Effects of Differently Rating Stock

[Chris D Walker]

This may read convoluted. What different effects would there be on a negative stock rated and developed in the manners I have written below?

 

For the sake of argument I'll use two stocks to ask my question despite one having been discontinued: 5277 Vision 320T and 5260 Vision2 500T.

 

1) Using 5260 500T, rating at 320 ASA and developing as normal.

2) Using 5260 500T, shooting a gray card rated at 320 ASA and developing.

3) Using 5277 320T, rating at 500 ASA and developing as normal.

4) Using 5277 320T, shooting a gray card rated at 500 ASA and developing.

 

I have rough ideas about each outcome but would appreciate from the knowledge to be found on this forum. Any information about the effect on sharpness, contrast, grain, colour saturation and resolution in any of the four examples would be most gracious.

[David Mullen ASC]

Most labs don't develop to grey cards, they can't see an image until after processing is finished -- so you have to tell them to develop "normal", or "push" or "pull" -- but usually by whole stops.

 

5277 has been long-obsolete, so any stock you find will be outdated, so you can't really talk about how it would look developed normally versus some other method, who knows how it has shifted or fogged by now.

 

So it may be better to phrase this differently.

 

For example, you could compare Vision-3 500T 5219 to low-con Expression 500T 5229, which is the closest thing to 5277 unless you want to consider Fuji Eterna 400T instead.) And work in whole stops because that's what the lab will do.

 

For example, you could ask:

 

5219 rated at 500 ASA, developed normal... versus rated at 250 ASA, pulled one-stop, etc.

 

The basic thing is that overexposure tightens grain structure, reducing the appearance of graininess... and pull-processing lowers contrast.

 

Underexposure increases graininess and push-process increases contrast.

 

Although the amount of pull or pushing doesn't have to match your rating. You could, for example, rate a 500T stock at 320 ASA and pull-process one-stop -- since you rated the stock at 320 ASA instead of 250 ASA, you'd have a net gain of a 1/3-stop in extra density, which is always safer. Same goes for pushing, rate a 500 ASA stock at 800 ASA when asking for a one-stop push and give yourself a safety margin for density.

[Chris D Walker]

Thanks for your response. I suppose using an outdated stock such as 5277 against 5260 isn't the best comparison. Having read in other forums how underexposure heightens grain I was thrown off by what I read in the American Cinematographer Manual, 9th Edition Vol I, under 'The Future for Traveling-Matte Composite Photography' by Jonathan Erland. Here is the extract:

 

"...re-rate the film stock to half its normal rated speed, thus overexposing it by one stop, and then compensate for this overexposure by instructing the lab to pull-process one one stop, thus reducing the development. This maneuver results in a negative with a normal density range but with noticeable reduction in graininess and improved resolution. (Push-processing on the other hand, increases grain and contrast while lowering resolution, and should be absolutely avoided in composite photography.)"

 

I'm reading two differing explanations. Although this extract concerns composite photography, the processing of film stocks used in visual effects is the same as when shooting sync sound. Have I missed something in my reading?

Edited January 17, 2009 by Chris D Walker

[David Mullen ASC]

Thanks for your response. I suppose using an outdated stock such as 5277 against 5260 isn't the best comparison. Having read in other forums how underexposure heightens grain I was thrown off by what I read in the American Cinematographer Manual, 9th Edition Vol I, under 'The Future for Traveling-Matte Composite Photography' by Jonathan Erland. Here is the extract:

 

"...re-rate the film stock to half its normal rated speed, thus overexposing it by one stop, and then compensate for this overexposure by instructing the lab to pull-process one one stop, thus reducing the development. This maneuver results in a negative with a normal density range but with noticeable reduction in graininess and improved resolution. (Push-processing on the other hand, increases grain and contrast while lowering resolution, and should be absolutely avoided in composite photography.)"

 

I'm reading two differing explanations. Although this extract concerns composite photography, the processing of film stocks used in visual effects is the same as when shooting sync sound. Have I missed something in my reading?

 

I don't see the contradiction. Overexposing reduces the appearance of graininess (it actually doesn't technically make the big grains any smaller, but it fills in the gaps between the big grains with smaller grains -- only slower speed stock has less large grains) and pull-processing reduces density, thus compensating for the overexposure. What they don't mention is that pull-processing also lowers contrast.

 

But you could just process the overexposed negative normally and print down the extra density that would result -- you'd get a slightly different look, more snap, deeper blacks, but less extreme highlight detail. Pull-processing on the other hand would lower contrast.

 

What exactly do you think you are missing? I don't see a different explanation so I don't understand.

 

Overexposure -- decreases grain, increases density

Pull-processing -- decreases density, lowers contrast

Underexposure -- increases grain, decreases density

Push-processing -- increases density, increases contrast

 

As for resolution, well, that's technically true but it's a subtle difference -- underexposure means that more detail exists only on the larger grains, the rest being washed away in development due to lack of exposure, so there is less detail, whereas overexposure means that more detail is captured on both the large and small grains (larger grains are more sensitive to light because they have more surface area, smaller grains are less sensitive to light). What's more confusing is what happens to "sharpness", which is different than detail or resolution -- because generally higher contrast images appear sharper to the eye. So while underexposure may reduce resolution, push-processing may increase contrast and thus give the impression of more sharpness even though amount of detail has been reduced.

 

Conversely, an overexposed image that is pull-processed may have more detail but it will also have lower contrast, which may make the image look less sharp.

[David Auner aac]

I'm reading two differing explanations. Although this extract concerns composite photography, the processing of film stocks used in visual effects is the same as when shooting sync sound. Have I missed something in my reading?

 

Hi Chris,

 

which two differing explanations? Pushing and underexposing both increase grain while pulling or overexposing both reduce grain.

 

Regards, David

[Chris D Walker]

I think I've got it now. If you were to use rate 500 ASA speed film as 250 and pull process by one stop, you would get a normal density exposure; you would be pull-processing an overexposed negative. It would look less grainy and that subtle difference in resolution would be there also. Concerning low contrast, wouldn't that issue be reduced by duplicating to an interpositive stock and so on?

 

My confusion lay in mixing scene underexposure with pull-processing an overexposed negative when in development.

 

Thanks for clearing that up. My mind was going in circles for a moment.

Edited January 17, 2009 by Chris D Walker

[Chris Keth]

I think I've got it now. If you were to use rate 500 ASA speed film as 250 and pull process by one stop, you would get a normal density exposure; you would be pull-processing an overexposed negative. It would look less grainy and that subtle difference in resolution would be there also. Concerning low contrast, wouldn't that issue be reduced by duplicating to an interpositive stock and so on?

 

My confusion lay in mixing scene underexposure with pull-processing an overexposed negative when in development.

 

Thanks for clearing that up. My mind was going in circles for a moment.

 

It takes a round of thinking to get it straight. Sometimes it's a little bit like a mental math problem that goes "1+1-1+1-1-1+1-1-1-1+1+1-1=" :lol:

 

You are correct. Overexposing a stop and pulling a stop would result in normal density and a little bit low contrast. The printing process does add a tiny bit of contrast.

[James Steven Beverly]

So, how much can you "print down"? the reason I'm asking is on Blood Moon, we'll be shooting older Lomo anamorphics with an effective stop of 3.3 but to get more DOF we may need to open them up to a stop of 4.6 to 5 which is where anamorphic lenses work best, now much of the film takes place at night with a fair amount of red light (which is problematic for focus anyway) and we want a saturated, contrasty look so if we rate the film (5279 or 5260) at 350 ASA and process normally, then print down, will that give us a believable night "look" or do we Pull-process and print to kill some of the highlights and drop a stop or 2 then print down and increase contrast? OR do we open it up the 4.6, under expose and push process and if so will there be enough detail n the shots and what role could printing play? OR do we selectively light areas for a 4.6 f-stop and leave the surrounding areas with less light and process normally (which because we are using negative space in our composition, might not matter if it's a fair bit darker)? I'm not so much asking anyone to direct the film for me but I wold like to know what the different combinations might look like under these conditions. This whole f-stop thing on these lomo anamorphics and using them at night has been bothering me for a while. I know Carpenter used anamorphics from the 70s on Holloween and they worked well so I know it can be done as these are newer lenses than the ones he used (though his were western lenses and HOPEFULLY that won't make too much a difference) B)

Edited January 18, 2009 by James Steven Beverly

[Satsuki Murashige]

Steve, I think with the stops you want to work at, you should consider pushing the film stock by one stop. Just to give you an idea, lighting to a f/5.6 at 400ASA would require 100 footcandles to get a normal exposure. It's a lot, but doable if you have large lighting package or don't mind a hard lit look. Shooting at f/4 and 800ASA would let you use only 25fc. I suspect you'll probably end up working close to wide open with those lenses. Plus pushing the film should increase contrast and saturation versus just underrating the stock by one stop and processing normally. Pulling the film is a bad idea if you are already struggling to get enough light since you'll have to pump twice as much light into the scene to get the same density on the neg after processing.

 

But you should shoot a test with these options on the stocks you want to use to get a real idea of what it's going to look like. Carpenter probably used Panavision C-Series lenses on "Halloween." Those generally open up to around T2.8, some are faster.

[James Steven Beverly]

Yes, shooting f 4 and f5.6, I had a brain sitting around here somewhere anyone see it? :rolleyes: Thank for the advice Satsuki. David Mullen had said pretty much the same thing on a thread a few months ago and though I was not second guessing him in any way, it is nice to hear you say it as well, it really sets my mind at ease knowing it seems to be the consensus of people in the know. We've got 4 4-36 Dinos and the gennies to run them to work with and a lot of other lights so we SHOULD be OK as far as lighting goes even at 5.6. I guess I was just worried there would be too much light poured onto the desert to make it look like night even at night OR that we'd have to loose most of our DOF which would limit composition. I guess these were unfounded. I will, as both you and David suggested, of course do some tests. I may even be able to do some while we're shooting the trailer for Blood Moon. Thanks again for the info on Halloween, I was wondering! The Lomos I believe open up to 2.5 but the evective is 3.3 as I mentioned before. B)

Edited January 19, 2009 by James Steven Beverly