An idea for cutting down levels of grain

[Guest Daniel J. Ashley-Smith]

I have an idea, I'm not sure if it would work or not but here goes:

 

Use positive film to shoot the footage with, but then use a pull process so the contrast levels are about the same as a negative but the grain has been cut right down because of the pull process.

 

Been done before? Doesn't work?

 

Thanks for any suggestions.

[Kevin Zanit]

Well, as a rule using any push or pull process on a reversal stock creates fairly un-predictable results. Reversal is far less forgiving than negative, thus any tinkering with the chemistry tends to cause blacks to either go real milky or get real blocked up. Also an increase in grain will result from this work.

 

The other issue is that reversal stock is inherently grainier than negative, thus not really helping your cause.

 

Also, reversal has far less latitude than negative, so it is gives one much less room to work.

 

And finally, if you want additional prints, you actually will need an additional lab step to end up with a negative. Or you could print onto a reversal print stock from the camera original, but this will result in even more grain.

 

Pulling really doesn?t lower the amount of apparent grain by much. If I pulled a 500 speed negative film by one stop, effectively shooting at 250, I might as well just shoot on 250ASA film without the pull. The results would be finer grain, and simpler lab work. This is assuming you don?t want the effect that a pull has on color.

 

 

Kevin Zanit

[Guest Daniel J. Ashley-Smith]

Oh... Two words: That Sucks.

 

Tnx Kevin. What do people mean when they say a film has more or less latitude?

[Filip Plesha]

Think of it this way:

 

Negative film has more range than you actually use in prints. No matter

how you set your printer, you always have "leftovers" on either (or both) sides

of your printing range. You have a black and a white point in your print (or an electronic image scanned from film) but your pure black is not the darkest

part of the image from the negative, and your pure white is not the brightest

part of the image from the negative. On both ends of your image you have

additional information (in the negative) that you could have used by migrating your

black-to-white range upper or lower on the full scale of the densities (tones) on the negative.

 

For example, when you scan the film into cineon 10-bit format you have

1024 levels of light, but by default you only use 590 levels for viewing on the monitor. It is your choice which 590 levels you will chose depending on wheather you want a brighter or a darker image.

Or you can choose more or less than 590 levels if you want more or less contrast

for your image on the monitor.

 

Same goes for analog printing. Let's say that you have again 1024 levels of brigthness in the negative (it's only for the sake of picturing it more clearly, film is continuous rather than divided into levels of light)

The choice of print stock will determine how many of these 1024 will you use

for your projected image. A more contrasty print stock will perhapse extract 500 or 450 levels, and a lower contrast print stock will perhapse extract 650 levels etc.

And by adjusting your printer light you "migrate" your "window" of let's say 550 levels) up or down the scale of full 1024 levels so you can get a brigter or a darker print.

 

I used numbers only to make it clearer, they could be any numbers because

film is analog and it's range can be devided into as many numbers as you wish.

 

Now latitude is the USEFULL range in which you can move your printing range up or down the scale of the full range and get a high quality image.

 

If you try and move your printing range of tones up the full scale by two stops

on a low latitude film stock you will probably get a grainy image.

And if you do the same on a high latitude film stock you will get less grain, less

color shift in shadows, and a higher quality image overall.

 

I hope you get the picture.

 

Oh one more thing...

 

If you are wondering how to picture higher and lower dynamic range in your head,

imagine a lower and higher contrast image.

But by low contrast I don't mean milky blacks and gray whites as if when you turn down the contrast on your TV.

The white and black and still white and black, it's just that a higher dynamic range

image has pure white assigned to stronger light levels (sun reflections etc.) rather

than what we percieve as white with our eyes.

so it looks like it's a milky low contrast image, but it is rather different that when

you lower the contrast of an image by force and reduce the brightnes of its darkest and brightest pixels.

 

And therefore your theory of reducing reversal to the contrast of negative is wrong because you can not retrieve something that is beyond the white point (Dmin) of reversal film becsause it is not recorded. Reversal film is "what you see (on film)is what you get (on screen)" while negative is "what you see (on screen) is only a part of what you see (on the negative original film)"

[Christian Appelt]

Daniel, did you actually mean REVERSAL film? I'm asking because POSITIVE film is quite another thing, it's the stock on which your negative is printed.

 

Positive film has indeed very fine grain, but it is so slow that you will need more light than a 3-strip Technicolor production back in the 1930s!

 

A friend of mine did some shooting on black&white positive film, but he processed it in a special developer calles NEOFIN DOKU which is normally used to get grayscale pictures from high contrast films like Kodak Technical Pan. I think Kodak TECHNIDOL LC does the same.

 

I always wished I could shoot 35mm on Technical Pan film, with proper developers it can be rated at about 50-80 ASA and gives a grainless look.

 

But all of this is strictly black & white, and if you want an almost grainless color image, use low speed 50ASA stock and talk to your lab, maybe it can be further pull processed when you overexpose it a bit. But the grain is already so fine that overexposing/pull processing may not do anything at all.

 

Of course, there's always 65/70mm... ;)

[Kevin Zanit]

Simply put:

 

Latitude is the range from light to dark (highlights and shadows) a film can record.

 

Reversal film has less range than negative.

 

 

Kevin Zanit

[Filip Plesha]

Simply put:

 

Latitude is the range from light to dark (highlights and shadows) a film can record.

 

Reversal film has less range than negative.

Kevin Zanit

<{POST_SNAPBACK}>

 

 

Actually that's dynamic range, latitude is how many stops you can print up

or down and still get detail in your image.

[Kevin Zanit]

I am going to disagree with you:

 

Latitude, as defined by Kodak (http://www.kodak.com/US/en/motion/students/support/glossary/glossary1.jhtml?id=0.1.4.13.10.4&lc=en#L)

 

Is: ?In a photographic process, the range of exposure over which substantially correct reproduction is obtained . . ."

 

Dynamic range is just another way to refer to latitude. Maybe the meaning of these terms differ in still photography, but when referring to motion picture films, latitude and dynamic range are essentially one in the same.

 

 

Kevin Zanit

[Filip Plesha]

I am going to disagree with you:

 

Latitude, as defined by Kodak (http://www.kodak.com/US/en/motion/students/support/glossary/glossary1.jhtml?id=0.1.4.13.10.4&lc=en#L)

 

Is: ?In a photographic process, the range of exposure over which substantially correct reproduction is obtained . . ."

 

Dynamic range is just another way to refer to latitude.  Maybe the meaning of these terms differ in still photography, but when referring to motion picture films, latitude and dynamic range are essentially one in the same.

Kevin Zanit

<{POST_SNAPBACK}>

 

Well, I agree with that definition, but I dissagree with your interpretation of it.

 

Let's get back to that fameous S curve. The middle straight line is the portion

in wich you can get normal grain and normal color balance in your image, right?

The lower curved part and the higher curved part are not recommended places where you can "put" your image, right? Why? Because you get anomalies in grain structure or color balance or contrast.

Now, for me, latitude is the range on wich you can record the image

without these anomalies, and with "correct reproduction" as Kodak put it.

 

But the dynamic range is the range from Dmax to Dmin, including these both

curved portions of the S curve.

 

Dynamic range and latitude do come together, but for me latitude is a lower

term than dynamic range, or in other words, dynamic range "contains" latitude.

 

So if you have your dynamic range from somewhere between 10 and 20 stops (as Kodak says) on the negative, your latitude is usually about 5 stops, two stops under and three stops over the 18% gray for example.

You cant overexpose something by 5-6 stops, print it down and get a decent image, but you still have density variations 5-6 stops over 18% gray if

you are photographing something like a sun reflecting on the surface of watter.

 

This is my basic understanding of latitude vs. dynamic range, and untill now

I had the idea that it is allso how these terms are used in this community, I guess I was wrong. But I did see a lot of different uses for the word latitude, so I guess it's a matter of choice.

 

but as for the official definition, I believe my explanation is fully compatible with Kodak's definition.

[Kevin Zanit]

I think the problem is you are over complicating the concept.

 

The characteristic curve is simply representing how much information a film can record.

 

When the curve reaches D-Min and D-Max, you simply have run out of information on the film. Where the D-Min and D-Max fall is how much latitude you have for a film. This is the same as dynamic range.

 

The straight-line portion of the curve is just where the "optimal" place is for exposure. It does not mean that the film will not record detail beyond this point. As a rule you find more visible grain in the strait line portion because it tends to fall more in the "midtones" where grain tends to be more visible.

 

I think we are actually in agreement as to the meanings of these words, but it is just how we are phrasing it that we differ.

 

Dynamic range (or latitude) in printing is a completely separate issue from a negative's latitude. You should only look at the negative for the dynamic range (or latitude) of a negative. If there is information on the negative, it fell within the dynamic range (or latitude) of the film.

 

Now, when I telecine footage, the video format usually can not record all of the information that the negative contains from white to black. Thus, I throw away some information that was on the negative. Does this mean the negative has lower latitude? No, it just means the film's latitude exceeded that of the printing steps.

 

Dynamic range is a term I hear more in the video world, and latitude tends to reside in the film world. These terms mean the same thing though.

 

I am sure John could shed more light on the subject.

 

 

Kevin Zanit

[Filip Plesha]

I think the problem is you are over complicating the concept.

 

The characteristic curve is simply representing how much information a film can record.

 

When the curve reaches D-Min and D-Max, you simply have run out of information on the film.  Where the D-Min and D-Max fall is how much latitude you have for a film.  This is the same as dynamic range.

 

The straight-line portion of the curve is just where the "optimal" place is for exposure.  It does not mean that the film will not record detail beyond this point.  As a rule you find more visible grain in the strait line portion because it tends to fall more in the "midtones" where grain tends to be more visible.

 

I think we are actually in agreement as to the meanings of these words, but it is just how we are phrasing it that we differ.

 

Dynamic range (or latitude) in printing is a completely separate issue from a negative's latitude.  You should only look at the negative for the dynamic range (or latitude) of a negative.  If there is information on the negative, it fell within the dynamic range (or latitude) of the film.

 

Now, when I telecine footage, the video format usually can not record all of the information that the negative contains from white to black.  Thus, I throw away some information that was on the negative.  Does this mean the negative has lower latitude?  No, it just means the film's latitude exceeded that of the printing steps.

 

Dynamic range is a term I hear more in the video world, and latitude tends to reside in the film world.  These terms mean the same thing though.

 

I am sure John could shed more light on the subject.

Kevin Zanit

<{POST_SNAPBACK}>

 

 

Here is another example:

I don't know how it is in cinematography, but it is often said that reversal film

has "zero latitude", same goes for prints.

This is basicly my understanding of latitude. Reversal film has no headroom beyond the white and black that you use in projection or scanning.

But it does have it's dynamic range.

 

Reversal film is used from Dmin to Dmax, and negative is used by extracting a portion of the dynamic range, so here is an example of what I mean:

 

You shoot day for night, the subject is the surface of the sea, it is lit by direct sunlight, and there are bright reflections on the surface. It idea is to get

a night shot of a sea lit by moonlight.

If you used reversal, you would not be able to print it down in reversal-reversal printing because you would loose highlights in the reflections.

If you print it down, the white reflections will turn to gray.

 

but with negative, you would print it down, your midtones would turn to shadows,

your highlights would turn to midtones, but you would still have the extra

information that would come out as highlights in a print, and you would have

a scene or an image looking like night, but with a natural distribution of tones, from

Dmin to Dmax.

In case of reversal copy, you would get a distribution from Dmax (black) to

some gray shade which was once your white (Dmin)

 

This is is what I call latitude, reversal has no latitude at all, and negative has

some. So in other words, latitude is like a synonime for headroom of the original.

 

Or to put it another way:

Latitude is flexibility of the dynamic range, or a ratio of original D range vs. D range extracted for prints/electronic imaging.

 

It is all a game of words I think,

You mention dynamic range of print vs negative, well like I said

for me it's like this:

 

latitude=(full dynamic range of negative)/(portion of that dynamic range used for viewing)

 

Of course not literally, you don't actually express latitude in numbers, it's just symbolic.

 

I am really currious about how other people on this forum define latitude..

 

 

P.S. I think by now we have totally confused Daniel, who asked about latitude in the first place

Edited December 27, 2004 by Filip Plesha

[Wendell_Greene]

My questions after reading the initial post were "What's the look that you're going after? Are you planning on just projecting the reversal, or transferring the print in Telecine? Why not shoot a test?

 

I know of a director [Chris Milk] who shoots almost all his music videos on older ektachrome reversal stocks, his DP works very hard at making sure not only that his exposures are dead on, but also that the print will be as low contrast as possible, since they know they always add contrast [as well as reduce grain] when they do the transfer in telecine.

 

We can often complicate matters by getting bogged down in the mire of overwhelming sensitometric information , debates on interpreting the characteristic curve, semantics, and most often, by applying parameters that are more applicable to still photography than motion picture cinematography. The bottom line is "What do you want it to look like, and what's the best way to accomplish that, in the most simple and direct way possible?

 

Just my dos centamos.

[Michael Nash]

Or to put it another way:

Latitude is flexibility of the dynamic range, or a ratio of original D range vs. D range extracted for prints/electronic imaging.

 

I am really currious about how other people on this forum define latitude..

<{POST_SNAPBACK}>

 

We had this discussion a while back. As I understand the terms, "latitude" and "dynamic range" are NOT the same thing. Latitude is the amount you can over- or under-expose a negative (or reversal) and recover an acceptable image. Dynamic range is the range of brightness the film can capture.

 

This distinction becomes important when you start to talk about "rating" a film at something other than the manufacturer's recommended ASA. For example, a given film may be able to capture detail down to 5 stops under exposed, but I wouldn't DARE tell someone the film has "5 stops of underexposure latitude", and have them think they can underexpose the image 5 stops and still recover an image! The underexposure "latitude" of that film might really be more like ONE stop.

 

But I'm aware that many, many people use the term "latitude" when they're talking about "dynamic range." So if that's the case, is there ANOTHER term that describes the amount of exposure deviation a film can handle? Like "exposure margin" or something? I really thought that's what the term "exposure latitude" was for, after all...

 

I'd really like to hear from John Pyltak and Dominic Case on this one.

[Filip Plesha]

We had this discussion a while back. As I understand the terms, "latitude" and "dynamic range" are NOT the same thing. Latitude is the amount you can over- or under-expose a negative (or reversal) and recover an acceptable image. Dynamic range is the range of brightness the film can capture.

 

This distinction becomes important when you start to talk about "rating" a film at something other than the manufacturer's recommended ASA. For example, a given film may be able to capture detail down to 5 stops under exposed, but I wouldn't DARE tell someone the film has "5 stops of underexposure latitude", and have them think they can underexpose the image 5 stops and still recover an image! The underexposure "latitude" of that film might really be more like ONE stop.

 

 

<{POST_SNAPBACK}>

 

Yes, this is what I was talking about, that's the way I understand latitude.

[John Pytlak RIP]

I'd really like to hear from John Pyltak and Dominic Case on this one.

<{POST_SNAPBACK}>

 

The Kodak website has some really good tutorials on film tone scale, latitude, exposure, etc.:

 

http://www.kodak.com/US/en/motion/support/h1/exposure.shtml

 

http://www.kodak.com/US/en/motion/support/h1/structure.shtml

 

http://www.kodak.com/US/en/motion/support/h2/

 

http://www.kodak.com/US/en/motion/students...0.1.4.9.6&lc=en

[Michael Nash]

From the Kodak website:

 

EXPOSURE LATITUDE: Degree to which film can be underexposed or overexposed and still yield satisfactory results.

 

CHARACTERISTIC CURVE: Shows the relationship between the exposure of a photographic material and the image density produced after processing.

 

I think "characteristic curve" is the closest useful description to "dynamic range," although it also expresses where densities fall within the total range, and not just the upper and lower limits.

 

The term "dynamic range" does not appear anywhere in the Kodak descriptions, and Kevin's probably right that it's a term carried over from the video world. But in any case there are two different concepts here, described by two different terms.

[Filip Plesha]

From the Kodak website:

 

EXPOSURE LATITUDE: Degree to which film can be underexposed or overexposed and still yield satisfactory results.

 

CHARACTERISTIC CURVE: Shows the relationship between the exposure of a photographic material and the image density produced after processing.

 

I think "characteristic curve" is the closest useful description to "dynamic range," although it also expresses where densities fall within the total range, and not just the upper and lower limits.

 

The term "dynamic range" does not appear anywhere in the Kodak descriptions, and Kevin's probably right that it's a term carried over from the video world. But in any case there are two different concepts here, described by two different terms.

<{POST_SNAPBACK}>

 

Dynamic range is a term usually used in audio terminology, in video terminology too. I think the term used for film is more often "contrast range" or more accuratly "density range"

But "dynamic range" is a very general term and can be used for many things, including film talk.

[Laurent Andrieux]

I second the fact that "dynamic range" is used in electronics in general, not in cinematography.

 

I'll try to put this as "easy" as possible...

 

One shouldn't miss the difference beetween film latitude : the ability of a film stock to "record" satisfyingly a range of tones and the ability of such a film to be over or underexposed...

 

As someone explained it fairly well somewhere here, a film can have a latitude of, let's say 6 stops, it doesn't mean you can over/under expose that much, because that depends on the shot subject.

 

If you shoot a subject that has only one or two stops of contrast (middle tons), then you can over or underexpose (let's say 2 stops for instance) and get your contrast and details back at the printing, over or under printing. If the subject has already 6 stops contrast beetween blacks and whites and you shoot with a 6 stops latitude film, then you have no exposure latitude ie possibility of under/over exposing because you will loose detail in the low lights or highlights.

 

If you shoot a subject that has 5 stops of contrast with a film that handles 6 stops, then you have a possible offset of 1 stop around the supposed "perfect" metering (that would render the densities the best, with no need for correction at the printing) that allows you to over/under expose of 1 stop.

 

The word "latitude" can be used for both things (film latitude / exposure latitude) and than can make a mess in someone's mind...

 

Could somebody remind the correct terms for both concepts, in english ?

Edited December 30, 2004 by laurent.a

[K Borowski]

Daniel, have you ever shot a roll of MP film in your life, even 8mm? If you don't know what latitude is, you should consider learning the more traditional methods of shooting rather than coming up with ridiculous new alternative processes, which usually serve only as brief clips in movies anyway, for things such as flashbacks.

[Kevin Zanit]

The question of latitude was not raised by Daniel; it was a discussion that originated in response to my answer in regards to a problem with Daniel?s theory.

 

I would actually encourage this kind of thinking. It shows an interest in really learning the lab side of cinematography. When one can begin to think about how to mix up the post process for a specific effect, it usually means one has a certain understanding of the process. This is good.

 

As for the problems with a specific approach, well that just comes from experience. But Daniel asked a very valid question, and raised a valid theory. And anyone with insight to this theory should respond as to why or why not it might not work as expected.

 

No reason to jump on Daniel for a simple question.

 

 

Kevin Zanit

[Phil Rhodes]

Hi,

 

Laurent - your English explanation is fine, but I think that possibility for confusion is why people are increasingly using the term "dynamic range" to describe that particular sense of "latitude" The term is accurately applied and also makes direct comparison with video cameras easier.

 

Phil

[Filip Plesha]

Daniel, have you ever shot a roll of MP film in your life, even 8mm?  If you don't know what latitude is, you should consider learning the more traditional methods of shooting rather than coming up with ridiculous new alternative processes, which usually serve only as brief clips in movies anyway, for things such as flashbacks.

<{POST_SNAPBACK}>

 

 

Don't be so hard on whoever asked the question, as you can see

we had a long debate here over what actually is latitude, it turned out to be

less obviouse that everybody thinks.

[Laurent Andrieux]

As to come back to Daniel's first post, I was wondering if the fact that reversal is considered like "zeo latitude" is not only due to the fact that, originally, reversal is designed to be projected after processing, and since there is no printing, there is no possibility for correcting the original...

 

I wonder if this approach of "zero latitude" is not linked to this and not to the reversal stock caracteristics. I read somewhere that when you duplicate or print an intermediate from a reversal, both giving the ability of correcting the exposure, then you get a huge latitude with it...

[David Mullen ASC]

I read somewhere that when you duplicate or print an intermediate from a reversal, both giving the ability of correcting the exposure, then you get a huge latitude with it...

<{POST_SNAPBACK}>

 

The gamma of reversal stock is designed for a lot of contrast so when projecting the positive image onto a white screen, blacks look dense enough. Copying a reversal onto a low-contrast film element like an internegative does not increase picture information and therefore does not increase your ability to make color-corrections. You are always limited by what originally was recorded on film.

[K Borowski]

I asked this question in a still photography class once, and I remember there is opportunity for a small ammount of correction, but maybe only a stop of underexposure and a half stop of overexposure. This is with something like 7250. With the 100D reversal film that's out, there is probably even less latitude than that. With Kodachrome 40A, there's almost no latitude at all. You'd be lucky to correct 1/3 of a stop in either direction.

 

Regards.

~Karl Borowski