Want to see Film and Digital Compared?

[photophorous]

I was thinking some folks might be interested in what I'm about to do, but I'd like to see a show of hands. I'm not trying to deliver the final blow that will forever end this debate, nor am I trying to push my opinions on anyone. I just want to do this comparison so I know for myself what I should expect from my own equipment. I was thinking, since I'm going to all this trouble, I might as well share it with the group. However, I've read many threads, here and elsewhere, in which some people seem to get offended by the mere fact that these discussions still take place. I for one think there is still a lot of confusion surrounding this topic and it might benefit some people to see the differences for themselves, particularly those who don't have much experience with both formats. I will be doing the test for my own personal education regardless of how you vote, but please do vote and let me know if you would like to see a write up when I'm done.

Here are some details:

This is meant to be a practical comparison using equipment that is within financial reach for the average photographer. It will not be the end all of scientific tests. My equipment consists of a current consumer model 12MP crop sensor DSLR with a very sharp third-party zoom lens, and a 35mm film SLR with top notch prime lenses from the 1970s. Because of the crop sensor, I will not be able to use the same lens on each camera. Photos will be taken at several different aperture settings surrounding the expected sweet spot for each lens and the best images will be chosen for comparison. I will use a sturdy tripod and a self timer, and I will personally scan the film at 4000 dpi on a dedicated film scanner. Digital will be shot in RAW, film will be slide film, both will be 100 ISO, and all image files will be minimally processed for final output.

Please vote, and feel free to ask questions or add comments / suggestions. It will probably take me 2-3 weeks to get the results posted, but I'm asking this question now so I can take into account any suggestions you have before I start shooting.

Thanks,
Paul

[Xia_Ke]

Bring it on I'd be interested to see how the latest batch of digi's are comparing, not that I'll be switching though. I'm curious as to which DSLR you will be using and which slide film. Also, I would like to see a comparison shot with the same lens on each camera, regardless of the crop factor. What are you planning for subjects? Everyday things or will you be using resolution/color charts. Will be interesting to see how this comes out when the person shooting isn't really biased one way or the other.

[A.M.D.A.]

I voted "red", bring up the comparisons! :thumbsup:

[photophorous]

I'm glad to see there is some interest. This is going to be fun...gives me a reason to go shoot.

I'm not sure which film I'll be using yet. I'll probably take a roll of Sensia 100 and Velvia 100F along and then decide based on the subject matter. I think the two are so close in terms of resolution that it wouldn't matter either way...just a little different color pallet. The digital camera will be a Canon 450D with Tamron 17-50 f/2.8 lens. I'm very impressed with this camera/lens combo, but I'm still not convinced it can outdo slide film for my intended subject matter. I won't be shooting any charts. I'm planning to go to a park that I'm very familiar with and hopefully get at least a couple of decent landscape compositions, one wide, one normal. If I can think of a good location I might take a few city shots too.

I have an adapter that will let me mount my film camera lenses on the Canon body, so I can try to do a test with the same lens on each. I'll try that with a 50mm and see what happens.

I'm honestly not sure what to expect, but I can assure you that I'll try as hard as I can to get the best out of each camera, because I honestly want to know which I should use for landscapes. I don't want to cheat myself.

Paul

[Loupey]

I would like to see your results but I have a question with this part:

...Digital will be shot in RAW, film will be slide film, both will be 100 ISO, and all image files will be minimally processed for final output...

Digital can handle much more dynamic range than positive film and so RAW requires (sometimes extensive) post-processing to bring it back to realism. So how much will be enough?

I think you should post-process the digital as you normally do to how you feel the original scene actually looked before you see any results from the film. Not processing the RAW file to its "proper" level will automatically handicap the results.

[brmill26]

I'd be interested to see your results, Paul, just to see a practical user's comparison. If you're looking for subjective qualities, such as the "look" of film vs. the "look" of digital, I think it will be interesting. But with that said, if you are looking at resolution only, as your landscape statement suggests, this test has been done very scientifically by many different groups. Even APS-C digital resolves much more detail than 35mm. Although 35mm has a higher technical resolution simply because the negative is larger, the digital will always produce a sharper and more detailed result. Most of what I've read says APS-C at 8Mp has enough to match or surpass 35mm film.

For example, I've printed my XTi (10.1Mp) at 20x30. The image was processed from a RAW file, and shot with my 70-200 F/4 L at about F/8. It has gobs of detail. It was an image of a race car, and even with the whole car in frame, there is enough detail that you can see which direction the screw heads are facing in the fascia. There's no way 35mm can do that at that print size, because the (relative to digital) high level of grain extinguishes fine detail.

On the other hand, I'd be very interested to see a comparison b/t Medium Format and digital in terms of resolution. I was reading Joe McNally's blog the other day, and he was saying he thinks the new Nikon D3x provides better resolution than MF. Of course, at the moment that still comes at a huge premium, but that's still a cool fact. Obviously, in terms of pure resolution, it will be many years before digital matches 4x5 sheet film, much less 8x10.

All that said, I still shoot film b/c it's different. I have two MF cameras and one 35mm SLR. Honestly I don't think I'd ever use the 35mm for color because there's simply no point - my XTi does that better, cheaper, faster, and easier. But I use the 35mm for B&W film, b/c digital B&W and film B&W have very different looks, and that's what I'm after. And for me, the MF has higher resolution than anything I've got, and it offers the square format that's different as well.

[photophorous]

I would like to see your results but I have a question with this part:



Digital can handle much more dynamic range than positive film and so RAW requires (sometimes extensive) post-processing to bring it back to realism. So how much will be enough?

I think you should post-process the digital as you normally do to how you feel the original scene actually looked before you see any results from the film. Not processing the RAW file to its "proper" level will automatically handicap the results.

That's a good question, and a tough one to answer. I'm still open to suggestions on this one, but here's what I'm thinking at the moment.

First, I plan to shoot a scene that has relatively low contrast, so there should be no need for neutral density filters, layer masks, HDR or any other fancy editing. Normal levels and saturation adjustments should suffice. I want to illustrate the differences between the two formats. If I were to shoot a more challenging scene then my editing skills would play too large of a role.

Whether or not I edit the digital images before seeing the film should not effect opinions of what is seen in 100% crops. The close up views will basically illustrate sharpness and detail rendition. Sharpening will be applied as you would expect for web posting.

When I show the full image, I can show multiple edits to illustrate different takes on color and contrast. I can do as you say and make an initial edit before seeing the film file to eliminate any conflicts from the film image. But if I see fit, I may also show another edit of the digital image made specifically to mimic the look of the film image.

From past experience I almost always prefer the "look" I get from film. This is something that is entirely subjective and I don't expect my results to change that. What I'm really most interested in seeing in this test is whether or not I can get that same look from digital, AND whether or not digital will record as much or more detail.

Whatever I end up doing, I will disclose, and if I forget to mention something, feel free to ask. I may even go back and make changes if there are concerns about the way I edited something.

Paul

[photophorous]

I'd be interested to see your results, Paul, just to see a practical user's comparison. If you're looking for subjective qualities, such as the "look" of film vs. the "look" of digital, I think it will be interesting. But with that said, if you are looking at resolution only, as your landscape statement suggests, this test has been done very scientifically by many different groups. Even APS-C digital resolves much more detail than 35mm. Although 35mm has a higher technical resolution simply because the negative is larger, the digital will always produce a sharper and more detailed result. Most of what I've read says APS-C at 8Mp has enough to match or surpass 35mm film.

For example, I've printed my XTi (10.1Mp) at 20x30. The image was processed from a RAW file, and shot with my 70-200 F/4 L at about F/8. It has gobs of detail. It was an image of a race car, and even with the whole car in frame, there is enough detail that you can see which direction the screw heads are facing in the fascia. There's no way 35mm can do that at that print size, because the (relative to digital) high level of grain extinguishes fine detail....

Brad, I hope that my comparison sheds some light on what you're saying here. You may be right, but based on my own experience I don't completely agree. But I don't want to get into it here. We'll just have to wait on the results and talk about it then. But, this makes the need to include a city shot in the comparison seem more important. I think the two formats handle different subjects in different ways.

Paul

[Sushigaijin]

Please don't take offense, but I'm not sure a "comparison" is worth the effort. The way I see it, this is merely a lens comparison. We already know that film and digital will behave the same and differently at the same time. We know that both are sharp and colorful. What more are you trying to prove? Overall resolution? then you better use the same lens! price? You either pay now or pay later. Contrast? color? You can't really do that shooting RAW, better to shoot jpg and use the same lens. If you're just looking to prove that a certain film and lens combo can out resolve a certain DSLR lens combo, what's the point? It probably can. Or, it probably can't. Define "better" and the comparison will make a lot more sense.

Again, I'm not trying to be as aggressive as it sounds...it's hard to convey exact meaning on the interweb!

[Skyman]

I would be interested if you compared what processes gave YOU the best results for digital and what gave YOU the best results for film. That is what combinations of film type / file format, printing / post processing Give you better results. If you came up with your best technique for both then compared your best vs your best in terms of things like cost, ease of access, time you could have a very interesting study. The only other way to do it would be to get a full frame dslr and the equivalent slr with the same lens, but even then your post processing plays such a big part that the results are a little meaningless. Film and digital are different and each has their place. Just like neg vs slide film or jpeg vs tiff vs raw. Handled correctly all will give great results, each with their own inherent feel and that is exactly why people buy holgas or produce fine art prints from early 600X400 digicams - no offence to the holga people they are people too (I think)

[photophorous]

Please don't take offense, but I'm not sure a "comparison" is worth the effort. The way I see it, this is merely a lens comparison. We already know that film and digital will behave the same and differently at the same time. We know that both are sharp and colorful. What more are you trying to prove? Overall resolution? then you better use the same lens! price? You either pay now or pay later. Contrast? color? You can't really do that shooting RAW, better to shoot jpg and use the same lens. If you're just looking to prove that a certain film and lens combo can out resolve a certain DSLR lens combo, what's the point? It probably can. Or, it probably can't. Define "better" and the comparison will make a lot more sense.

Again, I'm not trying to be as aggressive as it sounds...it's hard to convey exact meaning on the interweb!

No offense taken. I think I understand what you're saying. To answer your question, I'm not really trying to prove anything. I'm trying to find out. I know a lot of people think they already know what the results will be and therefore this is a pointless exercise, but what a lot of people have so confidently told me in the past doesn't jive with my real world experiences thus far. So, I'm going to find out for myself, using my equipment, shooting my usual subjects. Better? I don't recall using the word, "better."

Here is what I said:

"I just want to do this comparison so I know for myself what I should expect from my own equipment. I was thinking, since I'm going to all this trouble, I might as well share it with the group."

"What I'm really most interested in seeing in this test is whether or not I can get that same (film) look from digital, AND whether or not digital will record as much or more detail."

Considering that all of the lenses I'll be using are regarded as very good lenses and I will be shooting them at their optimal apertures, I think it's safe to say that the differences in the test results will mostly be due to the differences in format. If the results are too close to call, I will have to do further testing before I can decide how I want to use my equipment.

Paul

[brmill26]

Well, whether or not 35mm would resolve the screw head may be a point of contention, as may be exactly why or why not, but the point about ultimate resolution is not. As I stated above, that point has been very scientifically tested by many people and the issue is well settled. Personally I really don't care about the ultimate resolution of 35mm b/c that's not why I use it.

From past experience I almost always prefer the "look" I get from film. This is something that is entirely subjective and I don't expect my results to change that. What I'm really most interested in seeing in this test is whether or not I can get that same look from digital, AND whether or not digital will record as much or more detail.

I'm very interested in seeing the first part of this quote - how the film and digital render the same scene. I agree that from my own experience, it's usually quite different. Though, I've never sat down and thoroughly compared multiple exact shots for test purposes, so I think that would be very cool to see. Because like I said, that's why I use my 35mm - to achieve a look that digital cannot reproduce. So seeing that borne out would be cool. I think the films you listed above were color, but I think it would be extremely interesting to see a comparison of (true, not C-41) B&W film to B&W digital as well. Obviously that would depend even more greatly on the exact film(s) used, but I think it would still be an interesting comparison of how the two render a scene.

But as Erik pointed out, the latter part of the above quote creates loads of issues. You really have to use the same lens on both cameras. Even two great lenses can render light quite a bit differently, and of course the difference b/t zooms and primes is even greater. So in order to examine detail at the finest levels, it really has to be the same for reliable results. Otherwise, it's impossible to tell whether the image was rendered differently because of the medium or the lens.

Finally, shooting in RAW will definitely give the digital even more of an advantage. But I'm fine with that b/c most digital users shoot critical work in RAW, so for me that actually makes perfect sense to bring that in.

[readingr]

How about being a bit more rigorous and documenting the post processing applied to each photo and using normal post processing to each so that you end up with realistic photos as many have pointed out.

How will you post the results? the pics will loose a lot in downsizing!

The Gadget show in the UK printed a film v DSLR picture to the size of a two floor building and compared the prints and came down on the side of Digital for smoothness over the best film you can buy.

Roger R.

[danic]

I'nm going to be a party pooper, and say "Don't bother...."

I feel there is already too many threads/topics/arguements on this topic already.

If you like to shoot film, thats great. If you like digital, thats good too!

Just enjoy your photography!

[another view]

I think it would be interesting from the standpoint of one person making their own comparison, who is experienced with both media and using more common equipment for people like the average photographer on this site (not the $5k+ latest greatest DSLR and a drum scanner, etc). Personally, I'd mainly be interested in your bottom-line conclusions moreso than anything else and if I were to do the same experience I might come to different conclusions. That's what's interesting to me.

However, there are a lot of pitfalls too. Comparing a good darkroom print vs a good digital print is a totally different experience. Differences between what's possible with film and digital are yet another (thinking more of photoshop and other techniques than of equipment). Differences between my monitor and yours will make me wonder what you're really seeing.

[drg]

There have been several good comments made already many of which I would echo regarding that this may not be worthwhile.

One of those most to the point is that at best it will result in comparing two different prints most likely made by an entirely different process.

Then it get really complicated.

Even with the same lens on the same tripod in the same position there are several other factors that could 'mess up' any meaningful comparison.

- there's no current slide film (reversal) worth scanning at 4000dpi.
- a zoom vs a prime lens results in all kinds of mismatches.
- a 1970's vintage lens has degraded coating and won't produce the same quality of color as it did even 10 years ago.
- if the 35mm image was cropped to the same central portion of the image as the digital and rendered at the same dpi it might be somewhat more accurate.

- What confusion surrounds this topic, if done on a direct comparison basis?

I still shoot film, rarely though side by side. When I do still shoot film it is for aesthetic (not quality) reasons. For those with no experience in one medium these tests usually do far more to confuse than illuminate because they rarely address what someone needs to know or do so ineffectively.

If you want to really compare the two medium, don't pixel peep or allow anyone else to do so. Just have the two best prints of a reasonable size (10x15 is more than big enough) made and look at those as the comparison. Everything else too easily becomes unrelated to film and digital.

[photophorous]

Thank you all very much for commenting. I'm continuously revising my testing plan as I read your comments. There are a lot of good points here. But just to be clear, the main reason I'm doing this is to help me decide which of my cameras and my lenses will best meet my personal requirements for shooting landscapes. I'm confident the test will give me exactly what I need. All of the discussion taking place here is just about what you folks can or can't glean from my results. That's fine. I want to do what I can to make it informative for everyone, if you are interested. But I never meant to imply that this would be any kind of definitive test. It will be just one more example to consider in the context of all the other comparisons you may have seen. Take it for what it is.

One point I think several people have made is that I should process all of the images as I normally would, to get the most out of each format. I think this is a good suggestion, but there is no way I can shoot enough different scenes to cover all scenarios. So, I will probably just go out and find what I find. Whatever I end up with, I'll process accordingly. And, I'll do my best to document and share my post processing, as long you promise not to make fun of my lack of photoshop skills.

The fact that I have to use different lenses seems to be a big point of contention, but I think that is due to a fundamental misunderstanding about the whole point of my comparison. I keep using the word "practical" on purpose, because in real life you would not use the same lens on two different sized formats for the same type of shooting. The test I'll be doing is a real world comparison of how most people could/would use these different formats. Therefore the results will represent two options that are practical for most people. However, I will do what I can to alleviate this problem by mounting one of the film camera prime lenses on the digital body and shooting the same scene with the same lens on both cameras. But you must keep in mind that this will require moving the tripod between shots, which will change the focus distance, the perspective, and the DOF characteristics. This will only be a little closer to equal and only from a technical, not a practical, standpoint. One thing I can do that might give us a little more practical sense of the differences between the lenses is to compare both lenses, at the same focal length, on the digital camera only. But even this is not exact, because the crop sensor will only use the best part of the full frame lens. I don't think there is any possible way to do a technically perfect comparison for these two different formats. That's why I'm going with "practical," and shooting real world scenes instead of test charts.

In regards to previous tests / comparisons that many of you have mentioned.... I remember that UK Gadget show comparison in particular and there were several things that bothered me about it. First, it was done like a fashion shoot. I don't shoot fashion. Second, they were shooting 400 ISO color negative film, which I would never expect to perform on par with digital. They had studio lights. Why weren't they shooting Astia? Doesn't make sense. Third, their processing of the film image was suspect because it had a color cast. Fourth, the comparison was done using a printing technique that 99.9% of us will never use. I've probably seen most of the other tests floating around the internet and as far as I can remember there was something similar that bothered me about each one of them. That's why I'm doing my own.

Thanks for voting.

Paul

[photophorous]

Hi Everyone,

I finally finished my comparison. I've been posting it in pieces in the new blog I created. If you're interested in the results click here.

Being in blog format, the most recent post is the Conclusions post. If you want to see it from the beginning you'll need to scroll down to the bottom and work your way up.

Hope you enjoy it.

Paul

[Photo-John]

I'm sort of on the fence about this. I think comparing quality is sort of a waste of time at this point. But I'm sure there are people who disagree or haven't had the opportunity to compare for themselves. At this point, I think the difference mostly comes down to the process involved in shooting film vs digital. For me, I shoot very differently depending on whether I'm shooting film or digital. I actually haven't shot film for quite a few years. But I am aware of the different approach I have when I can't check my histogram or composition on the LCD display.

One thing to consider are the lenses you're using. Lens coatings have hugely improved in the past decades. So it would be good to shoot in a very controlled environment where flare and saturation won't be compromised. Make sure to use lens hoods on both cameras, too. A lens from the 70s might be sharp. But it will very likely have less contrast and saturation than even a cheap current lens.

[photophorous]

I'm sort of on the fence about this. I think comparing quality is sort of a waste of time at this point. But I'm sure there are people who disagree or haven't had the opportunity to compare for themselves. At this point, I think the difference mostly comes down to the process involved in shooting film vs digital. For me, I shoot very differently depending on whether I'm shooting film or digital. I actually haven't shot film for quite a few years. But I am aware of the different approach I have when I can't check my histogram or composition on the LCD display.

One thing to consider are the lenses you're using. Lens coatings have hugely improved in the past decades. So it would be good to shoot in a very controlled environment where flare and saturation won't be compromised. Make sure to use lens hoods on both cameras, too. A lens from the 70s might be sharp. But it will very likely have less contrast and saturation than even a cheap current lens.

John,

I'm already finished.

I know I'm kind of long-winded with all of this, so I can't blame you for skimming over it. Let's just say, this is something I had to do for myself. It's done. I learned from it. Now I can move on and feel good about the direction I'm going. I posted it all for everyone to see, because I think some other people may learn from it too, within the narrow scope of what it is.

Thanks,
Paul

[another view]

I learned from it.

That made it worthwhile.

I could say a lot about why, but it doesn't really matter. It's an individual choice and there isn't a wrong answer. Sounds like you did it objectively, too.

[brmill26]

As AV said, at least you confirmed the information in your own tests.