Again: MF vs Digital vs 35mm

[a_petkov]

I know there are folks here who like to look at those tests, so here

is one more if you are interested:

 

http://www.ales.litomisky.com/shootout/analogversusdigitalshootout.htm

[marbing]

Nothing new here.

 

'Comparisons' such as this are a sort of a shell game. They shift around resolutions (resizing, sharpening, cropping) so nobody can tell what they are actually looking at...is it a comparison of film to digital OR is it a comparison of digital to a scanned copy of film?

 

This makes the results meaningless since they are NOT comparing what they claim to be comparing.

[nikos peri]

So much for rigorous test conditions... what is that POS lens he slapped onto the D200?! He

had presumably the 50mm he used on the Nikkormat. The distortion is nauseating.

[markplawchan]

..as they say, "Different horses for different courses"....I say shoot 'em all, (they do shoot horses, don't they?)

[mike dixon]

Their results are consistent with what I've seen from comparing 10" X 15" cheap digital prints from my 5D to 8.5" x 11" and 11" x 14" expensive custom, color "wet-darkroom" prints from 35mm and 645 negative film. The enlargements from 35mm film (100-speed) weren't even in the same league as far as grain and apparent sharpness (they weren't even as good as prints from a 6 megapixel digiRebel); the 645 prints were much closer, but the digital prints still held up better under close inspection.

 

I still enjoy the look and the process of making b&w prints in a traditional darkroom, but if your main concern is sharpness and fine grain, digital is ahead at this point.

[Ed_Ingold]

Meryl said, "is it a comparison of film to digital OR is it a comparison of digital to a scanned copy of film?"

 

It is perfectly legitimate to compare a film scan to a digital image, provided a good scanner is used. How else do you propose to display the results on a web page? If you can see the grain, how much more detail do you want? Most optical enlargements are not grain-sharp anyway, especially at the edges and corners.

[mario_mazariegos]

Yeah, right....

[jorge_jimenez1]

I certainly agree with Meryl. We cannot compare on a computer monitor the direct result on paper. The ultimate way to compare is a direct print from the negative on paper versus a direct print (without manipulation) of a digital image on to paper of the same subject at the same scale. In my experience, digital has eliminated 35 mm film from the equation, but it will take a few more years before it will replace medium format.

[mike dixon]

<i>The ultimate way to compare is a direct print from the negative on paper versus a direct print (without manipulation) of a digital image on to paper of the same subject at the same scale.</i><P>

If you're making large, high-magnification prints, you're likely to get a better result by using a high-quality, high-resolution scan of the film than you are by making a direct optical print. At high magnifications, losses in resolution/sharpness due to diffraction can significantly degrade the image; these losses don't occur with scanning. The quality of the final print from film is NOT being reduced by printing from a good drum scan.

[nigel d]

When these tests are run, why do they always compare crops of the centre of the frame rather than the edges?

[eric_brody]

I am impressed with the choice of a Tango drum scanner for the film tests and they seem valid; they pass the "sniff test."

 

It is really unfortunate, however that the tester chose to compare a good quality Canon prime 50mm lens on the 5D with a marginal quality high ratio zoom on the D200, when it appears, as another poster noted, that he had a Nikon 50mm f/1.4 prime available. This simple detail is critically important given that the lens is a critical link in the chain and weakens that chain irrevocably. The 5D vs D200 test is an apples vs oranges comparison.

 

I suspect many people are in the situation of considering a D200 at its attractive price point, with its excellent feature set, against the 5D, with its full frame high quality sensor. It is really a shame that this test is critically flawed.

[bob_the_builder1]

Wow, what a completely useless and biased test.

[blakley]

There can't really be a fair comparison between apples & oranges no matter what the methodology, so these are always going to be useless.

 

Given that this is the case, what I'd really like to see is a "challenge test" - like the ones Kodak used to determine "proper exposure" many years ago.

 

It would go something like this:

 

Pick 5 shooters of a particular film camera and a 5 shooters of a particular digital camera (I don't care which; let's say Hasselblad vs EOS 1DS mk II).

 

Give each shooter the following assignment:

 

"Produce the best life-sized print you can of a head-and-shoulders portrait of a blond, female human subject between the ages of 20 and 25 with normal exposure and a lighting contrast ratio of no higher than 4-to-1. The picture must include the top of the subject's head and the subject's elbows and must be printed on 24x36" paper."

 

Now mat all the prints in identical archival white mats and frame them in plain black wooden frames behind museum-quality UV-resistant glass.

 

Display all ten prints outdoors (but protected from weather) for 1 year.

 

At the end of the year, invite 250 randomly-selected individuals to an indoor, evening show of the ten prints. When they arrive at the door, hand them a scoresheet with 2 columns: in the first column, tell them to rank the ten pictures according to "quality of photograph". In the second column, tell them to rank the ten pictures according to "quality of print". At the bottom of the questionnaire, in BIG LETTERS, tell them that they can turn in the COMPLETED scoresheet for 5 "free-drink" tickets. Make sure you provide premium beer and good quality wine.

 

Analyze the results. You probably won't learn much, but at least you'll be judging pictures instead of pixels or grains.

[aaron_lind]

My hasselblad will still make mind numbingly sharp pictures no matter how many megapixels

digital cameras get. It will still be more fun to use. And, in three years, it will still be worth

more than half of what i paid for it.

[gary_yeowell]

Yes I agree, what a pointless load of s+++. The conclusion that the 5D is better than the Hasselblad assumes that as image makers we are only interested in resolution, and the overall appearance of the image is of no consequence. I own and use a Hasselblad and a 5D and by quite a large margin prefer the output from the Blad - simple. Previous to the 5D I owned a Phase One P20 for the Blad, and guess what, I preferred the output of film - simple.

[Spearhead]

<i>My hasselblad will still make mind numbingly sharp pictures no matter how many megapixels digital cameras get.</i><p>

 

I'd rather take great pictures than sharp ones.<p>

 

People should take a look at Lauren Greenfield's recent prints, that's a great test.

[aaron_lind]

The point of testing cameras isnt about taking great pictures, thats the point of photography.

[pics]

Whats the point? I don't see obtrusive grain in my 11x14's shot with MF and they look plenty sharp to me. If it takes a high power microscope to prove the superiority of one medium over another I'm not impressed. I guess we should discount all the great photographs taken decades ago on lower resolution/grainier film since I suppose they would perform rather poorly under these meaningless extreme crop tests.

[hugh_sakols]

Here is another site that compares MF to Digital that gives different results. I found this in

the medium format forum not long ago.

 

http://www.shortwork.net/equip/review-1Ds-SQ-scantech/

[scott levine]

My two cents for what it's worth. I never really used 35mm, so I will give my impression

based on Medium format and Digital. I have a D200 which I use with prime AIS lenses. My

Medium format cameras are a Mamiya RZ or ProTL. When looking at C-prints, I have to say

that the 18mb scan from either Medium format camera still has a better feel (more 3-d), than

the digital with twice the file size. For my eyes the digital tends to compress the image too

much and exaggerate color. Fleshtones and color just look better with film. Just my

opinion.

[todd_k.]

When did sharp pictures and great pictures become mutually exclusive? Wouldnt you rather have the option of producing great sharp pictures vs. being stuck with great soft pictures? You can always get a soft print from a sharp neg.

[mark_l3]

seems to me that if you are going to compare a digital original to a film original that all other factors must be equal. In other words, you must have a print from the digital image and a DIGITAL print from the film file. otherwise you are also comparing the digital print to the analog print.

 

It's been my experience and the experience of many others here that a digital print has the potential to be much sharper than a tradtional print.

[Spearhead]

<i>The point of testing cameras isnt about taking great pictures, thats the point of photography.</i><P>

 

See Daniel Lawton's comments.<p>

 

No great photograph resulted from testing cameras. So what's the point of the endless testing and comments on sharpness? Nothing. Look at Lauren Greenfield's work, like I suggested. I know, real photographs by a great photographer don't matter, but her digital work is just as incredible as her film work. And in the end, that's what matters. All the comments about having a sharp Hasselblad won't make one a great photographer, or even a good one.

[dglickstein]

Anastas,

 

Why didn't you include a pinhole camera in your comparison tests?

 

dG

[Brad_]

I think Robert Frank should have spent a little more time testing cameras. Those unsharp

and weak on contrast photos in his book <i>The Americans</I> probably killed any chances

of success he might have otherwise had.