#9

The purpose of this comparison was to help me choose between film and digital for landscape photography, and since it was intended to be practical I used my own scanner and cameras in a way I normally would. Things might be different if I shot with L primes on a new 5D Mk II or if I had my film professionally drum scanned at 8000 dpi, but that isn’t practical for me. My main concern at the beginning of this comparison was with whether or not my 450D could capture as much detail as my Coolscan V can pull from 35mm slide film, but the process of editing the digital files first and the film second made it painfully obvious that I don’t do a good job editing digital files to have the kind of color I want. Generally I’m better at getting the color I want from film, but that can sometimes be even more challenging in less than ideal conditions. But since color is such a subjective topic and one that I think really does have more to do with editing skills than the capture method, I think it’s best to separate it. So for now, let’s talk about resolution, sharpness, grain, and dynamic range. I'll come back to color in a later post.

Resolution and sharpness are not the same thing. A sharp lens will give you higher resolution, but when you are talking about the final image, the sharpness of the image is not the same as how much detail has been recorded. Sharpness is just how well defined the edges are for whatever detail is there. In regards to sharpness, I didn’t learn anything here. I already knew digital was sharper and as far as I’m concerned this comparison confirmed that.

As far as resolution is concerned, I really wasn’t sure which would record more actual detail. Based on the overall results, I think digital has a very slight edge, but it is mostly seen in the third example, for which the film version required the most editing. In other crops, the main differences I see are in sharpness and the level of detail is very close to the same. I expected this to be close, but honestly I expected film to do very slightly better than digital, and for this practical comparison of my own equipment, I was wrong.

One other thing I discovered when doing this comparison is that my film scanner is not really scanning at the specified 4000 dpi. Yes, I am getting files in the neighborhood of 5500 x 3700 pixels, but if you look closely at the film versions of the “Film Sized” crops you can clearly see pixelation (or is it pixelization?). This is not to be confused with the resolution of the film. This clearly shows that the limits of the scanner are less than 4000dpi. Whether or not there is more detail in the film is hard to say. If I had to guess I’d say there probably is a bit more, but frankly it’s irrelevant to my “practical” comparison. I’m not even considering upgrading my film scanner or paying big bucks for drum scans, so I’ll probably never know.

Grain is another difference. The digital files are undoubtedly cleaner, but I’ve never been bothered by grain in 12x18 prints I’ve made using this scanner. The grain is very hard to see, even viewing the print from just inches away. Less grain could allow for slightly larger enlargements, mostly because grain can impose certain limits to the way the image is sharpened. You have to be careful to avoid exaggerating it. It’s a small issue for me, but never the less, another strike against film.

Dynamic range is another difference, as seen in example 4. What I see agrees with what I’ve read elsewhere. Digital has more dynamic range but it seems to be shifted towards the shadow end, while highlights clip very easily. Conversely, the slide film seems to handle highlights a little better than digital but has much less shadow detail. This has not been a problem for the type of shooting I do with slide film, and it’s also not the same for every slide film. Velvia is one with the least amount of range and Astia is a much better film for shooting high contrast situations. Even compared to Astia, I still think digital would have more shadow range. Overall, that's a good thing, but it’s also another thing that complicates post processing. You have to learn to underexpose and then pull up the shadows to prevent blowing highlights in a way that isn’t necessary with film. As far as I’m concerned, for shooting landscapes, this difference is not important.

In conclusion, what I’ve decided is that I need to spend more time learning to get the color I want from my 450D because if I can pull off that feat it will be worth it. I’m not sure I’ll ever be able to make digital look like Velvia, but I’m confident I can do better than these examples. If I can get the colors I want from my 450D files, I think I’ll be able to make prints at 12 x18 that look better than 35mm slide film scanned on the Coolscan, and I may even be able to push some images to 16x24 and still be happy. Color aside, and resolution very close to equal, I think digital will allow for slightly larger enlargements of landscape images because of the lack of grain and extra sharpness. Now I just need to make some photos worth printing large so I can see it on paper.

Comments (1)

#8

This scene was shot with the 50mm Zeiss and I tried my best to duplicate the framing with the Tamron with out moving the tripod in between shots. The best of the digital versions of this scene was the one shot at f/5.6, and the best of the film shots was at f/8.

Here are the resized edits of each shot: digital first and a film second.





As you can see there are several things about this scene that are significantly different from the previous examples. Obviously, the subject matter is very different, but the difference in lighting is also important. This shot is an interesting example of dynamic range.

Once again, I edited the digital version of this shot before seeing the film, but this time I actually made the digital version a little too warm. I plan to go back and work on the digital version again to see if I can get that nice blue sky from the Velvia shot and hopefully make that yellow paint look a little less like Velveeta.

I shot the film version first and I went with the exposure the camera chose in Aperture Priority Mode. You can see the fender over the rear wheel is reflecting the sun and is overexposed. When I started shooting the digital camera, I figured I’d take advantage of the histogram and see how far I had to go to keep that spot from blowing out. Well, I stopped at –2 on the exposure compensation and it was still slightly blown out. The shot I chose to work with was –1. Unfortunately, I didn’t record shutter speeds on the film shots, so they’re not directly comparable. But, I can tell you that in the first few shots with soft lighting, the meters were pretty much identical. Because the digital shot was mostly underexposed I had to pull up the shadows in post processing and interestingly, I still got more shadow detail than I did from the slide film. With out the hard data it’s kind of a stretch to say this, but as far as I’m concerned this supports what I already knew: digital has much more dynamic range in the shadows, but slide film still holds more detail in the highlights. For high contrast scenes, you just have to underexpose the digital and take advantage of that shadow detail. This example is so extreme even that wasn’t really enough. For slide film, you have to choose where you’re going to loose detail. However, a polarizer filter and/or a better selection of compositions would probably have helped equal things out.

Here is a shot that shows the location of the upcoming detail crops:



For both images, the point of focus was on the “WP” sticker near the middle of the shot.

Here is the first set of crops: digital first and film second.

Digital Sized:




Film Sized:




Here is the second crop area:

Digital Sized:




Film Sized:




And the last crop area:

Digital Sized:




Film Sized:




Well, that’s the last example I have. I hope this has been informative or at least a little interesting. If you’re one the many film shooters who just likes film for “the look,” hang on for a bit. My next post will be a conclusion of what I think I saw in these four examples, but after that I plan to do some more work on editing for color to see if I can get the most out of the film shots and then to see if I can duplicate “the look” of the film shots in the digital shots.

Comments (2)

#7

This scene was shot with the 28mm Zeiss and I tried my best to duplicate the framing with the Tamron with out moving the tripod in between shots. The best of the digital versions of this scene was the one shot at f/11, and the best of the film shots was at f/16. This is more along the lines of what I expected, considering the differences in the DOF for each format and also from the results of the 28mm lens test.

Here are the resized edits of each shot: digital first and a film second.





For this scene the editing was significantly more difficult than the last two. The cloudy sky behind the tree presented quite a challenge…one that I probably should have just avoided all together. Of these two, the film images where actually quite a bit more difficult to edit because the cloudy sky cast a very blue color on the scene. That’s easily corrected with the digital RAW conversion but for the film image I had to spend a lot of time in Photoshop tweaking the color balance layer. And, I still don’t like it. I think I overdid the saturation a little too.

Now, here is a shot that shows the location of the upcoming detail crops:



For both images, the point of focus was set to infinity, which was somewhere just in front of the trees.

Here is the first set of crops: digital first and film second.

Digital Sized:




Film Sized:




Here is the second crop area:

Digital Sized:




Film Sized:




And the last crop area:

Digital Sized:




Film Sized:




That’s all I have to show for this scene. The next scene is very different.

Comments (2)

#6

This scene was shot with the 28mm Zeiss and I tried my best to duplicate the framing with the Tamron with out moving the tripod in between shots. The best of the digital versions of this scene was the one shot at f/16, but for some reason the best of the film shots was at f/8. I would have expected this to be opposite. I checked closely and there was nothing in any of the shots to indicate any errors could have caused this. Oh well. That’s why I’m showing several examples.

First, here are the resized edits of each shot: digital first and a film second.





For this scene the editing was pretty similar to the last scene, because they’re very similar subjects and were shot in the same light. I bumped up the contrast for both images, using the Levels tool and masking out the whitest part of the blurred water. I also bumped up the saturation and tweaked the color balance a little, but again, I’m not happy with that part. I think the colors should probably be somewhere between these two edits.

Now, here is a shot that shows the location of the upcoming detail crops:



For both images, the point of focus was on the rock just right of center, also known as infinity.

Here is the first set of crops: digital first and film second.

Digital Sized:




Film Sized:




Here is the second crop area:

Digital Sized:




Film Sized:




And the last crop area:

Digital Sized:




Film Sized:




That’s all I have to show for this scene. The next two scenes will be a little different.

Comments (0)

#5

This first scene was shot with the 50mm Zeiss and I tried my best to duplicate the framing with the Tamron with out moving the tripod in between shots. Both shots were taken at F/11.

First, here are the resized edits of each shot: digital first and a film second.





As I stated in the intro posting, I edited the digital file before seeing the film, but after looking at these, I plan to go back and redo the edits to try to improve the colors on both images. For this scene in particular, the editing was pretty simple. I bumped up the contrast for both images, using the Levels tool and masking out the whitest part of the blurred water. I also bumped up the saturation and tweaked the color balance a little, but that’s the part I’m not happy with. More on this later.

Now, here is a shot that shows the location of the upcoming detail crops:



For both images, the point of focus was on the close edge of the rocks near the center of the image.

Here is the first set of crops: digital first and film second. The first two are sized for 100% crops of both images at 12MP and the second set are 100% crops of both images at about 20.6 MP. From now on I will refer to this as Digital Sized and Film Sized.

Digital Sized:




Film Sized:




Here is the second crop area:

Digital Sized:




Film Sized:




And the last crop area:

Digital Sized:




Film Sized:




That’s all I have to show for this scene.

Comments (1)

#4

Some of you probably already know what this is about, but if you don’t, please see my previous blog entries, starting with the very first one. Basically, I wanted to see the differences between my own 35mm film capabilities and my own DSLR capabilities, specifically for shooting landscapes. I don’t think I followed exactly what I said I was originally planning to do, so the purpose of this introduction post is to explain what I did and to give the usual disclaimers. There will be several posts following this one that show the actual results. Please withhold comments and judgments until you’ve seen all of the examples.

First and foremost, this comparison is not meant to declare a winner or to make any general claims about which is better. You may catch me saying which I think is better for a specific purpose for a specific reason, but even then it’s only what is better for me. Film and digital are both great tools and I plan to continue using both. I recommend you do the same.

My methodology for this comparison was simple. I went out and found several different compositions and shot them with both film and digital. I shot six different scenes and I’m going to show you four of them. The film camera used was a Yashica FX-D (Contax mount) with a Zeiss 28mm f/2.8 Distagon and a Zeiss 50mm f/1.4 Planar. Film used was Velvia 100F and I scanned it with a Nikon Coolscan V. The digital camera was a Canon 450D with a Tamron 17-50 f/2.8 lens.

I did my best to duplicate the framing as exactly as possible for each shot but they are not perfect. One reason they’re not perfect is because of my scanner. It always chops a little bit off the edges. I shot each scene with each camera/lens at three or more different aperture settings and I later compared the shots to pick the sharpest. I also bracketed exposures a little bit to make sure I got a decent exposure, and again, I picked which was best for each camera. So, I ended up with at least six shots of each scene on the film camera, and probably more on the digital.

The Canon 450D produces 12.2 MP images and the Coolscan V scanner produces film scans of about 20.6 MP. In order to account for this difference I plan to show two sets of of all the detail crops. The first set will be a 100% crop of the digital image and a 100% crop of the film image downsized to the same size as the digital image. The second set will be a 100% crop of the film image and a 100% crop of the digital image upsized to the same size as the original film scan.

Editing for the photos was intended to be realistic, but frankly, I’m not totally happy with all of them. I edited all of the digital files before seeing the film. After scanning and editing the film files, I realized that I have a tendency to edit the digital files “cooler” than they should probably be. Because of this, I like the general look of the film images better. BUT, I realize this is basically a deficiency in my own editing skills. Getting good color from the film scans just seems easier to me, most of the time, but after I finish posting all of the detail shots, I plan to go back and re-edit all of the film and digital files to try to make some improvements. This part is obviously very subjective, so take it for what it is.

Considering that I have invested about the same amount of money into film and digital photography over all, I hope that you can see that I have no particular agenda. I’m honestly just trying to make the differences between these two formats apparent to myself so that I can make the best use of the equipment I own. I have definitely learned something from this process. I hope that you can learn something from seeing my results.

Comments (0)

#3

If you're wondering what this is all about, please refer to my first two blog entries. I'm going to cut to the chase and start posting comparison shots. These were taken at the same time as the 50mm comparisons, under the same conditions with the same details except as mentioned. Instead of using a C/Y Zeiss AE 50mm f/1.4 Planar, this comparison is done with a C/Y Zeiss AE 28mm f/2.8 Distagon and the Tamron lens set as close as possible to to the same focal length.

First, here are the full frame shots of the juniper tree, resized:

Tamron near 28mm


Zeiss 28mm


Sorry about the sky. I let it blow out on purpose to make sure the dark green tree was exposed properly. Exposures are identical to the 50mm test, and as far as I'm concerned, the same difference in contrast and color is exhibited here.

Now for some details. These are all 100% crops from the center.

Tamron on the left and the Zeiss on the right.

f/5.6


f/8


f/11


Once again, I see less contrast and color saturation from the Zeiss lens. Besides that, I see very little difference in sharpness.

Now for the painting shots.

Here are full frame images, resized.

Tamron at 28mm:


Zeiss 28mm:


Now for some details. This first set is 100% crops from the same center portion of the painting...just a little left and above center:

Center comparison:


Again, I can barely see any difference in sharpness but there is ever so slightly less contrast in the Zeiss images.

Corner comparison:


Now there are some obvious differences. If you've done much research on this Zeiss lens, as I have, you have probably read that the corner performance is less than stellar, but I was still surprised to see how big the difference is. Not only is there less contrast and less saturated colors, but sharpness is way off and even at f/11 the Tamron still looks better. Let's hope the Zeiss does better on film.

Once again, this completes my non-scientific comparison of lenses. I hope someone was able to get something out of this. What I got out of this is that for shooting subject matter like landscapes, where image sharpness is more important than some intangible "character" quality, the Tamron is a clear winner on the 450D.

Thanks,
Paul

Comments (1)

#2

If you read my Introduction entry you'll know what this is about. In short, I decided I want to do a film and digital comparison and one of the main concerns was that I won't be using the same lenses for the film shots as I'm using for the digital shots. My intent is for the comparison to be practical and in practical situations we don't use the same lenses for the same tasks on cameras with different format sizes. But to ease some of those concerns I decided to shoot some comparison shots to show how these different lenses render the same scene on the digital camera only. Even if you're not interested in the film and digital comparison I'm working on, you might find this lens comparison interesting.

Equipment Details:
Canon 450D
Tamron 17-50 f/2.8 lens
Contax mount Zeiss AE 50mm f/1.4 manual focus lens
Roxsen AF confirm Contax-EOS adapter

Technique Details:
All shots were taken on a sturdy Bogen tripod using the mirror lock up function.
All focusing was done using live view zoomed in to 10X
Raw files were converted to TIFF files with default DPP settings, "Faithful" picture style, and no sharpening.
The first set of the juniper tree was taken with auto white balance and the second set of the painting was a custom white balance.
In photoshop, I applied the same level of sharpening to all the 100% crop shots and made no other changes except for cropping.

For the first test scene, I was thinking only about center sharpness. I thought comparing corner sharpness wasn't really fair considering that the cropped digital sensor is chopping off the corners for the full frame film lenses. Here are those shots, but hang tight, I went back and shot another test later to show the corners.

Full frame images, resized:

Tamron at 50mm


Zeiss 50mm


First, you can tell that the Zeiss 50mm is actually a little longer than the Tamron at 50mm. The only other thing that I noticed is that the Tamron seems to have better color and contrast.

Now for some crops. These are all 100% crops from the center of the frame.

Tamron on the left. Zeiss on the right.

f/5.6


f/8


f/11


You can see that the difference in contrast and color is still apparent in the 100% crops. This was kind of a surprise to me after everything I've read about Zeiss lenses. This is supposed to be one of the best 50mm lenses you can buy. My guess, and it is just a guess, is that this is caused by internal flare reducing micro-contrast, and maybe if the Zeiss were updated with newer coatings it might not be an issue. But, who knows. As far as sharpness is concerned, they look pretty equal to me, both peaking out at f/5.6 and decreasing as diffraction sets in at smaller apertures.

Now for some tests that show center and corner performance. The subject this time is an original oil painting done by my mom and donated to me when I went off to college. (Check out her blog if you want to see more of her paintings.) For these shots, I tried to correct for the difference in focal length by moving the tripod slightly, but as you can see, framing is not exactly perfect.

Here are the full frame images, resized:

Tamron at 50mm


Zeiss 50mm


These shots show the same differences that are apparent in the first set, the Tamron seems to have better color and contrast. Now for some detail shots.

The first set shows Center sharpness. They are 100% crops from just to the left and slightly above center. I chose this spot because it had a little more texture than dead center, but it's close enough.

Center comparison:



As far as I'm concerned these images are so close that the differences couldn't possibly matter in real world results.

This second set shows Corner sharpness. These are 100% crops taken from the top left corner of the painting.

Corner comparison:



The corner shots show a little more of a difference than the center did. In my opinion, I'd say the Tamron is slightly sharper all the way through, with both performing best at f/5.6 and softening every so slightly as they are stopped down further.

Considering the respective reputations of Zeiss and Tamron, I'd say Tamron has something to be proud of. I knew the Tamron was a good lens when I first started to play with it, but frankly, I'm surprised it held up this well against the Zeiss primes.

Well, there it is folks. Nothing scientific about it, but you should have a reasonable idea of the differences between these lenses now. How it will all pan out in the film and digital comparison, I don't know.

Stay tuned for another comparison between the Tamron 17-50 and my C/Y 28mm Distagon.

Oh, and please feel free to leave some comments and let me know what you think.

Thanks,
Paul

Comments (3)

#1

Howdy Folks,

Welcome to the first entry of my first blog. I suppose I should try to lure you in with promises of regular posts on topics that every photographer is just dying to read about, but for the sake of getting things off on the right foot I'll let you know right now, this is not that kind of blog. I'm not promising anything. I could tell you what my long term plans are but lets face it, we're just getting to know each other. No one wants to jump into a commitment this soon. Heck, I may not even finish this next sentence out of pure boredom. Okay, maybe the next one. Well, whatever. If you made it this far, I suppose you at least deserve to hear my short term plans.

If you were lurking in the Viewfinder forum last week you might have seen my poll titled, "Want to see Film and Digital Compared?" You can see it here if you missed it. The thread explains the details for a friendly little comparison which I was planning to do for my own good before deciding to poll the group to find out if anyone else would be interested in my results. But it occurred to me that maybe the forums aren't the best place for this kind of thing. Thus, the glorious birth of what the internet really needs more of, a new blog.

Before I get to the real comparison I want to address one of the main points of contention from my poll thread, the fact that the comparison photos will be taken with different lenses. This will be the topic of my first real blog entry, which should follow shortly.

After I'm done with this silly comparison thing, I'm sure I'll find some other useless information to bore you with. What will it be, and when will I post it? Your guess is as good as mine, maybe better.

Paul
aka: photophorous

Comments (2)