The great digital v film shoot-out... well, a bit of a test anyway!

Mr Badger

Mr Badger

Didymus
Messages
7,457
Edit My Images
No
A few weeks ago in another thread I came up with the idea of doing a direct comparison of two ‘prosumer’ type Canon SLR cameras; a latter-day classic film camera pitted against a full frame model from Canon’s current range, namely an EOS 3 and a 6D. These two cameras were launched onto the market some 14 years apart, almost to the month, with the EOS 3 being launched in September 1998, and the 6D in November 2012.

To keep this test as even as possible, a reasonably challenging indoor scene was chosen. This was illuminated purely by natural daylight, coming from windows on three sides and two skylights. A cloudy but bright day was chosen, so the light would be as consistent and comparable as possible during the test, and the photos were taken just a few minutes apart. The scenes had both light and dark areas to test dynamic range, as well as the cameras’ ability to meter correctly for the situation. Matrix type metering was selected for both cameras. Several shots of the same scene were taken using each camera (mounted on the same tripod), and the best looking image (comparing sharpness, even exposure, etc.) was then selected from each camera for comparison.

The same Canon EF 24-105 IS L lens was used on each camera body to rule out any difference in lens quality between the two sample shots. The lens was set to an aperture of f8 (with both cameras being set to 'AE' aperture priority mode) and the zoom length was kept as similar as possible (at around 32mm) for each of the two comparison photos, so the whole image could be used from each camera. The focus point chosen was the same for each camera, with the centre spot focus aimed at the corner of the oak dresser in the centre of each photo.

The EOS3 was loaded with Kodak Ektar 100 ISO colour print film, and the 6D set to 100 ISO. Ektar was chosen due to the fine ‘grain’ of the film. The image from the 6D was a JPEG file downloaded straight from the camera using Canon’s EOS Utility download software. No post production was carried out once the image was downloaded. The film from the EOS3 was developed by a commercial film lab and scanned by them at high resolution using a Noritsu professional scanner. This gave a closed file size of 13.2mb, and the closed file from the 6D was 8.85mb. The objective of this was to obtain the sort of results the average 'home user' of each camera might normally experience in everyday 'snapshot' type circumstances.


So which camera will take the crown, will digital triumph or will silver halide win the day? Let battle commence!

The first photo is from the 6D and shows the full frame photo downloaded from the camera:



The second photo is from the EOS 3, once again featuring the full size scan received from the lab:




So what do you think? Pretty similar, aren’t they? Perhaps it’s going to be a draw? Let’s go in a bit closer, first a screen snip from the 6D:



And now the EOS 3:




Hmmm, without the 'noise' there might be a bit more detail from the EOS 3 than the 6D? Nice colours too! Let’s try another one:

6D



EOS3




A win for the 6D for detail here… it looks like it's going to have to be 'best of three' to decide?

6D



EOS 3



The 6D clearly takes the detail crown in this last one.


Well, both cameras managed to expose the image well, and the dynamic range appears to be similar (as far as this particular scene tests that aspect), but when it comes down to capturing fine detail the 6D digital appears to win. So do we have a clear winner then? As with most things in life, it’s not quite that simple... Out of curiosity I printed both images out at 5x7 to see how they’d look. At that size the lack of fine detail isn’t obvious and I think the average person would be pleased with either of them. I also printed them off at A4, and the difference in fine detail still isn’t instantly apparent and I doubt the film print would attract many complaints from the average 'person in the street' (unless they put their specs on and had a closer look!).

However, when looking at both photos side-by-side from a 'normal' viewing distance, I think it’s the Ektar photo from the EOS 3 that looks the nicer of the two, those rich colours seem to suit this particular scene and in comparison, I think the result from the 6D looks just a little bit flat when placed next to the Ektar photo. It would be a simple job to change this in Photoshop or Lightroom, but that’s not allowed in this test!

So who takes the crown, digital or film? Well, of course we can’t tell from just one photo, it wouldn’t be a fair test. Also, we are comparing a scanned image of the film, rather than a direct print from the negative. I think a professional lab/darkroom print from the negative would probably reveal more detail, and if so, I wouldn't like to predict the winner then!

In this single ‘snapshot’ test the resulting image from the 6D shows noticeably more detail, but I think the inkjet print I obtained from the EOS 3 scanned image looks just as good, if not nicer, up to A4 (unless you look closely). It’s a shame I can’t show that effect on line, but despite computers and the digital age being an amazing invention, sometimes it’s just not quite the same as something physical… a bit like film perhaps?

I hope you've enjoyed reading this, feel free to debate or contribute to this thread as you see fit. If I don't get too badly mauled for doing this rather limited comparison, I might try a medium format TLR camera against the 6D next! :sneaky:
 
Good comparison, I think it may have been better if the film shots had been printed, as scanning the negs is bound to loose something along the way, still a goot effort, look forward to seing the 6X6 comparason.
 
Mr Badger said:
I hope you've enjoyed reading this, feel free to debate or contribute to this thread as you see fit. If I don't get too badly mauled for doing this rather limited comparison, I might try a medium format TLR camera against the 6D next! :sneaky:
Click to expand...

Nah, you're safe, we're just to damn polite to say what we really think...:)
 
As Dizidav said, a better comparison would have been with prints, the EOS 3 version being wet printed. As it is, we cannot tell what is a defect in the film shot and what is a scanning defect.
 
Thanks for your test...looks like the digi has the edge for quality of detail on blowups.
 
For me there isn't much advantage in shooting 35mm film vs digital. The real advantage comes from the medium/large format, it's not a resolution thing but the feel you get from larger formats.
 
Last edited:
so it sounds like, in your view, the film makes the better final picture print but the digital has more detail than the scanner. And based on that, digital wins. That's fine i guess if you're just interested in how much detail a piece of equipment can render, but for me, i got into photography for the images.
 
What's the actual physical resolution of the 35mm scan? Was it a TIFF or a jpg?
 
It is an interesting comparison, but very predictable in 35mm. Pretty much as I would expect.
 
I think I agree with your conclusion. Although the fine detail is more apparent in the digital shot, overall the film shot has a much nicer feel to it and at the small size I'm viewing it the wall in particular looks superior in the film shot.
Nice comparison.

Andy
 
srichards said:
What's the actual physical resolution of the 35mm scan? Was it a TIFF or a jpg?
Click to expand...

It was a JPEG. Basically, the test replicates what the 'average' person would probably get if sending a film for developing and a hi-res scan. I think that once we get into the territory of saving as TIFFs, drum scans, or pro-quality wet printing, then I imagine that in the interests of fairness we'd have to shoot the D6 in RAW and send the file off for professional digital processing to wring the best possible result out of that image too? I know this is the F&C section and we like film but I see no one has suggested getting the very best out of an image from the 6D yet?

Trouble is, the more variables we introduce, the more they could combine to potentially skew the end result, so we get to the stage that we won't know if its the camera, the film, the processing, the scanning, the printing, or the computer screen we are using that's influencing our own (and possibly subjective) judgement? Perhaps film can't reliably be compared with digital? It was quite interesting to try though.


AshleyC said:
so it sounds like, in your view, the film makes the better final picture print but the digital has more detail than the scanner. And based on that, digital wins. That's fine i guess if you're just interested in how much detail a piece of equipment can render, but for me, i got into photography for the images.
Click to expand...

I didn't actually pick a winner, and I pointed out that it wouldn't be a fair test to try to choose a 'winner' using just one photo! I also mentioned that the scene I used seemed, to me, to look a little bit nicer on the 35mm film photo version when printed out using my inkjet printer, probably due to the colour rendition of the Ektar film, which seemed to suit that particular scene. I also mentioned that it would be an easy job to adjust the digital image from the 6D to match, using something like Lightroom or Photoshop.

If anything, this simple 'snapshot' comparison shows just how difficult it can be to compare film with digital on a totally level playing field. Still, it was interesting to do, and I'm sure someone will be able to improve significantly on what I did. So which camera do I prefer? Both of them! I've been really pleased with the results I've had from the 6D and don't regret buying it for a second. I really like the 'eye controlled focus' system on the EOS 3 too, and would love to see that on a modern full frame digital SLR, particularly if it could be extended over a wider area of the viewfinder. However, if I could choose just one of them, it would have to be the 6D for the convenience digital brings, and its low light performance. I still love to use film though, and the 'feel' and atmosphere it can give.
 
Last edited:
Mr Badger, would it be possible to post a few pairs of images shot with the same bodies but without labelling them so a blind comparison between the end results could be seen? IIRC, "Save For Web" in PS(E) strips any EXIF out so people won't be able to see which is which for a fairer comparison.
 
Given that full frame is the largest sensor size that could be described as in common use (in the amateur world, anyway) and given that 35mm film could be described as the smallest film size in common use (very few I suspect use half frame or sub miniature sizes), wouldn't it be a more level playing field if the comparison was made against 5x4 film, which probably occupies the same size position in the film world? That is, larger is still possible, but comparatively few use the larger sizes.

Well done for making the test though :clap:
 
StephenM said:
Given that full frame is the largest sensor size that could be described as in common use (in the amateur world, anyway) and given that 35mm film could be described as the smallest film size in common use (very few I suspect use half frame or sub miniature sizes), wouldn't it be a more level playing field if the comparison was made against 5x4 film, which probably occupies the same size position in the film world? That is, larger is still possible, but comparatively few use the larger sizes.

Well done for making the test though :clap:
Click to expand...

Any debates digi ver film and the digi guys would only mention the big guns of digi cameras i.e. Medium format. :eek:
 
I thought this was interesting, though scanning with no sharpening seems to be widely described as an issue. Of course, we don't know what sharpening, if any, the lab did.

But to me, it's indicative of some of the reasons that some people use analogue film rather than digital that you could conclude the digital has more detail, but the analogue looked better! :) :) :)

Not for everyone of course. Some people really like pixel peeping. Each to their own!
 
StephenM said:
Given that full frame is the largest sensor size that could be described as in common use (in the amateur world, anyway) and given that 35mm film could be described as the smallest film size in common use (very few I suspect use half frame or sub miniature sizes), wouldn't it be a more level playing field if the comparison was made against 5x4 film, which probably occupies the same size position in the film world? That is, larger is still possible, but comparatively few use the larger sizes.

Well done for making the test though :clap:
Click to expand...


I reckon that using FF and 35mm is the fairest test - same recording medium area, same lens, same size body. Using a larger film size would mean that extra variables would come into play. Of course, it could work the other way, using one of the Lomographers' favourite MF cameras and one of the super high MP count FF bodies from Canikon...
 
TBH from the 1:1 crops it looks to me like either the EOS3 focused in a different place (see the bowl of flowers as an example) or a lot of detail that should be present has been lost along the way. In this particular case the image from film looks very flat and unappealing, while the digital version has a depth that is pleasing: I would have expected a substantially better performance from film than that presented.
 
wet print the neg
inkjet the digital image
compare prints, its the only level playing field.
anything else is either a bit of fun or pointless rubbish depending on your perspective...:)
 
joxby said:
wet print the neg
inkjet the digital image
compare prints, its the only level playing field.
anything else is either a bit of fun or pointless rubbish depending on your perspective...:)
Click to expand...

You'd have to say what size...can you do say a 30" X 24" colour wet print these days?
 
excalibur2 said:
You'd have to say what size...can you do say a 30" X 24" colour wet print these days?
Click to expand...

dunno, 24x16 is a decent standard size

I think you can buy 50 inch rolls but its just another pointless thought thrown in to a thread already overburdened with pointlessness.
 
Last edited:
16x12 is a 4:3 ratio - just like some digital formats...

On A3+, an 18x12 (inch) leaves a 1/2" border. 30.5cm wide Fuji wet colour paper is close enough to 12".
 
So you need a source for A3+ colour darkroom paper then...:)
 
No thanks. I usually shoot in 3:2 and either print pretty much direct or scan and print digitally. No longer have access to a good darkroom and never managed to do colour P anyway, although I did do a bit of colour (E6 and C41) D when I did have access to a good darkroom back in the '70s. Do many people do their own colour printing (wet) these days or does pretty much everyone farm it out?
 
Did something similar a few years ago but not because I was trying to compare the two formats. I just happened to take the same picture on both formats. One of them was taken using fuji velvia 100 slide film and the other an EOS 20D.
16041545823_135d9f4e82_b.jpg


16660075261_d16f62a5c6_b.jpg
 
Nod said:
No thanks. I usually shoot in 3:2 and either print pretty much direct or scan and print digitally. No longer have access to a good darkroom and never managed to do colour P anyway, although I did do a bit of colour (E6 and C41) D when I did have access to a good darkroom back in the '70s. Do many people do their own colour printing (wet) these days or does pretty much everyone farm it out?
Click to expand...

You suggested A3+ instead of 16x12, well I'm not a darkroom colour printer so I dunno what the sketch is with paper availability, my point was that if you think it should be A3+......source it, cos I dunno if it even exists, I know 16x12 does .
Anyway, its all fairly academic, its unlikely to get done especially if peeps are gonna argue the toss about an inch of extra paper A3+ vs A3 and whether it has a flippin border or not...:D
 
Kei said:
Did something similar a few years ago but not because I was trying to compare the two formats. I just happened to take the same picture on both formats. One of them was taken using fuji velvia 100 slide film and the other an EOS 20D.
Click to expand...

Top one looks like roast beef, with string, fat and burnt grizley bits....(y)
bottom one looks like spam straight out the tin, artificial, processed, synthetic..:)
 
Seems to me that there are various trade-offs between film & digital (I use both). In terms of resolution, 'full-frame' digital may equate to medium format film (let's say). But the effect is different. Think of a vinyl lp versus a cd. There are many factors, and it's the overall synthesis that counts. The character of lenses can be an issue, for instance - some are more clinical and some are more gutsy, regardless of media, but it's the interaction of all these things that leads to the final result. Thus - it's hard to generalise.
 
joxby said:
You suggested A3+ instead of 16x12, well I'm not a darkroom colour printer so I dunno what the sketch is with paper availability, my point was that if you think it should be A3+......source it, cos I dunno if it even exists, I know 16x12 does .
Anyway, its all fairly academic, its unlikely to get done especially if peeps are gonna argue the toss about an inch of extra paper A3+ vs A3 and whether it has a flippin border or not...:D
Click to expand...


I suggested A3+ since it's the largest paper that will fit into home (inkjet) printers as well as taking a 3:2 print at 18" x 12" with no cropping - A3 is slightly under 12" wide so the image needs cropping (or less enlargement.) As you say, it's largely academic since very few people home wet print colour.
 
Last edited:
Interesting comparison.
I did this comparison a while ago, just for fun, not particularly critical. I also did it slightly differently in that rather than select a specific film, I just shot the film I had and then did a film-like processing of the digital file.
But it's more about the "look" of the image rather than the accuracy or sharpness.
This was a Canon 60D and an EOS100QD with a Canon 50mm f1.4 on a macro extension tube, though I've forgotten now which one is which.
Film vs Digital by Alistair Beavis, on Flickr
 
There was an old thread where we had some similar examples here. Not all the images showing up for me, though.
 
You must log in or register to reply here.
Back
Top