telecine

The telecine is the name given to various optical and electronic techniques to convert a movie theater turned on silver film video source. By extension, the term refers to the process of adapting a movie sequence to the chosen digital format. This processing is based on the addition of complementary frames on the still image, which allows the transfer of any silver film ( 8 mm, Super 8 , 9.5 mm , 16 mm , 35 mm , etc.), whatever its projection speed, towards a digital medium.

Television

Professional cinema films are filmed at 24 frames per second. Each image is a photogram, similar to a photograph. The usual television standards ( PAL , SECAM , and NTSC ) broadcast interlaced images . This means that each image is divided into two half-frames called “frames”: the first is composed of the even lines of the image, the other of the odd lines. The two frames of the same image are transmitted (and reproduced on the screen) one after the other. Thus, in television, the important cadence is not the cadence of the images but the frame rate: 50 frames per second in PAL and SECAM, 59.94 framesper second in NTSC. So there is usually no exact match between the rate of movies and that of television, so we need to find a way to convert the rate of movies to match the pace required by the television standard without variation of speed nor perceptible jerks ..

Principles and functioning

A viewing apparatus adapted to the cinema film is coupled to a sensor ( video camera or computer scanner). Each image composing the film is captured by the camera, then processed and saved. Image processing and saving can be analog or, like today, digital .

The very first type of telecine was a projector and a television camera . Before the invention of the video recorder , the broadcast was necessarily live. This principle was exploited by the television channels between 1930 and 1955. In France, the live broadcast on the air continues at least until 1983: all advertising and feature films for the 35 mm , documentaries, French productions and American series for the 16 mm (even if these are shot in 35 mm). With the invention of the VCR and the possibility of recording a video image, it was then replaced by an autonomous device combining both the projection system and the video recording unit.

Since the 1980s, electronic circuits and digital systems have facilitated the conversion of the cinematographic source. These devices compensate for speed variations, image artifacts and undesirable effects.

”  Pull up  ” and ”  pull down  “

Table of usual pull-ups and pull-downs for audio

The 2: 2 pulldown

The 2: 2 pulldown is the technique used in PAL and SECAM .

The principle is to increase the frequency of the film to 25 frames per second. To do this, we accelerate the film very slightly, this acceleration will not be noticed by the viewer because it is only 4% (about 2.5 seconds less per minute). The film becomes shorter when it is broadcast on television in countries using PAL and SECAM 1 . Turning the movie faster means that all sounds are transposed one semitone up. To overcome this disadvantage, we use nowadays a “harmonizer” which goes down the sounds of a quarter of a tone without slowing them down.

This can only be noticed by comparing the film with an NTSC version or by comparing the film’s music on television (accelerated and higher) with that of the original (unmodified) soundtrack.

To convert the film from 25 frames per second to 50 frames per second, it is now sufficient to separate each image of the film into two frames.

This technique is also used for encoding movies on VHS and DVD (PAL or SECAM). It is called “2: 2 pulldown” because for two consecutive images of the film, each is transmitted on two frames.

The 3: 2 pulldown

If we take the 2: 2 pulldown system to apply it to the NTSC standard , we would have to accelerate the film from 24 frames per second to about 30 frames per second, which would mean an acceleration of about 25%, or 15 minutes for 1 hour . The viewer would be extremely disturbed, which is why the NTSC does not speed up the movie; on the contrary, he slows it down very slightly.

The NTSC has a frame rate of 59.94 frames per second, which is very close to 60. Now,{\ displaystyle 60/24 = 2.5} : each image of the film must be broadcast in 2.5 frames to be synchronous. Since this is not possible, instead of transmitting, for each pair of consecutive images, an image in three frames, then an image in two frames.

Implementation

This transmission of an image of the film in three frames instead of two may surprise, because if a frame is a half-image, then an image is necessarily composed of two frames.

In reality, the frames are independent: as they are transmitted and displayed one after the other, two frames of the same image can quite represent different moments. Everything happens as if you were transmitting 60 frames per second, provided you respect the alternation between odd-numbered lines and even-numbered lines.

Thus, when a movie image is transmitted in three frames, it is first separated into two frames (there are only even and odd lines). When each of the two frames of the image has been transmitted, the first transmitted frame (for example the frame of the odd lines) is retransmitted, the next picture starting with the other frame (here, the frame of the even lines).

For example, for four consecutive cinema images (numbered from 1 to 4), a sequence of this type will be transmitted:

  • 1i – 1p – 2i – 2p – 2i – 3p – 3i – 4p – 4i – 4p

If we group each pair “odd field / even field”, we obtain five interlaced images:

  • (1i-1p) (2i-2p) (2i-3p) (3i-4p) (4i-4p)

Only three frames represent actual images, the other two represent two different image halves (which is not a problem because the frames will be displayed one after the other).

Actual frequency

This calculation assumed that the NTSC frequency was 60 frames per second. Since it is actually 59.94 frames per second, you only need to slow down the original movie to 23.976 frames per second ({\ displaystyle 23.976 = 59.94 / 2.5}), a mere 0.1% decrease in the frame rate, which is out of all proportion to the 4% acceleration caused by the 2: 2 pulldown in PAL . The film rate is thus almost strictly transposed into NTSC.

Remarks

This technique is also used to code movies on VHS NTSC. On DVD and Blu-ray , you usually just encode the movie at 23.976 frames per second on the disc, the 3: 2 pulldown being played by the player.

If the speed of the film is almost equal to the original speed (unlike what happens in PAL and SECAM), this technique suffers from a major drawback: the images of the film do not have the same duration [why ?] . This can be noticed very easily during panoramics, which appear less fluid than in cinema.

Amateur telecine

The recovery of silver films, often one of the 8 mm variants, can be achieved with a relatively simple material.

It is necessary to have:

  • a cinema projector, preferably with variable speed
  • a camera, preferably digital
  • a translucent screen with mirror or opaque

Advantages and disadvantages of a translucent screen

  • With this equipment, it is quite simple to obtain an image without much distortion, because the camera is placed in front of the axis of the projector.
  • The grain of the screen may be visible, if it is not of poor quality and if the screen is big enough
  • The contrast is strong and the value of white tends to be saturated
  • There is a hot spot, that is, the center is lighter than the edges

Advantages and disadvantages of an opaque screen

  • Easier to use (a sheet of non-granular white paper, like photo paper for a printer, is enough)
  • Problems of parallax and deformation, even of focus (prefer manual mode in all cases).
  • Reflections and shadows

In fact, these defects are generally minor, which makes the two methods roughly equivalent.

Film scanner

The film scanner is the most advanced solution both for its performance and the quality obtained. This results in the disappearance of the main optical defects, both of the projector and the camera ( curvature of field , vignetting , etc.), improved synchronism of captured images, sharp edges, digital calibration, compensation of contrasts, etc.

However, this type of material is not available to the general public and the complete capture of a feature film requires complex handling and is a long-term job.

Telecine and Image Format

Main article: Projection format .

Among the settings of the telecine, one of the most complex – apart from the image calibration – is the adaptation of the proportions of the original film to that of the television. For example, video and television use the conventional 4 : 3 image format , which is 1.33 times wider than it is high. The “wide” version is 16/9 , which is 1.78 times wider than it is tall. On the other hand, the cinema formats most often exploit another ratio (the image is often wider, with typically a ratio of 1.37, 1.66, 1.85 or 2.35).

Here are the different methods used:

  • Letterbox (mailbox). The image is respected and black bands appear at the top and bottom of the screen: a certain “waste” of the surface of the screen exists but the artistic choice of the director is respected;
  • Cropping . The image is cropped: the entire TV screen is used but a portion of the image at the edges is removed;
  • Cropping and anamorphosis . The image is slightly squeezed vertically and is allowed to overflow horizontally. This compromise “fills” the screen but significantly modifies the original artistic setting;
  • Wide anamorphic format . The image is compressed (by anamorphosis ) and therefore distorted.

None of these methods is ideal, the choice of the technique depends on the available material, the habits or the preferences of the public or that of the technicians. Only the video projection allows to respect all the formats (the black edges are not projected, which allows an infinity of compatibility of source formats).

Notes and references

  1. ↑ For example, Revenge of the Sith , 2  h  20 cinema, takes only 2  h  15 on DVD PAL.

Leave a Comment