Resolution

[focusgroup]

Rumor has is that super 8 has horizontal resolution of greater than 1300 lines. If this is the case why arent optical blow ups to 35 mm ever done for super 8? My understanding is that super 8 is always digitally captured and transferred for blow up. I have also heard that mini dv and super 8 are similar in resolution limits. This doesnt make sense if DV NTSC is 525. I would appreciate any comments.

[mattias]

Rumor has is that super 8 has horizontal resolution of greater than 1300 lines.

that's where the emulsion sets the limit, yes. in reality you should get at least half of that but probably not more than 1000 or so. i tested k40 in my canon 814e once (search the google archives) and found that i got 900x700 or so if i remember correctly.

If this is the case why arent optical blow ups to 35 mm ever done for super 8?

but this is done all the time. that's how they cut super-8 into the 35mm features that use it.

I have also heard that mini dv and super 8 are similar in resolution limits.

mini dv uses edge enhancement, which makes it look sharper than it really is.

/matt

[tim]

This is one of those topics that goes round and round!

Resolution of fil depends on both the film itself, and the lens aperture. For a high contrast image and a wide lens aperture, K40 will resolve 100 line pairs per mm (Kodak's figures).

TV resolution depends on bandwidth and display device. Luminance bandwidth allows 600 or so line pairs across a screen, but chrominance bandwidth is lower so colour is less well defined. The colour stripes of a 26", 4x3 TV tube are in 0.8 mm wide groups of three. Smaller TV tubes do not have smaller stripes. Work all this out, and K40 super 8, transferred electronically with no loss of resolution, needs a VGA screen (colour stripe group 0.4 mm wide) to display it fully.

However, as has been pointed out, edge enhancement gives TV and artificial impression of definition.

Personally, I don't transfer to any electronic form. I go to a lot of trouble to get that definition and colour on S8, and I don't want to lose it.

Tim.

[jessh]

Rumor has is that super 8 has horizontal resolution of greater than 1300 lines. If this is the case why arent optical blow ups to 35 mm ever done for super 8? My understanding is that super 8 is always digitally captured and transferred for blow up.

as far as I know both optical and digital blowups of super8->35mm have been done. Digital seems to be common for 2 reasons, 1)it is more economical, getting an optical blowup just right may require doing it a couple of times, while results may be a little more predictable with a digital intermediate, also when shooting negative film doing an optical blowup to a negative requires use of an interpositive, which costs more money than using a digital intermediate. 2)most people who use super8 in a 35mm feature or with plans of blowing up to 35mm use super8 to achieve a "unique" look, by using a digital intermediate they are given much more control over the resulting look without the added cost of doing the blowup many time (plus things you couldnt do otherwise). Also there are places that do super8->35mm blowup using a digital intermediate and im not sure of any places that specifically offer an optical blowup (although I am sure many places could do it if you have the budget)

~Jess

[Brandt]

I believe that Monaco/Interformat in San Francisco still does S8->35mm optical transfers.

Digital Film Group in Vacouver, CA does a digital uprez for DV intended for 35mm release. I'm sure that anything captured using a Rank or Workprinter could be uprezzed for transfer by the same process.

Mattias, what was the test you used to determine that the film coming off your 814 had a specific resolution? Did you take a small section of the film and perform a crystal count? I'm not familiar with the process of quantifying film resolution.

I am finding that I can zoom in about 200% on a super 8 frame with my workprinter setup, which uses a JVC DV500U camera, and not see any appreciable degredation in image quality.

-Brandt

[mattias]

Mattias, what was the test you used to determine that the film coming off your 814 had a specific resolution?

i shot a resolution test chart. it's basically a bunch of converging lines and you measure the resolution by looking at where they're no longer lines but just a blur. search the archives here and at google groups for the complete report...

/matt

[Andreas Wideroe]

Click here to go to groups.google.com

[jessh]

i shot a resolution test chart. it's basically a bunch of converging lines and you measure the resolution by looking at where they're no longer lines but just a blur.

do you have any plans to do the same with other film stocks? I would love to see the results of Vision 200T, 50D and the B+W stocks, ektachrome would be interesting as well of course

~Jess

[mattias]

do you have any plans to do the same with other film stocks? I would love to see the results of Vision 200T, 50D and the B+W stocks, ektachrome would be interesting as well of course

i don't have to, since they all have almost exactly the same resolution built into the emulsion. kodak has the numbers on their site, and you can see that it only varies between 100 and 110 line pairs per mm or so. however, more important than resolution when comparing sharpness of a stock is contrast, grain structure and so on, which cannot be measured this way. perhaps the most common and effective way of testing that is to shoot a wide front lit daylight exterior with lots of different textures and contrasts. that will tell you how sharp the stocks are compared to each other, even though the test doesn't give you a "real" result such as a number...

/matt

[tim]

The link didn't work for me, so I couldn't find the answer to this question:

'What aperture did you use in the tests?' (At about f/11 lens resolution falls to about 100 line pairs/mm anyway.) Only tests at f/1.8 or better, and high contrast give true comparative results.

[jessh]

'What aperture did you use in the tests?' (At about f/11 lens resolution falls to about 100 line pairs/mm anyway.) Only tests at f/1.8 or better, and high contrast give true comparative results.

hmm, this contradicts with other things I have read, a ot of stuff I have seen says that the lens will actually be sharper(and have higher resolution) when stopped down that when it is wide open. I was looking at some tests earlier where someone shot reolution charts with a video camera and the camera indeed appeared to have a higher resolution when the lens was stopped down (higher fstop number)

~Jess

[Guest]

Jessh wrote:
also when shooting negative film doing an optical blowup to a negative requires use of an interpositive, which costs more money than using a digital intermediate.
Why can't they just transfer it directly to positive stock ( like a negative except that it has the complimentary masking colors)?

[jessh]

also when shooting negative film doing an optical blowup to a negative requires use of an interpositive, which costs more money than using a digital intermediate.

Why can't they just transfer it directly to positive stock ( like a negative except that it has the complimentary masking colors)?

well you could use your super8 negative and just make prints from it, but for something that will require many prints that is not a very good idea for 3 reasons: 1) you degrade the quality of the origional every time you run it through an optical printer 2) optical blowups are more expensive than contact prints 3) if you want to do any type of color or exposure correction you then have to do it for every print

the standard way of making prints when shooting negative stock is this:
-Make an interpositive from your origional negative(which when doing super8->35mm would be an optical blowup)
-make a protection negative from the internegative
-make release prints from this new negative using a contact printer (faster and cheaper than optical printing)

this method requires minimal wear on your origional, allows you to color correct and time the movie (during one of the first to steps), and allows you to make many prints for relatively cheap. by using a digital intermediate you don't have to pay for an interpositive, in a big budget production the blownup 35mm negative would most likely be treated like an origional (especially if being intercut with origional 35mm footage) and the above mentioned steps would be used to make release prints.

some people say that using a digital intermediate results in less grain because you have one less film generation, but you are adding a digital generation, but I really cant see a 35mm print stock (very fine grain 35mm) having more of a negative effect than using the digital intermediate, I havent compared results side by side, but I would guess that a 35mm printstock intermediate would beat even a 4k digital one in terms of quality.

hope that was of some help, if I am wrong about anything please feel free to correct me, and if you have any questions I will gladly answer them as best I can.

~Jess

[Roget]

No specific end-all-be-all answers here, but lenses are generally their sharpest in the middle of their aperature range. Commonly around f8 or f11. This is the general rule, though it varies by lens. All of the hard facts and figures I've seen backing this up were related to large format (4x5) lenses, not Super-8 cameras.

I don't know what the exact resolution of Super-8 is, but the American Cinematographer Manual, 7th edition has some interesting information about how to compare the resolution of 35mm film to video. The NTSC standard has a bandwidth of about 4.5 mHz; PAL about 5.5. 35mm film's bandwidth translates to approx. 35mHz, or roughly 7 times that of video.

Off the top of my head, I don't know the exact dimensions of a Super-8 frame vs 35mm, but if you know the difference in surface area, I suspect you could get a pretty good estimate of Super-8's approx. bandwidth and compare that to the video above.

Any time you make a print (optical or digital) this resolution is reduced somewhat, though I don't have any figures detailing how much. In the end, I think a lot of ends up being purely subjective. For example, "Oh Brother, Where Art Thou" was entirely scanned into a digital 2k system for extensive color correction. The film's DP noted that the resolution was softened, but the effect was well worth it.

Roget

[mattias]

The link didn't work for me

well, have you tried performing the search yourself? you know my name and the subject of the post? can't be that hard. you can try mike's board on hostboard as well.

What aperture did you use in the tests?

i don't remember, but it was somewhere in the middle, where the lens is the sharpest, and certainly in the 4-11 range.

high contrast give true comparative results.

yes, a test chart typically has black lines on white. or what do you mean?

/matt