Telecine

In countries that use the PAL or SECAM video standards, film destined for television is photographed at 25 frames per second. The PAL video standard broadcasts at 25 frames per second, so the transfer from film to video is simple; for every film frame, one video frame is captured.

Theatrical features originally photographed at 24 frame/s are shown at 25 frame/s. While this is usually not noticed in the picture (but may be more noticeable during action speed, especially if footage was filmed undercranked), the 4% increase in playback speed causes a slightly noticeable increase in audio pitch by about 0.707 semitones, which is sometimes corrected using a pitch shifter, though pitch shifting is a recent innovation and supersedes an alternative method of telecine for 25 frame/s formats.

2:2 pulldown is also used to transfer shows and films, photographed at 30 frames per second, like Friends and Oklahoma! (1955),[4] to NTSC video, which has ~59.94 Hz scanning rate. This requires playback speed to be slowed by a tenth of a percent.

Although the 4% speed increase has been standard since the early days of PAL and SECAM television, recently a new technique has gained popularity, and the resulting speed and pitch of the telecined presentation are identical to that of the original film. This pulldown method[5] is sometimes used in order to convert 24 frame/s material to 25 frame/s. Usually, this involves a film to PAL transfer without the aforementioned 4% speedup. For film at 24 frame/s, there are 24 frames of film for every 25 frames of PAL video. In order to accommodate this mismatch in frame rate, 24 frames of film have to be distributed over 50 PAL fields. This can be accomplished by inserting a pulldown field every 12 frames, thus effectively spreading 12 frames of film over 25 fields (or "12.5 frames") of PAL video. The method used is 2:2:2:2:2:2:2:2:2:2:2:3 (Euro) pulldown (see below). This method was born out of a frustration with the faster, higher pitched soundtracks that traditionally accompanied films transferred for PAL and SECAM audiences. A few motion pictures are beginning to be telecined this way. It is particularly suited for films where the soundtrack is of special importance.

When a TV station in an NTSC region airs a film or show that uses a PAL printing/version, but is being broadcast in the NTSC format, sometimes they do not perform the proper PAL to NTSC pulldown conversion or it is done improperly. This causes the program to be sped-up slightly and/or sound higher-pitched, due to the faster rate of the PAL 576 lines/50 Hz vs the NTSC 480 lines/~59.94 Hz format.

Mainframe Entertainment used a novel process for its TV shows. They are rendered at exactly 25.000 frames per second; then, for PAL/SECAM distribution, ordinary 2:2 pulldown is applied, but for NTSC distribution, 199 fields out of every 1001 are repeated. This brings the refresh rate from 25 frames to exactly 60,000/1001, or ~59.94, fields per second, with no change whatsoever in speed, duration, or audio pitch.

In the United States and other countries where television uses the 59.94 Hz vertical scanning frequency, video is broadcast at ~29.97 frame/s. For the film's motion to be accurately rendered on the video signal, a telecine must use a technique called the 2:3 pulldown, also known as 3:2 pulldown, to convert from 24 to ~29.97 frame/s.

The term "pulldown" comes from the mechanical process of "pulling" (physically moving) the film downward within the film portion of the transport mechanism, to advance it from one frame to the next at a repetitive rate (nominally 24 frames/s). This is accomplished in two steps. The first step is to slow down the film motion by 1/1000 to 24,000/1001 (~23.976) frames/s. The difference in speed is imperceptible to the viewer. For a two-hour film, play time is extended by 7.2 seconds. If the total playback time must be kept exact, a single frame can be dropped every 1000 frames.

The second step of the 2:3 pulldown is distributing cinema frames into video fields. At 23.976 frame/s, there are four frames of film for every five frames of 29.97 frame/s video:

${\displaystyle {\frac {23.976}{29.97}}={\frac {4}{5}}}$

These four frames are "stretched" into five by exploiting the interlaced nature of 60 Hz video. For every frame, there are actually two incomplete images or fields, one for the odd-numbered lines of the image, and one for the even-numbered lines. There are, therefore, ten fields for every four film frames, which are called A, B, C, and D. The telecine alternately places A frame across two fields, B frame across three fields, C frame across two fields and D frame across three fields. This can be written as A-A-B-B-B-C-C-D-D-D or 2-3-2-3 or simply 2–3. The cycle repeats itself completely after four film frames have been exposed:

A 3:2 pattern is identical to the one shown above except that it is shifted by one frame. For instance, a cycle that starts with film frame B yields a 3:2 pattern: B-B-B-C-C-D-D-D-A-A or 3-2-3-2 or simply 3–2. In other words, there is no difference between the 2-3 and 3-2 patterns. In fact, the "3-2" notation is misleading because according to SMPTE standards for every four-frame film sequence the first frame is scanned twice, not three times.[6]

The above method is a "classic" 2:3, which was used before frame buffers allowed for holding more than one frame. The preferred method for doing a 2:3 creates only one dirty frame in every five (i.e. 3:3:2:2 or 2:3:3:2 or 2:2:3:3); while this method has slightly more judder, it allows for easier upconversion (the dirty frame can be dropped without losing information) and a better overall compression when encoding. The 2:3:3:2 pattern is supported by the Panasonic DVX-100B video camera under the name "Advanced Pulldown". Note that just fields are displayed—no frames hence no dirty frames—in interlaced display such as on a CRT. Dirty frames may appear in other methods of displaying the interlaced video.

Similar techniques must be used for films shot at "silent speeds" of less than 24 frame/s, which includes home movie formats (the standard for Standard 8 mm film was 16 fps, and 18 fps for Super 8 mm film) as well as silent film (which in 35 mm format usually was 16 fps, 12 fps, or even lower).

• 16 frame/s (actually 15.985) to NTSC 30 frame/s (actually 29.97): pulldown should be 3:4:4:4
• 16 frame/s to PAL 25: pulldown should be 3:3:3:3:3:3:3:4 (a better choice would be to run the film at 16.67 frame/s, simplifying pulldown to 3:3)
• 18 frame/s (actually 17.982) to NTSC 30: pulldown should be 3:3:4
• 20 frame/s (actually 19.980) to NTSC 30: pulldown should be 3:3
• 27.5 frame/s to NTSC 30: pulldown should be 3:2:2:2:2
• 27.5 frame/s to PAL 25: pulldown should be 1:2:2:2:2

Also, other patterns have been described that refer to the progressive frame rate conversion required to display 24 frame/s video (e.g., from a DVD player) on a progressive display (e.g., LCD or plasma):[7]

• 24 frame/s to 96 frame/s (4x frame repetition): pulldown is 4:4
• 24 frame/s to 120 frame/s (5x frame repetition): pulldown is 5:5
• 24 frame/s to 120 frame/s (3:2 pulldown followed by 2x deinterlacing): pulldown is 6:4

The "2:3 pulldown" telecine process creates a slight error in the video signal compared to the original film frames that can be seen in the above image. This is one reason why films viewed on typical NTSC home equipment may not appear as smooth as when viewed in a cinema and PAL home equipments. The phenomenon is particularly apparent during slow, steady camera movements which appear slightly jerky when telecined. This process is commonly referred to as telecine judder. Reversing the 2:3 pulldown telecine is discussed below.

PAL material in which 2:3 (Euro) pulldown has been applied suffers from a similar lack of smoothness, though this effect is not usually called "telecine judder". Effectively, every 12th film frame is displayed for the duration of three PAL fields (60 milliseconds), whereas the other 11 frames are each displayed for the duration of two PAL fields (40 milliseconds). This causes a slight "hiccup" in the video about twice a second. It is increasingly being referred to as Euro pulldown as it largely affects European territories.

Some DVD players, line doublers, and personal video recorders are designed to detect and remove 2:3 pulldown from telecined video sources, thereby reconstructing the original 24 frame/s film frames. Many video editing programs such as AVIsynth also have this ability. This technique is known as "reverse" or "inverse" telecine. Benefits of reverse telecine include high-quality non-interlaced display on compatible display devices and the elimination of redundant data for compression purposes.

Reverse telecine is crucial when acquiring film material into a digital non-linear editing system such as Lightworks, Sony Vegas Pro, Avid, or Final Cut Pro, since these machines produce negative cut lists which refer to specific frames in the original film material. When video from a telecine is ingested into these systems, the operator usually has available a "telecine trace," in the form of a text file, which gives the correspondence between the video material and film original. Alternatively, the video transfer may include telecine sequence markers "burned in" to the video image along with other identifying information such as time code.

It is also possible, but more difficult, to perform reverse telecine without prior knowledge of where each field of video lies in the 2:3 pulldown pattern. This is the task faced by most consumer equipment such as line doublers and personal video recorders. Ideally, only a single field needs to be identified, the rest following the pattern in lock-step. However, the 2:3 pulldown pattern does not necessarily remain consistent throughout an entire program. Edits performed on film material after it undergoes 2:3 pulldown can introduce "jumps" in the pattern if care is not taken to preserve the original frame sequence (this often happens during the editing of television shows and commercials in NTSC format). Most reverse telecine algorithms attempt to follow the 2:3 pattern using image analysis techniques, e.g. by searching for repeated fields.

Algorithms that perform 2:3 pulldown removal also usually perform the task of deinterlacing. It is possible to algorithmically determine whether video contains a 2:3 pulldown pattern or not, and selectively do either reverse telecine (in the case of film-sourced video) or bob-deinterlacing (in the case of native video sources).

The parts of a flying spot scanner: (A) cathode-ray tube (CRT); (B) film plane; (C) & (D) dichroic mirrors; (E), (F) & (G) red-, green- and blue-sensitive photomultipliers

In the United Kingdom, Rank Precision Industries was experimenting with the flying-spot scanner (FSS), which inverted the cathode ray tube (CRT) concept of scanning using a television screen. The CRT emits a pixel-sized electron beam which is converted to a photon beam through the phosphors coating the envelope. This dot of light is then focused by a lens onto the film's emulsion, and finally collected by a pickup device. In 1950 the first Rank flying spot monochrome telecine was installed at the BBC's Lime Grove Studios.[8] The advantage of the FSS is that colour analysis is done after scanning, so there can be no registration errors as can be produced by vidicon tubes where scanning is done after colour separation—it also allows simpler dichroics to be used.

In a flying spot scanner (FSS) or cathode-ray tube (CRT) telecine, a pixel-sized light beam is projected through exposed and developed motion picture film (either negative or positive) and collected by a special type of photo-electric cell known as a photomultiplier which converts the light into an electrical signal. The beam of light "scans" across the film image from left to right to record the horizontal frame information. Vertical scanning of the frame is then accomplished by moving the film past the CRT beam. In a colour telecine the light from the CRT passes through the film and is separated by dichroic mirrors and filters into red, green and blue bands. Photomultiplier tubes or avalanche photodiodes convert the light into separate red, green and blue electrical signals for further electronic processing. This can be accomplished in "real time", 24 frames per second (or in some cases faster). Rank Precision-Cintel introduced the "Mark" series of FSS telecines. During this time advances were also made in CRTs, with increased light output producing a better signal-to-noise ratio and so allowing negative film to be used.

The problem with flying-spot scanners was the difference in frequencies between television field rates and film frame rates. This was solved first by the Mk. I Polygonal Prism system, which was optically synchronised to the television frame rate by the rotating prism and could be run at any frame rate. This was replaced by the Mk. II Twin Lens, and then around 1975, by the Mk. III Hopping Patch (jump scan). The Mk. III series progressed from the original "jump scan" interlace scan to the Mk. IIIB which used a progressive scan and included a digital scan converter (Digiscan) to output interlaced video. The Mk. IIIC was the most popular of the series and used a next generation Digiscan plus other improvements.

The "Mark" series was then replaced by the Ursa (1989), the first in their line of telecines capable of producing digital data in 4:2:2 color space. The Ursa Gold (1993) stepped this up to 4:4:4 and then the Ursa Diamond (1997), which incorporated many third-party improvements on the Ursa system.[9] Cintel's C-Reality and ITK's Millennium flying-spot scanner are able to do both HD and Data.

The parts of a CCD scanner: (A) Xenon bulb; (B) film plane; (C) & (D) prisms and/or dichroic mirrors; (E), (F) & (G) red-, green- and blue-sensitive CCDs.

The Robert Bosch GmbH, Fernseh Div., which later became BTS Inc. – Philips Digital Video Systems, Thomson's Grass Valley and now is DFT Digital Film Technology introduced the world's first CCD telecine (1979), the FDL-60. The FDL-60 designed and made in Darmstadt West Germany, was the first all solid state telecine.

Rank Cintel (ADS telecine 1982) and Marconi Company (1985) both made CCD Telecines for a short time. The Marconi model B3410 telecine sold 84 units over a three-year period,[10] and a former Marconi technician still maintains them.[11]

In a charge-coupled device Line Array CCD telecine, a "white" light is shone through the exposed film image into a prism, which separates out the image into the three primary colors, red, green and blue. Each beam of colored light is then projected at a different CCD, one for each color. The CCD converts the light into electrical impulses which the telecine electronics modulate into a video signal which can then be recorded onto video tape or broadcast.

Shadow telecine system, produced by Grass Valley (formerly Thomson, originated from Bosch-Fernseh's inventions), installed at DR, Denmark

Philips-BTS eventually evolved the FDL 60 into the FDL 90 (1989) / Quadra (1993). In 1996 Philips, working with Kodak, introduced the Spirit DataCine (SDC 2000), which was capable of scanning the film image at HDTV resolutions and approaching 2K (1920 Luminance and 960 Chrominace RGB) × 1556 RGB. With the data option the Spirit DataCine can be used as a motion picture film scanner outputting 2K DPX data files as 2048 × 1556 RGB. In 2000 Philips introduced the Shadow Telecine (STE), a low cost version of the Spirit with no Kodak parts. The Spirit DataCine, Cintel's C-Reality and ITK's Millennium opened the door to the technology of digital intermediates, wherein telecine tools were not just used for video outputs, but could now be used for high-resolution data that would later be recorded back out to film.[9] The DFT Digital Film Technology, formerly Grass Valley Spirit 4K/2K/HD (2004) replaced the Spirit 1 Datacine and uses both 2K and 4K line array CCDs. (Note: the SDC-2000 did not use a color prisms and/or dichroic mirrors.) DFT revealed its new scanner at the 2009 NAB Show, Scanity.[12] The Scanity uses time delay integration (TDI) sensor technology for extremely fast and sensitive film scans. High speed scanning 15 frame/s @ 4K; 25 frame/s @ 2K; 44 frame/s @ 1K.

With the manufacturing of new high power LEDs, came pulsed LED/triggered three CCD camera systems. Flashing the LED light source for a very short time span gives the full frame CCD camera a stop action of the film, allowing continuous film motion. With CCD video cameras that have a trigger input, the camera now can be electronically synced up to the film transport framing. There are now a number of retail and home made Pulsed LED/triggered camera systems. An array of high-power multiple red, green and blue LEDs is pulsed just as the film frame is positioned in front of the optical lens. The camera sends the single, non-interlaced image of the film frame to a digital frame store, where the electronic picture is clocked out at the selected TV frame rate for PAL or NTSC or other standard. More advanced systems replace the sprocket wheel with laser or camera based perf detection and image stabilization system.

Telecine technology is increasingly merging with that of motion picture film scanners; high-resolution telecines, such as those mentioned above, can be regarded as film scanners that operate in real time.

As digital intermediate post-production becomes more common, the need to combine the traditional telecine functions of input devices, standards converters, and colour grading systems is becoming less important as the post-production chain changes to tapeless and filmless operation.

However, the parts of the workflow associated with telecines still remain, and are being pushed to the end, rather than the beginning, of the post-production chain, in the form of real-time digital grading systems and digital intermediate mastering systems, increasingly running in software on commodity computer systems. These are sometimes called virtual telecine systems.

Some video cameras and consumer camcorders are able to record in progressive "24 frames/s" or "23.976 frames/s". Such a video has cinema-like motion characteristics and is the major component of so-called "film look" or "movie look".

For most "24 frames/s" cameras, the virtual 2:3 pulldown process is happening inside the camera. Although the camera is capturing a progressive frame at the CCD, just like a film camera, it is then imposing an interlacing on the image to record it to tape so that it can be played back on any standard television. Not every camera handles "24 frames/s" this way, but the majority of them do.[13]

Cameras that record 25 frames/s (PAL) or 29.97 frames/s (NTSC) do not need to employ 2:3 pulldown, because every progressive frame occupies exactly two video fields. In the video industry, this type of encoding is called progressive segmented frame (PsF). PsF is conceptually identical to 2:2 pulldown, only there is no film original to transfer from.