The Real Resolution of Film vs. Digital

From Kodak you can get some film info on the net so I have been reading about T-MAX 100 and EKTAR 100 negative film. For T-MAX you have this- resolution info 63 lines/mm (TOC 1.6:1) and 200 lines/mm (TOC 1000:1). TOC 1000:1 looks very exceptional to me, like what is possible with some laser method. But TOC 1.6:1 could be intresting for the normal photography. To resolve 63 lines/mm you need 126 pix/mm or 3034 for 24 mm. For the FF area this is 13.7 MP

For Ektar I don't get anything to compare this res. info but we can compare the Modulation Transfer Curves. For T-MAX it is close to 100 (RESPONSE%) up to 50 cycles/mm and falls down to 60 at 100 cycles/mm.

For Ektar the curve is close to 100 up to 20 cycles/mm and at 50 cycles/mm B and G are down on 55 and R on 30.

So BW and color are really different regarding resolution in films!

So far I have not seen anything from Fuifilm with RGB separated.
Another important part is the lens:
I found this lens test for large format:

http://www.hevanet.com/cperez/testing.html

Here is medium format:

http://www.hevanet.com/cperez/MF_testing.html

Some of these tests look strange to me- like wrong fokus.
Normally f/22 should never give better results in the center than f/16.
 
I found a lot of interesting points in the discussions here and other sites. I notice that is shown here (by Tim Parkin, nearly half-way down the page) how sufficiently small objects (like the individual red berries on a bush at a distance) can fall between between pixels and not register at all in a digital image.

I'm a casual amateur not constrained by the need for speed of workflow and other parameters where digital excels.
To paraphrase the (possibly apocryphal) response by Rolls Royce to what was the power of one of their cars - "Sufficient", I would say that the resolution I can get for landscape/townscape photography using 6x9 is 'sufficient'. I get my negatives developed and scanned by professional labs. With an appropriate image rendering program (one that can fully use the data output by the scanner), the pictures are quite crisp on a ~22" (~55cm) workstation monitor, even at higher display zoom-in so that you use the scroll bars to navigate the whole image.
That's shooting with anything in the range Ektar 100 to Portra 800, at stops down to f/32, with lenses from 64mm wide angle to 100mm.
 
I found a lot of interesting points in the discussions here and other sites. I notice that is shown here (by Tim Parkin, nearly half-way down the page) how sufficiently small objects (like the individual red berries on a bush at a distance) can fall between between pixels and not register at all in a digital image.

I'm a casual amateur not constrained by the need for speed of workflow and other parameters where digital excels.
To paraphrase the (possibly apocryphal) response by Rolls Royce to what was the power of one of their cars - "Sufficient", I would say that the resolution I can get for landscape/townscape photography using 6x9 is 'sufficient'. I get my negatives developed and scanned by professional labs. With an appropriate image rendering program (one that can fully use the data output by the scanner), the pictures are quite crisp on a ~22" (~55cm) workstation monitor, even at higher display zoom-in so that you use the scroll bars to navigate the whole image.
That's shooting with anything in the range Ektar 100 to Portra 800, at stops down to f/32, with lenses from 64mm wide angle to 100mm.

Digital is good but you have some limits with the Bayer method. Pixel shift is one way to make it better. Film is still interesting if light is good enough.
 
I shoot film and digital. I don't choose digital for speed. You can shoot film or digital either fast or slow. It's up to you.
Sure, whatever is the usage that applies.
For 'fast', I'm thinking of, say, wedding photography, or news reportage, where virtually instant visibility of results to the client is expected.
In contrast to my film workflow, which is, finish the film (might be weeks later ;) ), post it to the lab, get it back in the post; that's three days minimum, sometimes a week.
 
I used to parrot the line about how much resolution film had. Some guys on APUG did some great work quantifying it under absolutely ideal conditions.

Too bad the instant a camera touches my hands "ideal conditions" have goen out the window. So I did some experiments on my own.

For these tests* I used Kentmere 100 (traditional grain, not the finest grain possible, but finer than Tri-X or HP5) loaded in an Canon AE-1 with a FDn 50mm F1.4.
*The negatives I used were from my vacation, so not test images at all.

When I scan 35mm film on my V600 scanner, I get a result that is roughly similar to a 8MP sensor. I consider them lower quality than my old Olympus E510 or Nikon D80.

When I take that same negative and "scan" it with a light table, a good 1:1 macro lens, and my 20MP EM1-mk2, I cannot make out all the grain perfectly clearly (ie in raw data terms there is more than 20MP of data there).

When I use pixel shift on that same setup, I can very clearly make out all the grain (there is definitely less data than the 50MP captured by pixel shift).

So, while it's true I'll admit that Kentmere 100 results in more than 20MP of data, when I enlarge it in my darkroom through a fully analogue process, I find the sweet spot for image quality to be 5x7 or 8x10.
Meanwhile, images taken on my 16MP EM1 mk1, printed on a good inkjet (P800) or via a good photo service have more detail in them at 11x14.

I like film, I enjoy shooting it, I like the aesthetic, I like the process. But I'm done saying 35mm has more resolution than digital.
 
You guys might find this interesting. Below is a photo taken of the Times Square Area with a Leica Monochrom (M246) at ISO 800 with the WATE. The second photo is that same image enlarged for a wall mural in our offices approximately 14'x10'. There were no software tricks used for the enlargement. The image was res'ed up in PhotoShop to full size at 200ppi and printed.

I'm blown away every time I look at this image. The grain in the final print is very small - remember this was shot at ISO 800. The view through the building windows shows a lot of detail. I don't feel a 35mm film negative enlarged to this size would achieve these results - even if an 8x10 interneg was made for the blow up.

I can't speak to the technical tests and numbers that others have posted, but as an amateur photographer I pretty impressed with what digital can achieve.

NYC-1000633 by reddott2012, on Flickr

ddi RECEPTION-1001913 by reddott2012, on Flickr
 
Let me throw out a different idea for an old thread: Maybe digital is just too easy? and too good? It's like getting your steak at a restaurant versus hunting wild buffalo (or some such ) and cooking out on the trail. With film, there's the challenge part of the process that is more engaging and/or satisfying for the "hunter/gatherer" experience. The intangibles can and do matter but can be difficult to assess in terms of what their impact might be on the creative process. I think the impact is "real" and not totally imagined, but with most digital photographers of a certain age growing up with film, the differences are blurred.

As to the images themselves, in my mind there is little question that digital has finer detail. The thing is that today, this aesthetic is stylish. It's even cool to see the detail. No questions about that. BUT there is something that this accomplishes that is beyond authentic because this isn't really how the mind processes images. This is in part what makes reddot's image so amazing in my view. What a great shot! only the harsh light of the wall doesn't do the blacks as much justice as the screen shot does, but it's still pretty doggone cool! (Thanks for sharing btw). But in other places and my perhaps lesser experience, digital can become almost creepy or off-putting where the smoothness of the printed output looks more like a soup, cake icing or something inauthentic. I've never had this "problem" with film, and so I guess I find film seems more fun and "better" for my unstylish aesthetic, and the appeal of digital has just run its course. Yes it's there, but until I can afford a digital Alpa 12SW with a PhaseOne B&W back... or a Leica Monochrome... it's not likely to happen. And to be fair, it's not that the Leica Monochrome is unreasonably expensive itself so much as that the money is tied up all at once that made it a mental step too far at this stage. But no kidding, my hat's off to those who were able to take that leap.
 
I like both film and digital prints and for normal sized prints, say A5 in size.

But I do prefer film cameras to digital; with just one or two exceptions and both of them have red logos on them. Neither of them have a 236 page manual on a CD and a 44 page "quick" guide...

Edit as I hit the wrong button; as I've condemmed a good camera (despite its silly menu of scenes and changes to the previous model), I'll add that I was fuming at the thought of having to print some of the pages, which were 10% text and the rest blank. A couple of days after buying the camera I found a nice fat book about it for a pound in a charity shop. So what I pass off as sanity has been restored.

Regards, David
 
But I do prefer film cameras to digital; with just one or two exceptions and both of them have red logos on them. Neither of them have a 236 page manual on a CD and a 44 page "quick" guide...
Fortunately, you really don't have to read and retain the information in the manual if you just want to take pictures. And they are really small pages. I did skim the manual when I first got my camera, but I have never had to refer to it again. Most of the default settings are default for a reason.
 
Back
Top