Friday, September 13, 2013

The Final Word- Film Vs. Digital

Here now I offer you what should be one of the most fair assessments of "35mm Digital vs. Film" Photography. This is an opportunity for which I'm grateful to have had all the right equipment and software to be able to provide this comparison, and to make it as scientific as a non-scientist like myself can be. I would also like to mention I have taken into account some good comments I received from earlier comparisons I had presented, and shot accordingly.

The Equipment:

For Digital, I used a new (to me) 2006 Vintage Canon EOS 5D, original 12.8 Megapixel model, set for RAW + JPEG capture. For Film, I used a Y2K Vintage Canon EOS Elan-7 with Kodak Ultramax 400 film. The lens used for all but the first pair of shots was the Canon EF 28-105, f 3.5 - 4.5 USM Version 1 Zoom, of which I conveniently have two, so that I didn't have to swap lenses between cameras all the time. The first pair of shots was taken with the Canon EF 40 f2.8 STM. The settings used on both cameras was identical - 400 ISO, Evaluative Metering, Aperture Priority, and making sure Aperture and Shutter Speed settings were close enough to being the same for both cameras, and as close to the same focal length (zoom setting) as possible, although some minor errors are seen in some of the pairs of shots. For Film, I used my own Epson V500 set at 300 DPI for an image size of 11" x 17".

The Post-Processing (PP) approach was to use Photivo (Open Source Lightroom clone) for both the RAW output from the digital camera and the 16-bit TIFF output from the film scanner to get the pictures to look as good as possible (not as similar as possible), with no regard to colour adjustments. The parameters which I tweaked in Photivo were Lightness, Gamma, Local Contrast and Noise Reduction, although no noise reduction was necessary for the EOS 5D. In some cases, I would use one step of Dynamic Range Compression only if I found it helpful in either case. Once completed with Photivo, I created a new TIFF file, and then used GIMP (Open Source Photoshop Elements clone) to convert both the RAW and TIFF files down to a screen-friendly 8-bit JPEG, and to rotate and crop as necessary. GIMP was not used to make any further lightness, contrast or colour adjustments.

So, on with the show. If you look at every one of these in quick summary, you will see that with film, there is an obvious colour shift toward deep blue. So this is a standing comment that when comparing 35mm Digital to 35mm Film using Kodak Ultramax 400, Digital is the clear winner as far as colour accuracy is concerned. I've talked about film's colour characteristics before. As to why I did not make some colour adjustments to make the film capture more true - the answer is that I don't know how to do this well, and actually, I thought it would be more fair to show this particular film with it's own characteristics, because with film, they're all different, even the same brands from one roll to another can have colour variances.

Pair 1:
Digital

Film
In this picture, the Plaque on the stone gate can be read more easily with the Digital - I actually used the plaque as a focus point with both cameras. Either the Elan-7 does not focus quite as accurately, the film has lower resolution, or most likely of all, it could be a result of what my budget model scanner does, especially with "Digital ICE" (dust removal) turned on. However, the film version reveals a lot more shadow detail beneath the top of the stone arch, showing how well the Kodak Ultramax 400 favours shadow detail.

Pair 2:
Digital

Film
With this pair, there is more than one problem with the film picture. Aside from the colour shift, there is a lot less sharpness, and the sky detail is completely lost. However, the reflections of light throughout the picture - in the car windows, wet pavement and especially underneath the canopy came off far better in the film shot.

Pair 3:
Digital

Film
Once again, sky detail is all but lost in the film shot, and colour saturation seems unnaturally high. However, the film provides a much better separation of all of the picture elements - look especially at the sense of space provided between the two trees. The digital shot has some of the typical lack of vibrancy, which the film version provides in bucket-fulls throughout the picture.

Pair 4:
Digital

Film
Exactly the same comments as above - leading me to conclude that the "Evaluative Metering" works a bit differently between the two cameras, which is a fair statement, as the Elan-7 is probably at least five years older than the 5D, and we are told that Evaluative Metering is constantly improving. The film version again shows a very high colour saturation, and as I said in the introduction, I did not want to try messing with the colour.

Pair 5:
Digital

Film
Here, I have to declare film as the absolute winner, because I love reflected light. In the digital shot, the hay-mow almost gets lost into the background, but somehow, the film caught the light from this rusty old machine far better. Also, there are nice light reflections throughout the film version that are almost completely missing in the digital - look at the green flower pot, and also the side of the white VW parked in the background. The digital shot isn't bad, but the film shot is a perfect example of why I love film.

Pair 6:
Digital

Film
Exactly the same comments as with Pair 5, giving the film version much more life and vibrancy. Unfortunately, the composition between these two pictures is quite different, but there's enough here to show how different the two mediums are.


Pair 7:
Digital

Film
This pair is quite neat, in that it shows how a digital camera works with great caution, with it's built-in computer, to give everything a "dull perfection", while a film camera simply captures the light that is thrown into it with wild abandonment! Yes, the leaves of the plant are more washed out with over-exposure, but that's fine because the sunflower is given a lot more "pop".


Pair 8:
Digital

Film

These two are quite close - the film version is of course showing higher saturation, and again, the sky detail is all but washed out completely. I believe that a simple saturation increase in the digital shot would've made these two virtually identical. But what else do we look for. This "something" that I personally refer to as "separation between elements" is a little better in the film shot - the actual space between the front row and second row of cars is more clearly seen in the film version I believe.

Conclusion:

With this exercise, I took as much care as possible, although I'm sure there are people who will point out some things I could have done differently - and suggestions / criticism is always welcome here. I'm going to offer a real crazy conclusion - that is, I haven't proven anything. I have shown that film and digital photography are quite different, but that was proven long ago, and my demo here does nothing to improve on that. I will give you my overall opinion however. Digital photos are always dead-on accurate in terms of measurable values - they're always "perfect", especially with a camera as great as the one I'm using here. Good digital pictures give us a kind of "lab reference" so to speak. Film is unpredictable - I was not expecting the blue shift that can be seen in every one of these pictures. I was also not expecting the wash-out of highlight detail - this film seems to prefer shadow detail at the expense of highlights. Also, some deficiencies in my scanning is noticeable - and that's the way film photos are processed now, even at the one-hour photo counter - film is digitally scanned and  printed, and you can either do it yourself, or have Walmart do it for you.

But to continue with my opinion, film is often better than digital in terms of "immeasurable values", with "separation between elements" (i.e. - sense of space) being one of a few of these sorts of things that I notice. Another thing I notice is the quality and quantity of reflected light - film always seems to do this better. These are things which I learned to do using specific techniques as a painter - there are "tricks" to create a sense of space, and to add light to a painting. This may explain how I look for such things in photographs and expect them to be there. By and large, digital photography fails to deliver on things like this when compared to film photography, but to be fair, these are things that most photography enthusiasts seldom look for.

I won't say "never again" - I may do more digital vs. film comparison, but I rather strongly doubt it. I'd much rather simply be learning how to  take better picture and using this Blog to talk about them. This I think would be a far better, but much more challenging use of my time.

1 comment:

  1. Nice comparison. I alwas wanted to do this, but I never had 2 cameras which shared the same lens mount and sensor/film size.

    One comment though. I am not sure that the actual exposure was the same fir bot digital and film. Even if you set the same values the sensitivity could be different of the two mediums. I feel that the digital images a tac underexposed against the film. That is why digital maintains the sky better while the film has better shadow details. Of course it also could be the result of your scanners exposure settings.

    Anyhow, nice post. Keep it going.

    Gabor camerajunky.wordpress.com

    ReplyDelete

Reader's comments are welcome, and are subject to moderation by the author.