Jump to content

Using color correction filters to help balance old film.


Drew Bienemann

Recommended Posts

Hey,

 

Ive got about 4000' of old fuji that I am planning on shooting a short on. Ive had it for about 6 years in my fridge. its half Eterna 250T 8553 and half Eterna 250D 8546.

 

I just got the snip test back from fotokem and it looks like my layers are between .25 (red and green) and .70 (blue) off from what they should be.

 

Im wondering if, for instance, I could add a yellow filter to my lens when I shoot to make the overall shift of my blue channel get back more in line with the other channels? Does it work that simply, or am I missing something.

 

My thinking goes that if I add 1 stop of yellow to the overall image, then that would put my blue channel at roughly .20 over where it needs to be, and then if I over expose by a stop then that will help the overall quality of the image (though it will still be grainier than normal and a little funky in the colors. Is this reasoning sound?

 

 

Link to comment
Share on other sites

For refrigerated that stock really isn't that old at all. Personally I'd just shoot it normal and the adjust in post, either graded at print time or after scan in post.

The densitometer readings came back pretty far off. The blue channel is more than 3x over the amount the lab considers to be useable.

Link to comment
Share on other sites

The densitometer readings came back pretty far off. The blue channel is more than 3x over the amount the lab considers to be useable.

That's true, but a lab will always report conservatively, so that they are not held responsible for your results. What they consider to be usable is probably based on what would be correctable in a purely photochemical finish. Obviously, with today's technology, there is a far higher degree of control.

Link to comment
Share on other sites

That's true, but a lab will always report conservatively, so that they are not held responsible for your results. What they consider to be usable is probably based on what would be correctable in a purely photochemical finish. Obviously, with today's technology, there is a far higher degree of control.

For sure! I'll be happy with a far more degraded image than roger deakins would. But my question remains.

Edited by Drew Bienemann
Link to comment
Share on other sites

  • Premium Member

I have 15,500 feet of Fuji Super F and Eterna stock in my refrigerator that I'm going to use on a feature soon. I've tested several rolls and the stuff looks great. Fotokem's clip test says it's at the max fog level permitted. Yet, I've shot a few rolls all the way through and outside of the noise floor being higher then new, the colors were spot on. As long as each of the three color channels went up equally percentage wise, you should be fine.

 

The fotokem sheet will show you what they should be and of course, what they are. So all you need to do is calculate the percentage of shift.

 

I'd say shoot a test in a controlled setting. Do exposure tests to see what the latitude is of the stock and where the noise level is. I'd even get Fotokem to make a print for you, so you can see it projected. This way you can absolutely be assured what it will look like. Sometimes the digital scan hides the reality of the camera negatives condition.

Link to comment
Share on other sites

Im wondering if, for instance, I could add a yellow filter to my lens when I shoot to make the overall shift of my blue channel get back more in line with the other channels? Does it work that simply, or am I missing something.

 

Remember that any filtration is going to affect all 3 layers, not just the blue. You may be exchanging one problem for another. Shoot some tests with and without the filtration, and see which is easier to correct back to a 'normal' image.

 

I'd even get Fotokem to make a print for you, so you can see it projected. This way you can absolutely be assured what it will look like. Sometimes the digital scan hides the reality of the camera negatives condition.

That's not going to tell him anything, unless he's planning a entirely photochemical finish, in which case his options for color correction are greatly reduced, and the lab may be correct in saying that the film is unusable.

Link to comment
Share on other sites

  • Premium Member

A color filter will rebalance the exposures of each layer depending on what wavelengths are being filtered, but the problem is that filters only cancel wavelengths, so if your blue layer has become less sensitive and fogged with age, then all you can really do is filter the other two layers, robbing them of exposure, so you can get more relative exposure onto the blue layer once you rate the stock slower to compensate. And any fogging won't be helped with filtering.

 

Truth is that probably for color negative, you're better off just giving every layer more exposure. Certainly that's a simpler solution.

 

I don't think a digital scan would hide information -- a print would though, because of its contrast. If you are planning on a photochemical finish, then obviously a test print would tell you if the gamma of the print combined with your overexposure is enough to hide the aging that would be visible in a scan of the negative. But on the other hand, a photochemical finish would not give you access to digital color-correction tricks that could fix some aging problems like fogging or a color cast in the shadows, etc.

 

My general attitude is that given the high costs of filmmaking in general, taking a chance on the original negative doesn't make a lot of sense unless you are looking for some happy accidents for a dream sequence or something, where you can justify visual unevenness in quality.

Link to comment
Share on other sites

Thanks for all of the replays! This is great.

 

I spoke to a guy named Mark Woods who used to teach all of this at AFI and he said that while my reasoning was sound, he would probably opt to just overexpose and hope for the best, so that's what I'll do. I'll put up some scans when I get them.

 

Thanks again!

Link to comment
Share on other sites

  • Premium Member

That's not going to tell him anything, unless he's planning a entirely photochemical finish, in which case his options for color correction are greatly reduced, and the lab may be correct in saying that the film is unusable.

It tells you exactly what it looks like. If it's acceptable through a photochemical finish, it's fine for digital finish.

Link to comment
Share on other sites

  • Premium Member

For those of you interested, here is the sheet I got back from Fotokem.

Yea, those are WAY off. My guess is the film was stored improperly for quite a while before you got a hold of it, as Eterna isn't that old.

 

The good news is that it looks off by the same percentage per color channel. The bad thing is, over 20% off can't be compensated for through exposure increase. So no matter what, the fog level will be high which means, it will be very noisy even if over-exposed.

 

I was shocked how good my Super F stuff looked, for stock made in the late 90's. Outside of the noise level, which is very high for 64D, it looks damn good. I'm trying to dig up my Fotokem fog test sheet, haven't found it to compare.

Link to comment
Share on other sites

It tells you exactly what it looks like. If it's acceptable through a photochemical finish, it's fine for digital finish.

Tyler, your own statement was that sometimes a digital scan hides defects in the negative. If he is planning to finish via a scan, then having a contact print made is entirely irrelevant. As long as the image is digitally correctable, it makes zero difference what is in the original negative.

Link to comment
Share on other sites

  • Premium Member

Tyler, your own statement was that sometimes a digital scan hides defects in the negative. If he is planning to finish via a scan, then having a contact print made is entirely irrelevant. As long as the image is digitally correctable, it makes zero difference what is in the original negative.

Not saying this is the only way, but when you're doing tests you've gotta keep the variables to a minimal. It's very difficult to eff up a one light print and projection, but a digital scan or telecine, could be totally wrong and until you reference the film itself, you will never know.

 

Also... making a one light print is less expensive then scanning. I do it all the time for my projects to understand what things look like before I scan.

Link to comment
Share on other sites

Not saying this is the only way, but when you're doing tests you've gotta keep the variables to a minimal. It's very difficult to eff up a one light print and projection, but a digital scan or telecine, could be totally wrong and until you reference the film itself, you will never know.

 

Also... making a one light print is less expensive then scanning. I do it all the time for my projects to understand what things look like before I scan.

Why would a scan be totally wrong? Do post production facilities habitually mess their scans up? The point is that if you are intending a digital finish, then a contact print doesn't tell you anything relevant. If you want to know how your neg will look when it's scanned, scan it. If you can correct an acceptable image from that scan, then there is no point making a print "to see what's there", because you'll actually be seeing less not more.

 

A one light print from fogged neg is probably going to look awful, and may make you think the stock is unusable. A properly color-timed scan from the same neg may look perfectly acceptable. You need to test whichever route you are taking. I don't see the point of spending time and money testing routes you are not taking.

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...