Jump to content

Film Resolution equivalent


Adam Ray

Recommended Posts

18k = IMAX 15-perf (or 36K?)

12-13k = 70MM 5-perf

6K = 35MM 4-perf

3K = 35MM 2-perf

1.8K = S16MM

1.5K = 16MM

720p = S8MM

480 = 8MM


I had to share this with someone. I found so many people including myself looking for these numbers I had to compile this list. I cannot find anything close to this on the internet all in one place.


Digitally, IMAX is referred to as being three times better/larger than 35mm (18k / 6K).

On Film, IMAX is three times larger then 70mm and 70mm is twice as large as 35mm therefore it should be six times as large.


This is of course the best possible scenario of each format. These are all comparables, they cannot be absolutely perfect. I'm comparing category to category.


So much in the real world changes these numbers, for example cropping 16mm to a release format reduces the frame size and corresponding resolution or finding a camera and lens that can shoot S8mm or 8mm and come close to 720 or 480 quality.

Link to comment
Share on other sites

  • Premium Member

Welcome to the forum Adam!

 

We have the ol' resolution discussion on a regular basis.

 

The way you measure resolution on film is with a resolution chart, which is a bunch of lines that go from big to small. Some are lined up horizontally, some are lined up vertically. There are also circles of lines that do the same thing.

 

Film is also measured in "perceivable resolution", using the (MTF) Modulation Transfer Function process. Meaning, unlike digital which has a fixed pixel count that's easy to measure electronically, with film a human has to literally look for the lines of the chart and judge where you can't tell the separation between them anymore. So it really depends on the quality of the optics used to either project or the quality of the digitization system. Unlike digital, there is always a loss when measuring film resolution thanks to that issue, especially when it comes to negative first generation vs positive (projected) second generation.

 

The only real good test was performed in 2003, prior to our modern Vision 3 stocks, which are absolutely better in every way. It was done in both the US and in Paris, with different groups of researchers.

 

The test showed that a standard 4 perf 35mm frame in the Academy format contained 2400x2400 lines of visible resolution. This would mean 4 perf 35mm original negative contains 5.76 megapixel worth of information.

 

For every successive print though, the resolution dropped. So for IP it was 2100x2100. For IN it was 1700x1700. For release prints it was 1500x1500.

 

If you translate this math, which is as close to "accurate" as one can get without re-running the tests, you start to get a picture that looks like this:

 

15/70 = 70.41 mm × 52.63 = 7600x5700 (43 Megapixel)

5/70 = 52.63 by 23.01 mm = 5700x2500 (14 Megapixel)

4/35 = 22 by 16 mm = 2400x1750 = (4.2 Megapixel)

2/35 = 22 by 9.47 mm = 2400x1030 (2.47 Megapixel)

1/16 = 12.52 by 7.41 mm = 1360x800 (1.09 Megapixel)

1/8 = 5.79 by 4.01 mm = 631x430 (271 Kilopixel)

 

Now again, these numbers aren't very accurate based on modern film stocks. But they are "close" enough to get feeling of practical resolution. Obviously it would be nice to re-run the test with all of those film formats using modern stocks.

Link to comment
Share on other sites

35mm 4 perf and 35mm 2 perf are the same width. Why do they have different horizontal resolution?

2 perf is only using half the negative space of 4. I like the easy math on this one.

Link to comment
Share on other sites

Welcome to the forum Adam!

 

We have the ol' resolution discussion on a regular basis.

 

The way you measure resolution on film is with a resolution chart, which is a bunch of lines that go from big to small. Some are lined up horizontally, some are lined up vertically. There are also circles of lines that do the same thing.

 

Film is also measured in "perceivable resolution", using the (MTF) Modulation Transfer Function process. Meaning, unlike digital which has a fixed pixel count that's easy to measure electronically, with film a human has to literally look for the lines of the chart and judge where you can't tell the separation between them anymore. So it really depends on the quality of the optics used to either project or the quality of the digitization system. Unlike digital, there is always a loss when measuring film resolution thanks to that issue, especially when it comes to negative first generation vs positive (projected) second generation.

 

The only real good test was performed in 2003, prior to our modern Vision 3 stocks, which are absolutely better in every way. It was done in both the US and in Paris, with different groups of researchers.

 

The test showed that a standard 4 perf 35mm frame in the Academy format contained 2400x2400 lines of visible resolution. This would mean 4 perf 35mm original negative contains 5.76 megapixel worth of information.

 

For every successive print though, the resolution dropped. So for IP it was 2100x2100. For IN it was 1700x1700. For release prints it was 1500x1500.

 

If you translate this math, which is as close to "accurate" as one can get without re-running the tests, you start to get a picture that looks like this:

 

15/70 = 70.41 mm × 52.63 = 7600x5700 (43 Megapixel)

5/70 = 52.63 by 23.01 mm = 5700x2500 (14 Megapixel)

4/35 = 22 by 16 mm = 2400x2400 = (5.76 Megapixel)

2/35 = 22 by 9.47 mm = 2400x1030 (2.47 Megapixel)

1/16 = 12.52 by 7.41 mm = 1360x800 (1.09 Megapixel)

1/8 = 5.79 by 4.01 mm = 631x430 (271 Kilopixel)

 

Now again, these numbers aren't very accurate based on modern film stocks. But they are "close" enough to get feeling of practical resolution. Obviously it would be nice to re-run the test with all of those film formats using modern stocks.

Looks like the numbers aren't far off except for the lower end 8mm category.

 

What is the equation to get megapixels from resolution?

Link to comment
Share on other sites

Welcome to the forum Adam!

 

We have the ol' resolution discussion on a regular basis.

 

The way you measure resolution on film is with a resolution chart, which is a bunch of lines that go from big to small. Some are lined up horizontally, some are lined up vertically. There are also circles of lines that do the same thing.

 

Film is also measured in "perceivable resolution", using the (MTF) Modulation Transfer Function process. Meaning, unlike digital which has a fixed pixel count that's easy to measure electronically, with film a human has to literally look for the lines of the chart and judge where you can't tell the separation between them anymore. So it really depends on the quality of the optics used to either project or the quality of the digitization system. Unlike digital, there is always a loss when measuring film resolution thanks to that issue, especially when it comes to negative first generation vs positive (projected) second generation.

 

The only real good test was performed in 2003, prior to our modern Vision 3 stocks, which are absolutely better in every way. It was done in both the US and in Paris, with different groups of researchers.

 

The test showed that a standard 4 perf 35mm frame in the Academy format contained 2400x2400 lines of visible resolution. This would mean 4 perf 35mm original negative contains 5.76 megapixel worth of information.

 

For every successive print though, the resolution dropped. So for IP it was 2100x2100. For IN it was 1700x1700. For release prints it was 1500x1500.

 

If you translate this math, which is as close to "accurate" as one can get without re-running the tests, you start to get a picture that looks like this:

 

15/70 = 70.41 mm × 52.63 = 7600x5700 (43 Megapixel)

5/70 = 52.63 by 23.01 mm = 5700x2500 (14 Megapixel)

4/35 = 22 by 16 mm = 2400x2400 = (5.76 Megapixel)

2/35 = 22 by 9.47 mm = 2400x1030 (2.47 Megapixel)

1/16 = 12.52 by 7.41 mm = 1360x800 (1.09 Megapixel)

1/8 = 5.79 by 4.01 mm = 631x430 (271 Kilopixel)

 

Now again, these numbers aren't very accurate based on modern film stocks. But they are "close" enough to get feeling of practical resolution. Obviously it would be nice to re-run the test with all of those film formats using modern stocks.

My mistake the equation is right in front of me.

Link to comment
Share on other sites

  • Premium Member

The "K" refers to horizontal resolution only. So 2-perf, 3-perf, and 4-perf all have the same horizontal resolution (ignoring the small difference between Full Aperture / Silent and Academy Aperture / Sound width, i.e. 24mm versus 21mm wide.)

 

There's a difference between optimal scanning resolution and measurable resolution -- in theory, to avoid aliasing, you'd have to factor in Nyquist and scan at twice the resolution (frequency) of the original. However, most real-world detail doesn't create the worst-case scenarios for aliasing so 2X is a bit overkill.

 

Traditionally, 35mm has been scanned at 2K, 4K, sometimes 6K. Some people who have shot line measurement charts using 35mm color negative get somewhere like 3.5K in terms of measurable horizontal detail, but much real-world photography can have less resolution than that. Nyquist would tell you that if the image resolves 3.5K, you have to scan it a 7K to have zero aliasing, but for most images, 4K is enough and 6K is considered optimal, what is sometimes used for archiving work. But most projects that scan 35mm at 6K will then downscale it to 4K for the rest of post work and mastering.

 

5-perf 65mm is around 48mm wide, roughly 2X the width of the 35mm frame.

 

15-perf 65mm (IMAX) is the equivalent of three 5-perf 65mm frames, but those frames are rotated by 90 degrees, the film runs horizontally through the camera, so the total width of IMAX is around 70mm, which is almost 3X that of 35mm, not quite.

 

So in terms of measurable resolution, if I want to be generous and say that 35mm can resolve 4K, then 5-perf 65mm becomes 8K and 15-perf 65mm IMAX is almost 12K. If you want to be less generous and say that 35mm resolves more like 3K, then 5-perf 65mm becomes 6K and 15-perf 65mm IMAX is almost 9K.

 

In terms of optimal scanning resolution, if 6K is optimal for 35mm, then 12K is optimal for 5-perf 65mm and almost 18K is optimal for 15-perf 65mm IMAX.

 

If you think that 4K is good enough for scanning 35mm, then 8K is good enough for 65mm, and 12K is good enough for 15-perf 65mm IMAX.

 

I also think that due to the MTF of actual lenses used for medium-format photography, I'm not sure you actually get double the resolution for 5-perf 65mm and 3X the resolution for IMAX.

Link to comment
Share on other sites

This is an old and tired notion - that there's a fixed "resolution" for different gauges. It really needs to die, because it's simplistic and doesn't actually contribute anything to the question of what resolution one should scan at. Part of the problem is that charts like the one in the original post are an answer to the wrong question.

 

 

Film doesn't have pixels so you can't say X film gauge is Y pixel esolution. For one thing, "resolution" means different things in different contexts. In the digital world, it's a number that describes the pixel count, and nothing more. In the optical world, it's about how much detail can be resolved on the film. This is something that's affected by everything from the quality of the lenses being used to the type of film stock, to the aperture and film speed settings, to the lighting, to how the film was processed. There are so many variables, that trying to say in a blanket way that a given gauge is some specific digital resolution is bordering on nonsensical.

 

Look at it this way: If you scan a film at both 2k and at 4k, then put the 2k image on a 2k screen, and right next to it put the 4k scan on a 4k screen of the same physical size, you will see no appreciable difference in quality between the two. That's why scans to SD can look pretty damned good on a good SD monitor - you're viewing the scan on the appropriate screen for the resolution of the source file.

 

Now if you take that 2k scan and put it on the 4k monitor, it will look softer. Why? Because something in the signal chain has to scale that image up about 4x to fit the larger resolution 4k screen. That's making something up out of nothing. Scaling algorithms can be pretty good, but it'll never look as good as the same film scanned at its native resolution.

 

This same idea applies if you want to do a digital projection of film on a large screen. You're going to get a substantially better image if you scan Super 8 at 4k and then project it 30 feet high then if you scan it at the recommended 480px in your chart above. The 480 scan is never going to hold up to that kind of scaling. The 4k scan will.

 

 

  • Upvote 1
Link to comment
Share on other sites

  • Premium Member

are the "resolution" numbers based on which mtf response? 30%, 20%, 15%, 10%, 0%, etc. ?? film and video handle fine detail contrast differently which is one of the reasons why film image may often look "softer" and less detailed even if it is technically more "sharper", its mtf curve just drops generally faster than video's so that the small details are presented with lower contrast compared to video which gives the impression of small softness compared to video even if the video image is much less detailed near the 0% mark.

 

the whole concept of resolution comparisons between film and digital is a bit unnecessary unless one has to convince a producer that shooting on film is still a viable choice. the audience does not care much nor notices the difference but for scanning purposes it may be useful to know how much one needs to oversample to get the most out of the negative and to avoid aliasing problems

Link to comment
Share on other sites

This is an old and tired notion - that there's a fixed "resolution" for different gauges. It really needs to die, because it's simplistic and doesn't actually contribute anything to the question of what resolution one should scan at. Part of the problem is that charts like the one in the original post are an answer to the wrong question.

 

 

Film doesn't have pixels so you can't say X film gauge is Y pixel esolution. For one thing, "resolution" means different things in different contexts. In the digital world, it's a number that describes the pixel count, and nothing more. In the optical world, it's about how much detail can be resolved on the film. This is something that's affected by everything from the quality of the lenses being used to the type of film stock, to the aperture and film speed settings, to the lighting, to how the film was processed. There are so many variables, that trying to say in a blanket way that a given gauge is some specific digital resolution is bordering on nonsensical.

 

Nonsensical, is you telling us what we already know and was covered in the original post. Best possible scenario. Lens resolutions etc, no one said "fixed" resolutions. I specifically stated this was a category it fell into. If you don't turn the 4k camera on it has 0 resolution. If Alfalfa shoves a fist full of bubble gum in the 35mm camera it will run at 0 frames a second etc etc. Not worst case scenarios, not common scenarios, Best case.

 

Any and all discussions on resolution are production decisions based on money...always.

 

Our eyes don't have lines of resolution either, so why do we talk about lines, pixels or grain? We have too.

 

A 4k camera shooting 4k and downloading at 1080p looks better than the same camera shooting at 1080p to start with. I don't even care why I just need to know this. If I shot 35mm with a low quality zoom lens on 500ASA I wouldn't expect it to be 6k.

 

Scanners, Monitors, TV's, Projectors and Digital Cameras measure in "K's". Film doesn't and must be matched to a corresponding equivalent. No one here is going to shoot 35mm and project it in their house -- and be a part of this discussion at least.

You aren't going to stand in the middle of a screening room and look at your dailies and say "Ah, yesterday didn't go very well, just scan this at 1.5K"

 

You need to plan ahead and do the math for hard drives and everything needed. Kodak is releasing a better quality Super 8 camera soon so even those numbers play into this. We didn't shoot on 35mm in the past because what we saw at the theater was sub 1080p if you were lucky, with the dirty scratched up lenses, projector lamps that were too old and running at half power to save money. We did it for the future.

 

Because people like you try and shut down resolution discussions so often, is why it took me so long to put together and provide this information. If you look at the great chart provided by Tyler Purcell and copy and paste it online you won't find it anywhere. Perhaps a million people have it already, maybe none, but if it's not searchable, it's not public. I even opened with the chart so no one would be forced to read anything to get to it when needed.

 

If you are happy with your DVD's on your SD TV that is great, just shoot on Beta-SP but try handing it to a neighbor to watch.

 

I'm not fascinated with the numbers, it all takes away from the art of it all. Basically you need a little digital overkill but not too much ($$$). Without the numbers, you don't have a budget and without a budget you don't have a film.

Link to comment
Share on other sites

  • Premium Member

Not sure why you'd want to trust Tyler's numbers, not when he says that 22mm X 16mm = 2400 X 2400 pixels... 22 X 16 is a rectangle and 2400 X 2400 is a perfect square so that doesn't make sense. And basically he's saying that 35mm is 2.4K, which is a somewhat low figure for horizontal resolution though certainly within the realm of some real-world photography. On the other hand, it does show you why 4K would then be high enough as a scanning resolution.

Link to comment
Share on other sites

  • Premium Member

Thanks Tyler, that was interesting to read. I just need to figure out how to convert "lines per picture height" into something more meaningful to me. Do they mean number of horizontal lines for the height of a 4-perf 35mm negative? For example, if 2100 lines per picture height is the peak they measured for the negative, and the negative is 4x3, then that works out to be 2700 vertical lines across if they mean 2100 horizontal lines up & down.

Link to comment
Share on other sites

  • Premium Member

Yea, I know it's a bit confusing. I just looked at the MTF chart, where the camera negative held all the way down to 2400 in both vertical and horizontal. Then I realized that data isn't the right stuff to look at. I was able to deduce from the numbers they used, that 109 lines per mm seemed a close approximation.

Link to comment
Share on other sites

  • Premium Member

What was interesting is that they confirmed that the resolution drops in half more or less when you make a print from an IP/IN, compared to the original negative resolution, which is why 2K projection doesn't seem softer than 35mm release print projection -- the trouble there isn't so much the resolution, is that if you sit closer to the screen, with a 35mm print you see the grain structure but with 2K projection, you see the pixel structure, which is much less organic, more distracting.

 

This is the main reason I'd like to see 4K replace 2K projection in theaters, not so much because the resolution will be higher but that the inherent grid pattern structure of the chip used in the projector will be harder to see, smoother to the eye. Trouble is that, other than 4K laser projection or double 2K IMAX projection, most 4K projection technology has a greater problem with light-scattering/pollution compared to 2K, due to the greater number of adjacent pixels, causing a loss of contrast.

Link to comment
Share on other sites

Yes, thanks Tyler for posting that.

 

Did anyone spot what units were used for screen height?

 

Before the digital pixelmania, "lines" would make sense as line pairs, line pirs per mm, and it wouldn't matter much how they were aligned on the screen.

 

I like lp/mm (line pairs per mm) as an idea, something you can see projected from a reticule, but as they get finer, the contrast does get lower, they're harder to see, just like with an MTF relationship.

Link to comment
Share on other sites

  • Premium Member

Hmm, yes confusing when they use "lines per picture height" to measure the horizontal resolution. I'm used to line pairs/mm, (or sometimes cycles/mm) which is the usual MTF standard.

 

But assuming there are two lines in a line pair or cycle (which they confirm in that paper), 2400 lines equates to about 55 lp/mm for a 22mm wide film frame. And the MTF response is 5% at that frequency, which is basically the cut-off for perception.

 

That's about what I would have thought, maybe a bit higher with modern stocks and lenses. (Although I also agree with Perry that there are so many variables as to almost make this an exercise in counting angels on the heads of pins). You're always talking about system resolutions, depending on the stock the film by itself might be capable of recording over 100 lp/mm, but in the real world it always needs a camera and lenses to expose an image, processing to develop it, more machinery and optics to view it, all of which affects the final resolution.

 

When I test cinema lenses on a test projector, I inspect the 200 lp/mm block in the centre and the 50 lp/mm blocks towards the edges, both saggital and meridional. That tends to be the benchmark for film lenses, if I can make out those frequencies clearly, the lens is good enough.

 

I don't think you can equate lines with pixels to generate a Megapixel count like Tyler has done. It takes more than one pixel to reproduce a line width unless the pixel happens to line up exactly with the line. If the line is half way offset, the pixel records a uniform grey. If we assume that it takes at least 2 pixels to reproduce a line, 2400 lines is nearly 5K. Which is roughly the optimum scanning resolution for 35mm, but not necessarily a good comparison to digital camera sensor resolution.

 

Link to comment
Share on other sites

  • Premium Member

In regards to lines per picture height, it's a document based on projection resolution more then anything else. So they're focused on how many lines the naked eye can see on screen.

 

In my eyes, that's the most important thing because if you can't differentiate between the lines, then the resolution just isn't there. I also assume they did all the testing with a moving system. Meaning, things like registration, shutter and lens would play into the results more then if they simply scanned a still frame with a 12k scanner or something.

 

I think it would be easy to re-create part of the test.

Link to comment
Share on other sites

Because people like you try and shut down resolution discussions so often, is why it took me so long to put together and provide this information.

 

 

The numbers you put together are basically the numbers that have been bandied about on the internet going on a couple decades now. In some cases they made sense when the cost of a 4k scan was calculated in the dollars per frame range. It made sense when the scanners available had fixed sensors (so a smaller gauge had to use only part of the sensor anyway. Someone had to figure out an "optimal" resolution, which really turned out to be more of a minimum than anything.

 

The notion that 8mm = 480 is completely ridiculous. Have you ever seen a 2k or 4k scan of 8mm? There's so much more on the film than you might think.

Edited by Perry Paolantonio
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...