Puts on M8 vs film

[giordano]

Advertisement (gone after registration)

Anyway the definition of a linepair is a black and white line next to one another. That s two adjacent pixels. That means dividing by two,not some other number.

 

Not really. With a conventional sensor you can only achieve one linepair per two pixels when the image of each succesive line falls precisely on alternate rows or columns of pixels. I.e. the lines in the target must be precisely parallel to the sensor grid and exactly the right distance apart. In any other situation you get aliasing.

 

So dividing by two is unrealistic. IMHO you need to divide by at least 3, maybe 4 to get a figure comparable with film, where resolution is independent of the direction of the lines.

[mjh]

Are you sure that diagonal resolution of the sensor is the same as horizontal?

With sensor pixels arranged in rows and columns, diagonal resolution can be up to 40 percent higher than horizontal and vertical resolution (by a factor of sqrt(2), to be exact). This theoretical limit is rarely reached in practice, and some cameras (Sony’s, for example) have fairly low diagonal resolutions, but generally the diagonal resolution is about 20 percent higher. According to my measurements the M8 resolves 51 lp/mm, averaged over the horizontal, vertical, and diagonal resolution.

Edited July 26, 2009 by mjh

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[sbelyaev]

1. Number of active pixels (3970) divided by 27mm is 147pix/mm.

147 divided by 2 is 73.5.

 

2. Diagonal resolution is always lower than horizontal in sensors with square or rectangular shape of the pixels, for the diagonal length of each pixel is always greater than horizontal or vertical.

 

3. I agree that the resolution of M8 is about 50-60lp/mm (as good as Kodak Gold).

 

4. Some should consider going back to basics and read an article from LL. Res-Demyst

 

5. Delta 100 is a winner )

[BerndReini]

Jamie, you know I love my M8 and use it 95% of the time, but as far as dynamic range goes with the M8 vs Ektar, I have to agree to disagree:

 

Ektar (not sharpened for output, sorry)

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[markowich]

The DR of the M8 is 11 stops, double of slide film and about the same as negative film.

 

top digital cameras have a DR of close to 13 stops, which beats slide film and negative film by quite a margin.

there is just no way back, except maybe for mr puts.

peter

[pgk]

I may have posted this elsewhere previously but anyway....

 

Last year I carried out a project looking at British marine gobies and the possibility of identifying them using digital photography. I used a Canon 5D for the project, however I also had some images from the same location shot some years before on a film camera using Velvia 50. Bottom line is that the 5D files provides more and better information, enabling identification to be confirmed where this was possible from the images, whilst the Velvia slides do not always allow a conclusive identification even from similarly composed shots. This was a 'real world test' and as far as I am concerned 10MPixels produces more information rich files than 35mm fine grain slide film (given similar optics, etc.). You can argue nuances with fine grain B&W film, careful exposure, processing, etc., etc., but IMHO a real world situation is far more revealing of bottom line actual usable information/detail. Given the choice I'd take an M8 file against a scan anytime!

[BerndReini]

Advertisement (gone after registration)

I don't know who does these tests measuring DR. Maybe it is just a matter of the roll off of highlights and how they gradually "bloom" with films vs sharply clipping and disappearing with digital formats, but this theoretical dynamic range of digital sensors simply falls short in real life applications. I shoot with the M8 for convenience but I always get excited when I capture a great photo on film in contrasty light because I know that the print will have a quality I still cannot achieve with any digital camera I have used. Michael Reichmann praises the PhaseOne 65+ back as the answer to his dynamic range struggles, but that is simply too large of a camera for my style of shooting.

 

I used to get into those discussions of film vs digital and I agree that in theory, the M8 has all the advantages. A lot of times though, I still go out with my M7 or MP and shoot Ektar because the results are amazing. I can only recommend that instead of researching this topic on the internet, everybody should go out, shoot a couple of rolls of film with their Leica lenses and get them processed, and maybe even one frame drum scanned or optically printed by a good lab. It will give you a definite answer to the question which you prefer, maybe not which is better.

[BerndReini]

The highlights in the car in my picture and the blue sky would not have looked this way with the M8! Maybe not worse to some people, but definitely different, and no amount of post-processing can change that.

[sbelyaev]

Even-though the article clearly shows superiority of Delta 100 many feel that it should not be happening based on their experience.

I guess quantum mechanics is correct when is comes to omnipotent reality. There is no single correct reality/truth. Our faith / believes can modify pretty much anything.

[microview]

I have a theory... Film emits an odourless, colourless gas that has a calming effect on photographers. Film photographers are hence far more laid-back and generally calmer people, getting a fix every time they raise their cameras in front of their faces. Those who have embraced digital, on the other hand, have the opposite to contend with l

 

I have a theory... the M8 emits an odourless, colourless gas that has a calming effect on photographers. M8 photographers are hence far more laid-back and generally calmer people, getting a fix every time they raise their cameras in front of their faces. Those who have embraced the M6 or M7, on the other hand, have the opposite to contend with, especially when they try to load or unload their films. That's why some make the interim move to an MP, which at least has a decent rewind knob.

[jaapv]

Not really. With a conventional sensor you can only achieve one linepair per two pixels when the image of each succesive line falls precisely on alternate rows or columns of pixels. I.e. the lines in the target must be precisely parallel to the sensor grid and exactly the right distance apart. In any other situation you get aliasing.

 

So dividing by two is unrealistic. IMHO you need to divide by at least 3, maybe 4 to get a figure comparable with film, where resolution is independent of the direction of the lines.

But there the resolution is less geometric, resulting in wavery lines, degrading the theoretical maximum.

I thought the megapixel wars were over, with resolution, once above a certain level being totally irrelevant to image quality (according to Mr. puts this is at about 30 Lp/mm) So what is the point of this whole discussion and of this article:confused:. and will somebody please explain to me the relevance of a resolution of 73 Lp/mm as opposed to 75 Lp/mm:rolleyes:

[xrogers]

I've got some 14x21" high-key silver prints on my wall, done in my darkroom, shot with an M6 on Delta 100. These prints have detail and luminosity I too often cannot achieve with the M8 at the same size. Film wins in these situation for ease-of-use as well (I don't have to shoot/chimp histogram/repeat to ensure that I hold the highlights).

 

The M8 is my primary photographic tool, and it usually does produce a superior result. But in my experience there are situations where even 35mm film can provide a better print with less effort.

 

Until later,

 

Clyde

[Jamie Roberts]

The highlights in the car in my picture and the blue sky would not have looked this way with the M8! Maybe not worse to some people, but definitely different, and no amount of post-processing can change that.

 

Bernd, this is a thread hijack and nothing to to with Put's observations, but I would absolutely challenge your assertion, at least from the posted shot; the look you've got in that photo can be very easily reproduced with digital (and IMO all-too-often is).

 

To get that look, you must expose for the sky / highlights, then flatten the contrast in the upper quartertone (shoulder) and open up the lower quartertone (toe)--just like film does. You also need a corresponding print curve (either for the screen or for a print).

 

The M8 at ISO 640 or lower absolutely has the shadow response to do exactly what you've done here. Lightroom--though generally terrible with colour and the M8--even has the funky "fill light" thing they inherited from RSP that does pretty much exactly that (I don't like the results there though because of what it does to the midtones, as well as the previously mentioned awful colour )

 

Personally, I'd prefer keeping the midpoint absolutely in the center of the resulting curve. But it absolutely can be done, and it's not "post-processing" either; you get can that look right at the RAW stage.

 

More work to do this in digital than film? Perhaps that's true. I think of it as different work, but with digital we're all our own lab technicians. That's more for a lot of people

[BerndReini]

I use Capture One Pro as my Raw Converter, and yes I have also tried Raw Developer. When you expose this shot for the sky, the kids faces are gone. This is a very high contrast situation with the family walking alongside a tall building in the shade. I have been shooting a project on Hollywood Blvd. for two years now and know exactly what the light is at certain times of the day. I have shot (and exhibited) shots taken with the M8 in similar light in the same location, and I know what I'm doing exposure wise. I expose to the right of the histogram as far as possible without clipping the highlights and then push the shadows to recover as much information as possible. What I can tell you, and I don't know whether you have shot a lot of Ektar, is that the curve of the M8 is a lot steeper than that of Ektar, meaning that the highlights and shadows are a lot closer together and the contrast a lot flatter for a significant range of exposure values. So when I scan, the highlights are darker and the shadows lighter than with the M8. Once I bring the shadows and highlights closer together in the M8 files for printing (by lightening the shadows) there is a significant amount of color noise to deal with. Most importantly though, look at the highlights in the car: they are pure white and the area around it gradually fades into white. This does not happen with the M8. When the whites clip, it gets ugly, and when you let this happen on skin tones, you might as well delete the image. On film however, you can overexpose certain areas of the skin to pure white, and the areas around it will naturally fade into it and create a beautiful image. This is especially important when you have multiple people in the shot with multiple complexions in bright sunlight.

[plasticman]

Bernd - I'm with you on this one - though I would've posted the image of the blonde woman running across the LA street that you posted a while ago, rather than the image in this thread.

 

I personally think these film vs digital threads are pretty much played-out by now, and I wouldn't have bothered were it not for the frankly laughable assertion that cellphones offer the same quality as an M7 loaded with film.

 

Film and digital are very different - both have advantages and disadvantages. Sure there's a lot of tweaking that can be done in post to achieve similar results from both (and by this I mean digital and film can be tweaked), but if you are shooting detail in deep shadow and hoping to retain the charm of the areas in bright sunlight, there's something about the way that film behaves that lends the image an aura or glow that digital doesn't match.

This isn't to say that digital cannot cope with the contrast when handled carefully, or when exposed for different areas. But film has magic. imho.

[georg]

Dynamic range isn't that easy to measure, so I won't say specific numbers (xx stops of DR) but it's a huge problem for electronic devices and the reason why there wasn't much progress over the years. Film is chemical and works on very different fundamental principles.

 

Negative film is rarely used in professional still photography today, but it's the leading medium in professional, medium/high-budget cinematography. The films are different and so are the scanners and digital cameras, but the basic rules of these technologies can be applied to still-photography.

It takes very skilled cinematographers to hide the shortcomings of todays digital-cameras with special lighting.

 

Here they made a comparison of various formats, it's about grain, resolution, DoF but also DR. the "HDCAM"-camera is a >100.000$ 3CCD-high-end-camera which represents a very high DR for digital sensors. The film is of course scanned and therefore limited in it's DR (high-end scanners take an additional image -2EV to make one 16bit-file out of two 14bit-scans - here they used a simpler version), but it's really simple to see the differences:

 

http://www.ecctv.de/videos/hdtv/mountain_top_reasons.wmv (650MB!)

 

Watch the skin tones @ 5:10 - 5:20

The bright parts in the background @ 16:25 - 16:34

 

Well, it's not still-photography, but it's the best comparison between these two worlds on the net.

 

By the way, ARRI (the leading company in the film-industry and regarded as very engineering-driven and trustworthy) started to work on very scientific, yet understandable tests to compare various aspects of image quality with various systems. They already created a very interesting paper regarding resolution. It focusses on cine-technology but it also explains fundamental aspects of resolution, MTF etc. very well. http://www.arri.de/fileadmin/media/arri.com/downloads/Camera/Tutorials/SystemsTechnologyBrochure.pdf

They measured about 80lp/mm resolution and about 60lp/mm @ 10% contrast in the scan.

They're currently working on a new method to measure DR, let's wait what they have to say!

Edited July 27, 2009 by georg

[BerndReini]

It's true, the film vs digital debate is much more debated in motion picture except that everyone seems to acknowledge in the film industry that film is still superior. Here the debate is based purely on budget considerations. I am a cinematographer and shoot digital almost exclusively now, and I definitely would prefer shooting film if I could.

[scaryink]

It's true, the film vs digital debate is much more debated in motion picture except that everyone seems to acknowledge in the film industry that film is still superior. Here the debate is based purely on budget considerations. I am a cinematographer and shoot digital almost exclusively now, and I definitely would prefer shooting film if I could.

 

Bernd, the other issue is viewing of the output. Most people here only look on the computer screen and judge results strictly from this display.

 

It is a well known that display technology is very very basic. Edward Tufte in his lectures strongly suggests that all computer users get as a minimum, 2 of the largest highest res monitors you can afford. As he describes it, our displays have the comparable resolution to cave men hammering on a rock. No kidding, I absolutely agree.

 

If your screen cannot display the resolution, it turns it to mushy noise. People view the noise as proof that the analogue capture is inferior.

 

When viewing a movie projected, there is an enormous difference in digital versus film output.

 

It is hilarious to see people spending those ungodly amounts of money on the new leica projector with such low resolution. My $ 200 film projector (a Leica may I add) just knocks the crap out of these digital cave dwellers.

 

I understand the convenience factor at the right price, but does anyone even care about final output quality other than a computer screen or digital display?

 

There is also no question that negative film has a much greater dynamic flexibility than digital capture. I also find that positive film has more latitude than any of my digital cameras, the Leicas and Canon eos 1ds included.

 

I was just in mesa verde national park and had to be extremely careful in exposure with M8. If I had taken my film camera there would have been alot more freedom and room for error.

 

The comments from digitalphiles are similar to those who can't hear the difference between mp3 and live music. It is the dumbing down of our native senses for the sake of convenience.

[BerndReini]

Hahaha, I have a feeling you will get slammed in a second, but thanks for being so direct. On a side note, I just got my new Macbook Pro with the matte LED screen. It is pretty amazing as far as displays go, and it really shows me a fairly realistic representation of my prints when sooft-proofed.

[Advertisement]

Hallo,

Look here: Leica M

[KM-25]

Film is difficult to master. A thorough understanding of the Zone system is often necessary for having complete control over B&W films...and working with transparency film really requires some form of consistent home processing system (like a jobo or photo therm) to obtain consistent color and reliable processing times. Once a photographer has a thorough understanding of his film-stock and processing workflow....Then he can achieve amazing print quality and enlargements with small format as long as he gets accurate exposures and sharpness during the capture process.

 

For me, it always used to take awhile to get to know certain film stocks before I was able to pull out great prints....And some films just never seemed to work well with my personal style of shooting and workflow. For example, I always had excellent luck with Kodak Tmax 100 and tech pan films, but not as much luck with Ilford. I love ilford prints from other people but could never vibe with it myself...That's just me and it took awhile to discover this.....Once I was able to commit to a handful of film stocks and know them well then my prints got much better. I've been shooting digital for 10 years now...and have still never made a print that can compare to my old Tech Pan prints

 

Anyway, I just wanted to post this for digital shooters that might be thinking of experimenting with film. Film takes a long time to master....and many of the film scans and images that we see floating around the internet are poor representations of what can really be achieved with film once a photographer has complete control over the capture and development process. It also takes quite a bit of experimenting before finding a film stock that vibes well with a photographer.

 

Film can be absolutely unforgiving for people that like to try a-little-bit of this and a-little bit of that. The shooters that just kinda pop any roll of film in their camera without doing much testing really have hit-and-miss results....generally miss results LOL

 

I am going to have to disagree with the whole film is difficult to master statement.

 

Yes, it takes a good photographer to get the most out of a given film, but it does not take as long as you think for everyone. For example, when I got my first roll of Kodachrome back from the lab, I was totally convinced it was the slide film of choice for me, I have images from that roll that when I look at them today, they really do look good. And I was 13 years old by the way when I shot that first roll.

 

As far as the slide film / control over processing thing goes, do it my self? No way. I have done that and all it does is get in the way of making more photos, that is what labs are for. A good E-6 lab or in the case of Kodachrome, K-14, will always be more consistent than most home brew labs. Black and white, yes, I do that my self, but even that gets left to labs at times. For example, I am doing a book project on a certain geological marvel right now. It is all being shot on either infrared or Tech Pan processed as slide shot in a Hasselblad. I am doing the infrared on my own, printing included. But the Tech Pan shots are simply stunning, consistently processed and then printed on Ilfochrome. I shoot, they process and print, nuff said.

 

While I do have and like to have control over image making and processing in photography ( not "capture", I hate that term. ) I also like to leave some things to chance, the wild experiment gone right. There is so much talk of control over things that I think many people miss out on the brilliant mistake and instead create some of the most boring and bland photographs you will ever see, all in the name of control.

 

There is often a ton of talk on these forums, but very very little in the way of powerful imagery that backs up these words with what really matters.

Edited July 27, 2009 by KM-25