— ROGERS

One of the implications of moving from film capture to digital capture is that images tend to have a different “depth” or “presence”. This is probably because of the way light scatters through layers of emulsion as opposed to the way it hits a cold, hard, digital sensor.

I’m not referring to sensor crop, or any other lens aspect-based change. Rather, this is more about why digital images look “flatter” or resolve distance/light differently than their analogue equivalents. This article is more about lens design, but has little nuggets of information that apply in a this context:

The emulsion layer that holds the light sensitive silver halide grains has a certain thickness and contains up to twenty layers of grains, any one of which can be struck by photons and therefore is part of the latent image. Light rays that strike the surface of the emulsion layer at an oblique angle will travel through the depth of the gelatine layer and will be stopped by some grains in the lower layers. So the angle of incidence is no problem at all. Ideally the film plane should be plane, but film is never flat at the film gate and will bulge. But the depth of the emulsion layer and the depth of field tolerance will offset this state of affairs and optical designers can use this characteristic to compensate the problem of the curvature of field…

…The main characteristics [of a] sensor are the fact that the sensor is flat (plane), constructed as a discrete matrix of pixels and not transparent (has no depth). The flatness of the sensor is bad for the curved nature of the image created by the lens. The opaque nature of the sensor cells implies that the oblique angle of incidence of the light rays striking the sensor surface must be limited. Otherwise only a few photons will be captured. [link]

So to paraphrase: film scatters the light in a gooey, flawed (but nice), analogue sort of way, whereas a sensor is a bit more “clinical”… yes, wild, over-simplification…

An interesting footnote is the Foveon X3 sensor, which is built more like film – red, green and blue all layered on top of each other (other sensors are usually one layer, arranged like a checkerboard). Sigma markets a series of cameras which use this sensor. Supposedly, it gives a more “film-like” look, as well as less noise and a higher dynamic range than other sensors. I can’t tell, from what I have seen online.

There seems to be a lot of concern about the footage coming from the RED One being soft. Discarding focal issues, this is pretty much the same complaint that used to come from photographers of digital RAW stills. The solution is that there doesn’t need to be solution. It is not that the images are soft, it is more that they haven’t been sharpened (digitally, in the camera). It is better that RAW footage is not sharpened before post-processing. Film has grain, which increases perceived sharpness. As digital images are usually low in grain/noise, they don’t carry this “advantage”.

Filters such as unsharp mask are used as a matter of course by stills photographers. And even though there are modern day digital equivalents, the technique originated as an optical process with the Germans in the 1930′s. Basically what they did was to use a negative combined with a soft (contact printed) positive as a mask, which was then used to increase the contrast of edges and details. The advantage was that it did not change the flatter parts of the image. This works well optically for our eyes, which is something Ernst Mach had a lot to say about.

Scanned film also looks soft, especially when you zoom in to 100%. With the proliferation of DI systems, techniques like Super 2k have become a popular way of getting better looking scans from 35mm film. Many scanning houses now offer this service. They scan at 4k, and then downsize to 2k, using a sharpening algorithm along the way (probably using filters such as sinc which tends to sharpen when scaling down). The images look punchier and the file size is relatively small. In addition to the sharpening, the perceived resolution is higher because all the extra detail of the pixels in the 4k image are blended into the pixels of the 2k image (see: sub-pixels). For transfers to video, manufacturers of telecine chains usually refer to sharpening in some sort of euphemistic jargon. It is all the same, of course.

The only problem is that while the sharpening process looks good, it’s absolutely horrible to work with from a visual effects point of view. Sharp spikes and artefacts appear in the grain structure, which are not visible on the run, but make it very hard to get a key or pull RGB channels apart. In these cases there are kinder ways to scale without the sharpening, such as lanczos or guassian. Sharpening has ramifications in post-processing and manipulation.

RAW footage ideally has the least done to it as possible, so you can do what you want with it later. RED Cine offers sharpening as one of the output functions, but basically you will be wanting to do the final sharpening when you are doing the final grade. There are lot of different sharpening functions, so the final “look” should be a consideration when deciding how much and what kind to wind in.

More reading on filters (pdf).