do in real world conditions is Velvia at 80 lpmm, though the 50% MTF point is still going to be lower than on a DSLR. Other color films like Provia 100F will top out at around 60 lpmm with a 50% MTF point much lower than a DSLR.

 

And that's on the film. To make a print you've got to pass that information through yet another analog phase. Which is one of the key reasons why digital shines. Once captured, a digital image can be faithfully transmitted and reproduced down to the last bit in the last pixel.

 

"It takes two rows of pixels to reproduce a line pair and usually at least 3."

 

You only need a sampling frequency of 2.2x to guarantee reproduction of the highest frequency you're after. 20 kHz CD audio has a sampling rate of 44.1 kHz, NOT 60 kHz. The same model governing digital audio (Nyquist) applies to visual systems.

 

"Therefore 300 dpi is about 4 line pairs per mm."

 

6 lpmm.

 

"P.S. believing scientific data over opinions is not "digital bashing".

http://www.fbi.gov/hq/lab/fsc/backissu/april2002/swgitfield1.htm"

 

There is NOTHING scientific in the article you quote. It's obvious the tech writer wasn't even a photographer, wasn't knowledgeable in Nyquist theory, and grabbed what little information he had from some other work on photography.

 

Believing an essay full of errors written by a tech writer over real pictures by real photographers with real world field experience is, indeed, "digital bashing".

"The best you are going to do in real world conditions is Velvia at 80 lpmm, though the 50% MTF point is still going to be lower than on a DSLR. Other color films like Provia 100F will top out at around 60 lpmm with a 50% MTF point much lower than a DSLR." -Daniel Taylor

 

Daniel, Tell me where I can find these DSLRs that have sensors capable of > 80 lp/mm? What is your opinion on the resolution offered by DSLR sensors in terms of lp/mm?

 

Thanks,

Vivek.

"Daniel, Tell me where I can find these DSLRs that have sensors capable of > 80 lp/mm?"

 

I don't know the resolution of all the various DSLR sensors off hand. The 10D has a Nyquist limit of 65 lpmm (slightly higher than Provia 100F), the 20D is some where around 75 lpmm. But these are also APS sensor bodies, which means there will be a bit more enlarging than with 35mm.

 

But comparing lpmm numbers is an oversimplification, as I'll get into in a moment.

 

"What is your opinion on the resolution offered by DSLR sensors in terms of lp/mm?"

 

My opinion is that lpmm is only one aspect of measuring how well a system records and reproduces images.

 

Of equal importance is the MTF curve of the film/sensor, which describes how well the line pairs are recorded. The 50% point is often quoted as a useful indicator of the threshold between really clear, useful details and fuzzy, not so useful details. If memory serves, Fuji rates Provia 100F with a 50% MTF of 45 lpmm in real world conditions. The 10D's 50% point is 60 lpmm. I don't recall Velvia's 50% point, but I do recall it was lower than a 10D's.

 

Then there's grain/noise. Those who like grain not withstanding, the more grain/noise, the lower the visual quality of the image. In my experience a 10D has less noise at ISO 400 than Provia 100F. At ISO 100 there is no discernable noise at common enlargement sizes. This is why people will comment that DSLR 8x10's look like MF 8x10's.

 

The printing stage is also very important. Once captured a DSLR image can be faithfully reproduced down to the last bit of the last pixel. But to print from film you must pass the film through a second analog system, either the lenses of a film scanner or the lenses of an enlarger. Either way you lose MTF, probably clip off some of those high frequency lpmm, and possibly even introduce more noise. And that's if everything is perfect.

 

At the end of the day I would rather use my 10D for 99% of my shooting than my 35mm body or even my TLR. The few times I do drag out one or the other are for additional wide angle options, or to use Velvia, probably in the same conditions Scott would. This is because all things considered, the 10D is a much higher resolution system than 35mm film.

Uhh, Jeff, you are answering a rhetorical question.

 

For non-native English speakers, the point is that people seem to want to convert others to their chosen medium. Who cares? You can get nice results from either film or digital. Or water colors. Or charcoal drawings. Use what you like!

I recently invested in a nice 6x6 system with three lenses for about $2000. Five years ago I could never have afforded such a nice set-up. I was seriously thinking of going with a DSLR but realized that one joy of photography for me is veiwing my transparencies on a light table and occationally projecting them. Keep in mind I photograph mainly landscapes and may only shoot a half dozen images over a weekend. I'd be happy to get 10 great images in a year. If it weren't for this style of photography, without a doubt I'd go digital.

 

If your serious about getting beautiful prints I think you really need to have some knowledge of Photoshop whether shooting film or digital. Of course you can always pay someone to do this for you but I find it takes the fun out of the creative process.

 

I imagine I will go digital when I can afford a digital body that is as well built as my Nikon F100. Furthermore I want to get the velvia look on overcast days and not be hindered when shooting wide angle images. I'm still learning about digital capture and have little experience other than using a Canon G5.

I posted some links but there are so many that it is impossible for anybody to have a definitive list. If it has stimulated the discussion so that we start taling about measurables instead of opinions then it has served its purpose.

 

Most of these discusions bog down because they degenerate into unsupported assertions hurled back and forth across the battlelines.

 

Perhaps, if we are going to be talking about resolution that we at least agree that lpmm is the standard unit of measurement that allows us to compare lenses, film and digital sensors directly.

 

Then we can say that a specific film has a certain resolution potential and that a specific digital sensor has a specific resolution potential as expressed in lpmm. No more apples to oranges comparisons!

 

You can, for example, demonstrate quite convincingly why a MF shot can be enlarged to a greater degree than a 35mm shot. We know the size of the MF frame and the 35mm frame and that the grain sizes of identical emulsions would be the same. It is easy to back up the anecdotal evidence (that you can enlarge MF more than 35mm) with real figures.

 

It is fine for people to have opinions and preferences but the people who can ALSO back up their opinions with figures are probably right.

Cliff -

 

I'll add my comments not as any sort of expert, but rather as a recent convert from the NPH/NPS film world to the world of digital and post processing, and as a photographing parent who shoots way too many pictures of his child.

 

The post processing is not bad at all. For me the main advantage is control - I choose the color balance, saturation, sharpness, etc, rather than the lab. I haven't gone the route of the calibrated monitor, ICC printer profile, home inkjet printer, Photoshop CS crowd. I'm working on a basic PC, using a copy of Photoshop 5.0 LE, and I'm able to do everything that I need to. A few trial print runs at the local Frontier lab, and I know what to expect my corrections to produce on the final print. I've read a few tutorials on the net about using Photoshop, and developed a basic understanding. I can now take a landscape shot and prep it for printing in about 5-10 minutes. I've got an 8x10 from my 10D on my wall that is every bit as good as my large prints from Provia and Velvia (printed by the same lab on the same machine). Post processing the family snapshots is much faster, about 30 seconds per print to do a quick set of simple adjustments.

 

Long story short, the workflow, postprocessing and printing are quite simple and don't take up too much time. I'll spend maybe 3 hours at the computer doing post processing once every 3 weeks for each print round.

 

As far as the quality of the results from digital - I cannot and will not speak to any sort of theoretical film vs digital or medium format vs digital debate. What I will say - my digital family snapshots are significantly more appealing to my eye than my film family snapshots. I'm getting crisper, more detailed, more saturated prints, that are noticeably different than the film shots in the same photo album. If I didn't like the increased saturation or sharpness, I can simply turn it down and reprint. My new digital workflow produces prints that I like more (and have more control over) than my film workflow did (which was to drop film off at a lab that hand color corrects photos for me).

 

My final thoughts...

 

Being the parent of a young child, I know how carried away you can get taking photos of your kid. Going digital has made me much more free concerning how many frames I take. It's great to be able to "work a scene" taking numerous photos from different angles and perspectives to capture the right expression. I find that my "keeper" ratio has gone significantly down, not because I'm shooting more carelessly, but because I have a higher standard of what a "keeper" is. With digital I now have the privilidge of keeping the one "great" shot out of 15 "good" shots after shooting 36 frames of my son in his high chair. And afterwards, I don't have to kick myself for spending $15 on film and processing, since it only cost me $0.27 for the print I want to keep. There's truth that taking more pictures doesn't necessarily mean you will capture the "decisive moment", but when you aren't worried about how many pictures you take, it gives you more tries at the "decisive moment". That freedom and the reduced costs associated with digital are the biggest benefits for my situation.

 

And as far as a medium format camera versus an SLR, I would recommend the SLR for the faster/lighter handling and more responsive frame rates and autofocus - I'm sure you know how hard it is to keep up with your daughter! I'd specifically recommend a 20D and Canon 16-35mm f/2.8 L zoom or 17-40mm f/4 L zoom plus a 50mm prime for the ideal balance of reasonable cost with high quality capture and and a fast responsive camera.

 

Hope this helps!

 

Sheldon

<i>Most of these discusions bog down because they degenerate into unsupported assertions hurled back and forth across the battlelines.</i><p>

 

I think most of them bog down because they have nothing to do with photography. <p>

 

I've met a lot of great photographers, some famous, many highly "successful," and not once have I heard any of them talk about lpmm or whatever (I confess to not reading 83% of this thread). Some won't even talk about equipment, except when they need something for a specific shoot, and most speak in general terms. <p>

 

This bogs down because it has nothing to do with photography.

"Perhaps, if we are going to be talking about resolution that we at least agree that lpmm is the standard unit of measurement that allows us to compare lenses, film and digital sensors directly."

 

Quite frankly, lpmm isn't sufficient as a measurement. It's like trying to determine which car is faster based on horsepower alone. My brother's truck has more horsepower than my Eclipse, but I guarantee you his truck will give out before I shift into my highest gear.

 

I would say you need a MTF curve (which would include lpmm as a component) at real world contrast combined with a grain/noise index. That would at least take care of the major variables affecting resolution in the camera, and it would also be relevant to real life photography. Problem is this still doesn't take into account printing, and film loses significant ground during the printing stage.

 

"It is fine for people to have opinions and preferences but the people who can ALSO back up their opinions with figures are probably right."

 

The problem with your links, and the reason I jumped on them, is because the figures are irrelevant to real world photography, don't tell the whole story, and they are incorrectly translated into megapixels by people who don't understand Nyquist. Generally film's resolution in MP is over estimated by 3-4x based on a couple simple mistakes. Skewed and incomplete figures are never right.

 

I realize I can be a smart a** about this, but it gets tiring. If the people writing these "film is superior" essays would bother doing some experiments with a DSLR, they would quickly realize how flawed their essays are. That might lead to more understanding on their part and a lot less clutter on the topic. But while they claim their theories are "scientific" they don't follow the scientific method at all. They are supposed to test their theories against real world experiments.

 

I would pay good money if Neal, for instance, would rent a 16 MP 1Ds mkII and test the theory in his FBI link that ISO 200 print film = 16 MP capture. Those of us actually using digital know first hand what the results will be, and it would be a joy to never have to see that stupid link again. If we could get away from crap like that then maybe these threads would be filled with useful information, like Scott's evaluation of digital vs. slide under different lighting conditions. Or of wide angle lenses on 35mm vs. MF.

 

I would much, much, MUCH rather read stuff like that than have to correct people on 1000:1 vs. 1.6:1 contrast tests and the translation of lpmm into pixels using Nyquist.