OT: Resolution of film

"The fallacy in all the film vs digital "tests" that find digital equal or better, is that in all cases the comparison is made to film that has been scanned--in other words, digitized"

 

Jay, even though I use digital for almost all my colour work I still maintain a traditional black & white darkroom and still make hundreds of traditional silver prints every year. However, I get the clear sense that in making a selenium toned fibre print from a Leica negative I'm in a small and fast declining minority.

 

I agree with many of your basic assertions, but they've taken me to different conclusions. The photographic doctrine that I can't understand are those Leica users who scan all their negs and print only inkjets, but somehow believe they're the guardians of a noble photographic tradition that places them a cut above the common Digital Rebel herd. It's a nonsensical idea to consider such a hybrid anything other than what it is, 100% digital photography but with an awkward front end.

<i>Jeff you are hilarious in a pathetic sort of way. The assertion that real

photographers don't give a crap about sharpness and resolution is simply

your way of bolstering your own sagging self-esteem by demeaning everyone

else. Fortunately your antics are as transparent as your point is groundless.</i><P>

That wasn't his point. I've heard pro/"real" photographers discuss sharpness/

contrast/resolution/etc. as issues affecting the suitability of their images for a

particular purpose (something which Jeff alluded to in an earlier post)--I've

never heard them discuss these issues as a theoretical issue of whether "the

resolution of digital SLR's has exceeded film." You shouldn't talk about

sagging self-esteem and demeaning others in the same paragraph where

you grossly misrepresent what somone else has said in order to demean

them.

"The nature of film is to attempt to resolve finer and finer detail less and less clearly. Digital on the other hand destroys any detail that it can't present clearly."

 

Absolutely right. And bit by bit that uncomfortable truth is surfacing. Uncomfortable because in the initial euphoric rush into digital there were some absurd claims made regarding the degree of enlargeability possible from a modest collection of pixels. And also uncomfortable because if a digital photographer wants to successfully make the big prints that the latest generation of inkjets are facilitating then photography becomes a big money game.

<<The photographic doctrine that I can't understand are those Leica users who scan all their negs and print only inkjets, but somehow believe they're the guardians of a noble photographic tradition that places them a cut above the common Digital Rebel herd. It's a nonsensical idea to consider such a hybrid anything other than what it is, 100% digital photography but with an awkward front end.>>

 

Excellent point and right on the money. But it's getting harder and harder to find a lab that still does traditional enlarging and does it well. I've had to drop back and punt, so-to-speak, by having my negs high-end drum scanned and printed on photo paper.

 

<<I've never heard them [pros] discuss these issues as a theoretical issue of whether "the resolution of digital SLR's has exceeded film.">>

 

Then it's only because you haven't been listening. Every time I'm somewhere where working pros are gathered (pro shop, pro lab, convention) it's the #1 topic. Again, I'm talking about working commercial professional photographers, not would-be "artistes" living out of their cars hoping to sell a few prints at a street fair or at a little storefront gallery somewhere.

Perhaps it's a matter of sampling bias in the pros each of us is talking to. I'm

not talking to them at camera stores or trade conventions; I usually talk to

them at workshops, on commercial shoots, and as personal friends.

</i>

</i>

ok jay, u keep yappin, i'll keep snappin

i don't think anyone really answered wai-leon lee's first question... so here is my shot at it..

 

if you think of camera shake as a low pass filter... it basically cuts off the high frequency out...(and if you have the same camera shake on both cameras, you'll basically have the same resolution... it doesn't matter if you had a million dillion lpmm, it gets cut out at the cutoff frequency)

 

which renders both cameras on equal footing in the resolution sense.

 

but which offers a smoother grain free image?

 

depends...

 

and that makes this discussion so interesting and complicated...

<<Perhaps it's a matter of sampling bias in the pros each of us is talking to. I'm not talking to them at camera stores or trade conventions; I usually talk to them at workshops, on commercial shoots, and as personal friends.>>

 

That makes a lot of sense Mike. There are venues where gear is on everyone's mind, and others where technique is more at the forefront.

 

<<grant . , jun 27, 2004; 02:20 p.m.

ok jay, u keep yappin, i'll keep snappin>>

 

u r definitely a snapper grant.

You can usually lookup a manufacturer's published lpmm ratings for their films at their web site. Fujifilm.com lists this information in their PDF datasheets.

 

Film fans just *love* to quote the 1000:1 test results for film, while completely ignoring the average contrast/1.6:1 results listed on the same page. As if photographers spend all day using color film to shoot B&W test charts under lighting so extreme color isn't captured, just black and white lines. You can forget those numbers. You will never see that resolution outside of very specific lab conditions. For real photos, look at the 1.6:1 number.

 

This leaves Provia, one of the absolute best color films on the market (and a film I'm very familiar with), with a resolution of 60 lpmm. This is going to translate into 12.4 MP for a 35mm frame. However, in my experience I've never been able to use an entire frame of 35mm, the edges always have to be cropped a bit to make the corners square, bringing the image down to around 10-11 MP.

 

In other words, theory used with the *correct* numbers would predict that Provia 100F in 35mm captures no more than a Canon 1Ds.

 

The other thing film fans *always* over look is the MTF response of film vs. a digital sensor. This is every bit as important as the lpmm rating, if not more. Provia 100F drops below 50% MTF at 45 lpmm. A Canon 10D is still above 50% MTF response at 60 lpmm, just shy of its cut off point. (I imagine most, if not all, digital sensors have >50% MTF response up to or very near their cut off point.)

 

Last but not least, digital sensors have a huge noise advantage over film at any given ISO.

 

This is why digital P&S cameras with sensors smaller than 110 film produce decent 8x10's where 110 film can't produce a decent 3x5. This is why prints from a 6 MP DSLR appear sharper and more detailed than prints from a frame of 35mm film at any size. And this is why Reichmann prefers his 1Ds to 6x7 MF film. Film may capture a bit more detail, but the important details which make the photograph are captured with far greater clarity by digital sensors.

 

When I read some of the responses to a question like this I can't help but wonder if the responders have ever touched a DSLR, or printed a side-by-side test, even when they claim they have. I recently purchased an Epson R800 and have been reprinting some of the work in my portfolio because the R800 can faithfully reproduce some tones that my older Epson printers struggled with. (Over the next couple months I'll probably reprint all of it for archival reasons.) My portfolio is a mix of mostly 35mm film work and 10D work. As I sit with the printer running looking over my favorite shots, I cannot help but notice how much better, clarity wise, the 10D shots are. On close inspection *it's no contest.* The 10D shots win hands down by every measure. They look like prints from 645, not a 35mm body.

 

Viewed at a normal distance the differences are less pronounced, but you can still tell. Where is all this glorious fine detail that 35mm was supposed to capture? Some would say I lost it in film scanning, but I also have some traditional wet lab prints, and it's not there either. You can preach theory based on 1000:1 contrast tests all day long, but I have to agree with the Reichmann-types who say trust your eyes. The first 8x10 I ever printed off my 10D is still in my portfolio, and I still remember how it dropped my jaw straight off the printer. At ISO 400 it's cleaner, sharper, and *more detailed* than my ISO 100 film work.

 

Below are some sites which address film vs. digital capture.

 

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.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

 

<<You can preach theory based on 1000:1 contrast tests all day long, but I have to agree with the Reichmann-types who say trust your eyes. >>

 

Yes it's always easier to prove a false notion by subjective evalutation rather than quantifiable data. That's how those pro-digital sites "prove" that digital=film, that's how that moron "proves" the 1Ds=medium format, and that's how Erwin "proves" new Leica lenses are all better than the previous ones. Skip the math and go straight to "well if you can't see it you're blind".

"Yes it's always easier to prove a false notion by subjective evalutation rather than quantifiable data."

 

Jay, did you not read my entire post? The quantifiable data proves the notion that a 6 MP DSLR will capture almost as much detail as 35mm and do so with more clarity. The personal choice between slightly more detail or considerably more clarity is the human "trust your eyes" part.

 

"Skip the math and go straight to "well if you can't see it you're blind."

 

I didn't skip the math. I *started* my post with the math. You're just not happy that I pointed out that the "film is vastly better" crowd always does the math with the *wrong numbers*.

Figures don't lie, Jay, but liars figure. Or if not liar, dunces who do naive math based on not understanding what the hell they're doing and making simplistic, idealized or just plain false assumptions, like Martin's computation for Provia, for example.

 

And when the subject is enlargeability, whether you understand it or not, subjectivity what ultimately matters. There is no objective way to quantify tradeoffs in grain versus resolution, for example. And if somebody's standing in a gallery thinking your print looks like crap I suppose you're going to whip out your calculator and film data sheets and prove him wrong?

<i>u r definitely a snapper grant.</i>

<br>

<br>

thanx

Daniel: I remember a routine where Abbott proved to Costello that 2+2=5 and his argument was much more lucid than yours, and his conclusion much more sensible. In addition you have completely and no doubt purposely ignored the fact that all digital cameras' outputs have to be salvaged by software to come anywhere close to exhibiting the quality of film. One way or the other, either there's artifacting or soft pixels due to strong anti-aliasing filtration, it needs to be doctored in the camera or in PS.

 

Mark: So now Martin is a liar and a dunce? My acid-test for an erroneous assertion is when the defenders finally resort in desperation to personal attacks when intelligence and logic have all proved them wrong.

Martin is more along the lines of a nut. In any case, he statement is total crap, and the fact that you don't know that speaks volumes about your technical ability.

 

As, for that matter, does your inability to follow Daniel's quite clear line of reasoning. He's an engineer. You weren't smart enough to get into medical school.

All very technically interesting. Jay made one point that no other person made reference

to, drum scans. 4 x 5 drum scanned yields about a 310mb file.

 

Large format users must be laughing their heads off at all this.

Gigabitfilm has a resolution of 720 lp/mm at 1000:1 and 240 lp/mm at 1.6:1 contrast. It will take ages before digital sensor can match, may be never<p>

Just two questions, do you all use your M's in a studio? Or are you some sort of supermen who can set perfect exposure and focus in a fraction of a second?<br><br>

 

The resolution and sharpnes of the pictures I take with a D60 is mostly better than what I get from my Contax SLRs or G2, The prints are made in the same <a href="http://www.cewecolor.de/">lab.</a><br><br>

Maybe one time I have a negative worthy of a 200 Euro drumscan or a 500 Euro 20x30 wet print from a real master printer. <br><br>

Volker <br>

Volker Spider-Man 2 opens on June 30, 2004; he will answer all the film verus digital questions!

The main plot of Spider Man 2 revolves around Peter Parker's dilemma in choosing Film or Digital...

Based on Gary's test I prefer the film version. However that was 6x9. Does it take that much film to equal an 11Mpx sensor? Surely not(?).

The ultimate resolution film camera:General Manager of Zeiss Ikon Dr. Emanual Goldberg's (1881-1970) invention: microdot camera of 1926<p>

Dr. Emanual Goldberg reduced a full page of printed text into 0.1mm x 0.1mm size microdot with his special film and microdot camera, each letter only one micron in size.<p> In WWII, both Zeiss and Leitz made

microdot cameras.

I just wanted to differentiate myself from the other Daniel Taylor. I am getting too many emails regarding these silly threads from postings I never made!

 

I love a great photographic print. I don't much care how it was made, as long as the final result pleases me on some level. I also enjoy the photographic process, for so many reasons, that using my Hassleblad's or 4x5 is still fun and rewarding.

 

my current problem is living on a sailboat in preparation for a circumnavigation. without facilities to develop film, there is a tugging on me to find a digital solution beyond my Canon G2. something will be missing for me, but what a lovely notion it is to be in a South Pacific cove on the boat and view/print images made that same day.

 

fair winds ...

"Daniel: I remember a routine where Abbott proved to Costello that 2+2=5 and his argument was much more lucid than yours, and his conclusion much more sensible."

 

Jay, all you're doing at this point is mouthing off. You're not offering anything for consideration which would change what I said, no corrections or new information. That's because you have nothing to add. You know the numbers I cited come straight from Fuji, you know the calculations are correct, and you've got nothing to counter with except insults which are both childish and ignorant.

 

"In addition you have completely and no doubt purposely ignored the fact that all digital cameras' outputs have to be salvaged by software to come anywhere close to exhibiting the quality of film."

 

I ignored this originally because it's not a fact, it's a myth pushed by people who have never even picked up a DSLR. Without question my 10D captures are closer to being perfect straight off the sensor than my film captures, and I spend (a lot) less time in Photoshop on a 10D image than I used to on a film scan. Having the correct color balance from the start is a *huge* time saver, not to mention having *no* noise to contend with.

 

And please don't try to tell me that Photoshop time on film is due to the scanning. Making color wet prints is very time consuming and difficult, centering around color and dust control. And most of the tricks in Photoshop came straight from the darkroom.

 

"One way or the other, either there's artifacting or soft pixels due to strong anti-aliasing filtration,"

 

If there are any artifacts in a 10D RAW capture at ISO 400 or below, they don't show up except upon 100% inspection in Photoshop (i.e. equivalent to a 40" poster). You can't possibly compare the silky smooth output of digital to the "artifacts" of film, including grain and dust.

 

I have no doubt that off-the-sensor my 10D images are sharper than Reala, and probably comparable to the old Supra 100. Provia 100F and Velvia are probably a bit sharper before post-processing and printing. But that anti-aliasing filter can't be doing too much damage because even a modest unsharp mask will bring the 10D image to Velvia levels of sharpness, with no change in noise.

 

I have several images shot with my 70-200 f/4L that I just didn't bother sharpening prior to printing. There just wasn't a need.

 

"Mark: So now Martin is a liar and a dunce? My acid-test for an erroneous assertion is when the defenders finally resort in desperation to personal attacks when intelligence and logic have all proved them wrong."

 

Heal thyself, Jay. *You've* resorted in desperation to personal attacks because intelligence and logic has proven you wrong.

Karim; look at Gary Ferguson's images from about 3 or 5 ft away. (~0.9m to 1.5m). What is interesting is that from a distance; the digital image looks "cleaner". But close up; the film image has better resolving power; the wording is clearly readable; with no guessing; the bands in the red logo are resolved. An "artsy" type might prefer the digital; for a rock poster. A military aerial bomb dropper would prefer the better resolving power; to pick out micro details.