Film versus digital

matt_needham · October 13, 2009

I have a simple test for you. Shoot with your digital gear and shoot with your film gear. Make prints. Hang them on the wall. Live with them for a while. Now which looks better to you? That's a much better test than pixel peeping someone else's results. It might not tell you which has better resolution, but it should give you an idea as to which works better for you.

james_symington1 · October 13, 2009

I finally did do a direct comparison a couple of weeks ago shooting a very detailed subject - a hillside covered with trees. I used a D700 and an F6 with an 80-200mm f2.8 set at infinity at F8. The D700 was set at its base ISO 200 and produced a NEF processed in NX2 and the film was Provia 100F scanned through my Coolscan 9000 using NikonScan.

The results were that film, on close inspection, yielded rather better resolution. Fine details in leaves came out much better and individual leaves were more separated from each other. The D700 couldn't match that and I tried to sharpen the pic using a variety of techniques. I printed both at the same size and judged the output from those.

There was more grain with the film of course but I don't mind that at all. The D700 brought out more subtlety in the colours (faint reds and yellows) whereas the Provia was a much greener rendition that was obvious from the slide even before scanning.

My conclusion from this experience was that you'll need something upward of 20MP to match film that is scanned decently well if all you care about is resolution.

john_liberty · October 13, 2009

Mike, It's really an apples and oranges thing. I finally realized that if I shot color digital is fine. Even my P&S digital takes great photos. But w/ digital you have to address other issues besides resolution. Crop factor, sensor noise, etc all need to be dealt with. And w/ film you have a lot of different looks you can get depending on your choice of emulsions. Pure resolultion is just one factor, and unless you're doing very large prints it's not that important in my view. On the other hand, my Leica 35mm camera has a lens that images better than my little P&S, so if I want to shoot color I take them both and nearly always prefer the Leica photos. If I shot a lot of color I would invest in a Pentax DSLR and get the Leitax Leica lens adapter and some wide angle Leica R lenses and be very happy indeed.

If you shoot B&W, or want large prints (16x20 and up), forget digital. Nothing is as good in B&W as a film print. It isn't even close, unless you like that smoothed over, narrow dynamic range look. Can be quite nice for portraits, but you'll not have a negative and have to inkjet your prints. Not acceptable in my opinion for B&W. And a MF or LF camera can have their negs blown up to very large sizes w/o any loss of sharpness. If you shoot 5x7 or larger cameras you also have the option to contact print your negatives, which opens the door to Platinum, Palladium, and a host of other beautiful styles of printing.

These resolution tests are a snipe hunt. Who looks at little zoomed in crops of an image? You can only see which look you like best by viewing the entire enlarged print.

James, you might want to invest in a shutter tester. That's the only way to know at what speed your camera is actually firing at. I suspect you may have an inaccurate shutter. Wise to bracket the important shots too.

anthony_zipple · October 13, 2009

The only way to do this is with blind comparisons. The few that I have seen do not suggest that film is chosen as "better" more often than randomly. In short, they global IQ in blind tests comparing good film and good digital images is about the same.

Dave Luttmann · October 13, 2009

If you shoot B&W, or want large prints (16x20 and up), forget digital. Nothing is as good in B&W as a film print. It isn't even close, unless you like that smoothed over, narrow dynamic range look. Can be quite nice for portraits, but you'll not have a negative and have to inkjet your prints. Not acceptable in my opinion for B&W. And a MF or LF camera can have their negs blown up to very large sizes w/o any loss of sharpness. If you shoot 5x7 or larger cameras you also have the option to contact print your negatives, which opens the door to Platinum, Palladium, and a host of other beautiful styles of printing.

</blockquote>

Right you are John. I find that for the color work I do for weddings, a DSLR is just fine....all you could ever want. I'm enjoying the low noise, good color and tonality, and the resolution from my 7D. For B&W though, I'll stick with MF and 4x5.

doug herr · October 13, 2009

<a href="http://www.photo.net/photodb/user?user_id=1722353">Anthony Zipple</a> wrote: "<a href="http://www.photo.net/member-status-icons"></a>The only way to do this is with blind comparisons."

I recently took several prints to a meeting with a gallery owner. The origin of the prints was not identified in any way (a mix of Nikon and Leica, several film types and the Leica DMR). The gallery owner made a big fuss over the color and detail of the larger prints from the DMR. At the end of our meeting he chose several prints for display in the gallery, all (without his knowledge) made with the DMR. YMMV.

desmond_kidman · October 13, 2009

One of the posters wrote "........if all you care about is resolution."

In matters of perception, which this debate is, it is rarely about just one aspect of perception. Therefore, the opinion of another is rarely applicable.

Everyone should run their own tests, get the best scan made from the film, and have great prints made. Decide for yourself. Expecting to get a valid answer from a forum is like getting an answer to what is the best wine, or what is the best perfume. The answer is going to come down to personal perception.

of the

spanky · October 13, 2009

I'm also a film shooter but not because I think film is better. I really don't care. Besides, the human eye cannot even come close to resolving detail as much as camera lenses, sensors and films are able to. So one can make the point all they want but in the end, most people won't be able to tell the difference. I also disagree with the notion that film is to be used for more important things and digital for less. Every shot you take should be important enough to make sure nothing is overlooked. Every time one takes a photograph, they are making a statement. They are saying "I think this is worthy of a photograph". Why not make all the effort necessary to ensure that the results are the best they can be?

iwao_sakurai · October 13, 2009

WElcome to the future when pics resolution, color, performance become so perfect, so flawless, so noise free. Instead of photographers we will have Image Perfectographers were digital holographic 3D tablet prints with zone crops of x1000 pixels will be the standard work output for professionals. With Microscopic zone crops not too far off.

Meanwhile the odd folks, used to be called Photographers, with their primitive "film" boxes try to sell their "abstract" grainy prints.

greg_peterson3 · October 13, 2009

I'm firmly in both camps.

I have a Nikon D3x that I use for most of my aerial work because it lets me shoot fast (helicopter time is expensive!) and get the quite respectable results to my clients very quickly.

I have a medium format camera that I use in the studio and for most of my architectural work. I have a 22 megapixel digital back for it, but I also often use it to shoot film. I scan the film with a Nikon CS 9000. That gives me a 49 megapixel image which I sometimes need for very large prints.

If the client absolutely insists, I shoot 4X5. I then do "proof" scans on an Epson 750 and have the final selection(s) scanned on a drum scanner. Makes great wall-sized prints.

And I agree with a previous poster, that the real image quality issue is dynamic range, not megapixels.

So I boldly say "Both!"

scott_murphy_photography · October 13, 2009

I am not at all surprised.

Digital has come a long way in the last 10 years but in my opinion, film still reigns supreme. I shoot a lot of black and white, in both 35mm and 120, using Ansel Adam's Zone System. I don't care how good the matrix metering system is in your camera, or how good you are at Photoshop, you will never be able to achieve with digital what you can with black and white film and top notch paper like Oriental Seagull, which although it is expensive, is all I use.

hal_b · October 13, 2009

Which one would win on Celebrity Deathmatch?

Dave Luttmann · October 13, 2009

<a href="../photodb/user?user_id=5012474">Greg Peterson</a> <a href="../member-status-icons"><img title="Frequent poster" src="http://static.photo.net/v3graphics/member-status-icons/1roll.gif" alt="" /></a>, Oct 13, 2009; 12:05 p.m.

And I agree with a previous poster, that the real image quality issue is dynamic range, not megapixels.

</blockquote>

While for the most part I agree, the problem I found was that while some people go on and on about the dynamic range of film vs digital, when you ask what film they are using, you find out it is Velva or Astia, or something along that line.....films with DR that was surpassed a long time ago with DSLRs. Heck, even my old Canon 10D beats what you can get in terms of DR from Velvia. That said, I'd take the Velvia scan any day over the 10D ;-)

iliafarniev · October 13, 2009

That can of worms has never been closed actually. But, not to say one is better then another, I suggest to consider these numbers:

Largest file size top DSLR can produce is less then 50Mb

Largest file Imacon scaner makes from one good 35mm frame is about 600Mb

Largest file Heidelberg drum scaner makes from one good 35mm frame is more then 1000Mb

steve_in_l.a. · October 13, 2009

Ignoring the source of the original poster's question (the blogger in question is an irritating publicity hound who garnered many new searchable mentions of his name in this very thread, which I'm sure was his goal), I think this will remain a theoretical question of some interest until we definitively answer it. To this point, I suspect if the camera companies have answered it (and I suspect they have), they aren't making the answers known publicly.

But the question remains, how much detail can the finest grain 35 mm film capture, and how many megapixels would you need to equal or surpass it? The manual of photography seems to suggest that 35 mm film is good for anywhere between 17 and 170 MP, if I'm reading it right, but they present no basis for their claim (as they fail to do often in that poorly written book). Meanwhile, the Koren reference given above suggests that 11+ MP "clearly outperforms 35 mm" using an end result assessment of print quality. For my part I find it very difficult to believe that 11 MP figure simply based on the much smaller size of photo sites in film (grain size?) vs digital.

My own sense, a vague one at best based on no hard data at all, is that somewhere well north of 20 MP, perhaps above 40 MP, we may actually completely defeat the best that 35 mm could ever offer. But on a related note, I think we must plateau fairly soon. The 2.6 MP D1 from 99 was increased to 24.5 in the D3x after only 9 years, a 9.4-fold increase. On that curve, we would arrive in the neighborhood of 230 MP in the D6x in 2017. I'm no engineer, but I doubt that even a century from now the 35 mm sized sensor will be capable of that, so if we make it to 30 or 40 MP in the 35 mm size, that may be all we ever get, eh?

dan_south · October 13, 2009

I seem to recall looking at that comparison sometime back. If it's the one that I'm thinking of, it was a photo of the countryside where objects were at a number of different distances from the camera.

At the time I though that this wasn't an accurate test, because you couldn't tell where the exact plane of focus was in each photo. Therefore, areas of the frame that were being compared may not have been at the exact focal point. If you compare a section of two photos, you should compare them at the plane of focus. If the section from one photo exhibits lens blur (a function of depth-of-field), then that photo will seem inferior based on only this test.

The way to test this properly is:

(1) Compare a flat surface (e.g. a wall)

(2) Make sure that the wall is exactly on the plane of focus.

(3) Make sure that the wall is lit from the side or the top in order to show contrast on the surface (no "front lighting")

(4) Use two cameras with sensors of exactly the same size (e.g., a 35mm film camera and one with a full-frame digital sensor)

(5) Use the SAME LENS on both cameras.

(6) Use mirror lock-up on both cameras. Both cameras mounted on the same tripod.

(7) Use SAME aperture on both cameras.

(8) Set the digital camera to its optimum ISO setting.

(9) Both cameras must use the same auto-focus settings, and the auto-focus system of each should be verified in advance.

(10) If manual focus is used, multiple shots should be taken at slight variations, and the sharpest one for each camera should be the one used in the comparison.

If someone wants to set up a more scientific test, I'll be glad to consider it. Anecdotal experiments aren't going to deliver a convincing verdict.

craig_sander · October 13, 2009

People want Ken Rockwell to be on one side of the film vs digital debate, when I believe he truly sees the value and virtue in both.

I've been going to his website for several years now and I honestly don't see him as being contradictory, two-faced, or any other negative comment that readers want to call him.

If he bashes Nikon for something, the Canon shooters rejoice.

If he talks up Velvia in medium format, the film shooters consider his words to be law.

But when he "switches sides" (ie mentions something positive about the "other guy") then he is a traitor, doesn't know what he's talking about, and is simply addicted to attention.

Come on now.

People grow and modify their opinions all the time. Ken offers his knowledge to a wide range of people, and I think all in all, he provides a good service.

stephen_asprey2 · October 13, 2009

The worm is turning....I'm glad to see that the discussion has been technical. The other thing to remember is that ken has the gear to do extreme pixel peeping, we don't.

From a photographers point of view, the top end offerings of Canon and Nikon are superb. They are both starting to reveal that some so called pro lenses are not that good after all.

However, one has to be very very good, with the best lens on the most expensive pro digital body to come close to the resolution of a low ISO film like Velvia.

patrick j dempsey · October 13, 2009

I really hope this stupid MP race is over soon. All it's done is to further this stupid film vs. digital war. It's been a huge waste of $$$ for the companies and the consumers that should have been spent on something meaningful... like getting digital sensors to do things that film absolutely CANT do. There is a little bit of work being done in this realm, such as the HDR or bracketing button that snaps off three different exposures at full FPS. How about better tonal range so that color images and color images converted to B&W don't look like muddy mush? How about programmable color spaces so the camera can record a wider range of middle tones for portraits or subtle differences in bright colors for flower photos... (like choosing between Kodak NC and VC films!). How about UV mode? How about better solutions for long exposures/high ISO than just in camera or out of camera post processing? How about recording distance information as part of the image file so any photograph can be used to generate a 3D map for doing computated focus correction? Any one of these would be a bigger leap that the difference between 10MP and 12MP.

sebastianmoran · October 13, 2009

Here's my story... People are getting amazing 4x6 prints from their digicams. Since buying a D70, I've been blown away by the image quality.

Then, threads here showed more resolution in film.

I did my own test. I took comparable not-exotic equipment out for a shoot of stationary objects. My test rigs: Nikon FG with 50mm f/1.8 Nikkor lens, Fuji 200 color negative film, scanned on a Coolscan V at 4000 ppi, producing a 20MPx file. Nikon D100 with 35mm f/2 Nikkor lens, shooting at ISO 100, 6MPx. Nothing exotic. Nikon primes of comparable quality.

Results:

a) Film has more resolution, you can see it.

b) The D100 has much less noise.

c) In some areas, the D100 had better rendition, of the surface textures, better feel (e.g., textured granite in grazing sunlight). In these areas, it was the film noise that wiped out the texture, leaving digital the winner.

d) Looking at 8x12 prints, they were very comparable. Both excellent image quality, in my judgment. Viewers preferred the digital saying it was "sharper." I'm sure this because I used pretty strong sharpening (thanks, Bruce Fraser!) on the print from digital and almost none on the print from film scan.

e) There is a subtle difference in feel of the light in the images. I can't say either is better or worse, just different. This should not be a surprise. Films, lenses, and sensors impart a feel to their images.

f) The scanned film probably had 10-12Mpx of information (the file size is larger, of course); the D100 only 6Mpx period. Yet the images are very comparable. The megapixel-count isn't everything. Resolution isn't everything in image quality. My film scan has more resolution than the digital, but the prints are comparable. I summarize it this way: In my experience, Digital pixels are better than scanned-film pixels. Maybe by 2:1.

Caveats:

- I used nothing exotic. You can get better film images with better film, better scan, maybe(?) with a better lens, certainly with MF or LF. But, you can also get a lot better digital camera today than the D100.

- All my tests were in sufficient light. Things are probably different in low light.

Corollary: If you are thinking about MF film for better image quality, it will take a really good scan of MF to beat today's DSLRs.

Images to follow.

sebastianmoran · October 13, 2009

Here's the comparison... Actual pixels. I think film shows more resolution in the writing on the blackboard. I think the texture of the granite in the red box is better with digital. Notice the noise in the film scan. Notice how clean is the digital image.

sebastianmoran · October 13, 2009

Here's the scanned film in Screen resolution:

<h4>Scanned film: Fujicolor 200, Nikon FG, 50mm f/1.8, Coolscan V</h4>

sebastianmoran · October 13, 2009

Here's the D100 capture in screen resolution. Even at this resolution, people have noticed the difference in the granite texture. And, the light has a slightly different feel, even though the images were taken within minutes (Note shadow positions are nearly the same.)

<h4>Nikon D100, 35mm f/2.0 Nikkor lens</h4>

mauro_franic · October 13, 2009

Quality is subjective. Detail not so much.

This topic is so long lasting because the answer largely depends on technique and experience .

35mm BW film can resolve 35+ megapixels of detail (It would take a 50MP+ DSLR after bayer pt, etc to approximate that).

35 mm color film like Velvia can resolve about 24MP (It would take a 35MP DSLR to approximate it in high contrast detail). That is what film can do. The rest is up to the individual's equipment and their ability.

35mm print film captures slightly more detail than the top end DSLRs today.

Yet:

Some people post they can only get detail comparable to a 10 or 12 megapixel camera, or that they cannot get more detail out of medium format film as they can with their DSLR. Some say their ability to use film yields no better detail than a 6MP camera. For some others it is 3MP. They are convinced of this because IT IS TRUE - yet the confusion is that they think it is a limitation of the film, not their technique-experience-lens-scanner-development skills, etc.

sebastianmoran · October 13, 2009

If anyone wants to play with the full-res images, let me know and I'll post links.

Sign In

Film versus digital

Recommended Posts

Link to comment

Share on other sites

Top Posters In This Topic

Popular Days

Top Posters In This Topic

Popular Days

Posted Images

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Create an account or sign in to comment

Create an account

Sign in