[Tig] Film Scan Issues?

Richard Kirk richard at filmlight.ltd.uk
Sat Feb 8 17:31:59 GMT 2014

> From: James Wicks <jim at jimwicks.com>
> Recently, I received a couple of 35mm to 2K .dpx film scans, and both scans rasied similar issues when I loaded them in my color suite.
> I have posted 2 examples (with help from Rob!).
> What I see in my color suite is that the border has the same color wash as the frame.
> It also appears that the border and the frame have the same brightness to them.
> Light leak, perhaps?
> To my way of thinking, a correct film scan should have a black border, or at least not the same color as the frame.
> I suspect that perhaps the scanner was not calibrated correctly?

I am not really the right person to answer this, but I can pose a few of the more obvious points to get the discussion along the right lines...

Are we looking at a scan of a negative or a scan of a print? I suspect you are talking about a scan of a print because a light leak on a negative would affect the highlights on a correctly-viewing image. Any bits of dirt on a negative image would also come out as light spots, and I don't see that here. On the other hand, we are seeing the camera gate, so it would be something like an answer print if it were a positive.

I don't think we are looking at a light leak because the colour cast is very uniform. I think the scanner was set up to give white and black values that are well away from the limits. Scans are typically done to give somewhat 'flat' looking images. You can always grade up the contrast, but you cannot rescue clipped highlight and shadow detail. I don't know whether this particular material has been scanned particularly flat, because I don't know what colour space the images are now in: I am guessing it looks a lot flatter to you then your usual work.

If this is a print, is it a genuinely old print? If it is old, then the film dyes lose their density. Some dyes bleach faster than others, which can give a colour cast. This does not need light, and can happen with the film in the tin. In which case, the scanner operator may be doing the right thing in scanning it flatter than usual: the fading might not be the same for all scenes. 

If we are looking at the scan of a negative, then the black value will be a measurement of the film base. If this is an unusual film stock and there was no clear base to zero the densitometer, then it may have been zeroed on different stuff, which will give you a cast in the shadows. Some negative stocks have a yellow dye that is bleached by blue exposure: the lowest yellow density is then not on unexposed film but on film exposed to blue. It's not much, but ages ago I remember seeing yellow densities a tiny bit below the base in the blue channel. 

Anyway, the images look in good condition: I bet it all comes back when graded up a bit.

Richard Kirk
FilmLight Ltd, Artists House, 14-15 Manette Street, London W1D 4AP
Tel: +44 (0)20 7292 0400  Fax: +44 (0)20 7292 0401

More information about the Tig mailing list