|
|
[Partners] [Contact] [French] [Top] [Previous] [Next] |
The BRAVA project aims at developing further the results of the AURORA project, to ensure the widest possible access to valuable TV broadcast archive material by significantly enhancing the efficiency of the video and film programme restoration process and preparing the restored material for dissemination via multi-resolution digital video broadcasting standards.
A pre-requisite to designing tools for correcting impairments affecting video-originated and film-originated material, was to prepare a complete survey of these impairments. When one tries to reference a problem on a video programme, there is often a vocabulary problem, as no standard dictionary exists for these problems. Different people use the same word for different problems (jitter, twitter...), or vice-versa. Additionally terms are borrowed from English and used improperly (scratch is commonly used instead of dirt and sparkle). Therefore this work intends to reduce the ambiguities by providing for each impairment, a code, one or several common names, both in English and in French, and a textual definition in a few lines.This document is the result of the refinement, within the scope of the Brava Project, of the impairments list prepared at the beginning of the Aurora project. It is very much oriented towards completeness. Every know impairment is described in a few lines. Therefore technical details may sometime be missing, and the readers are invited to look into the relevant technical reference documents if they want to get into more details.
Some impairments lists already exist, published by the broadcasters,
in order to be able to reference the impairments met during tape checking
before broadcast. These lists provide a good basis for starting, but are
in no way as complete as the dictionary presented hereafter. This dictionary
contains overall 190 different referenced impairments, although in some
cases some of them are only sub-categories (for example several types of
drop-outs are referenced).
This dictionary was originally intended for internal project use, but it is considered that it could be profitable to the whole broadcasting industry, therefore it is made available here. There are probably still errors, and even more likely divergences on the names or the signification or the employed terms, therefore this work should be considered and an ongoing process : please address comments, suggestions, and requests for changes to : brava_mgt@ina.fr.
The interested reader may also want to look at the following links :
http://ourworld.compuserve.com/homepages/PeterFinklestone/2inchQuad.htm is a complete guide of impairments due to 2 Inch VTR origination, with picture samples.
http://users.ox.ac.uk/~photo/facts/defects.pdf is a list of film faults and their solutions.
Copyright : copying and exploiting part or totality of this page for
internal use is authorised, provided an explicit reference is made to the
origin of the information. Do not publish on the web the totality of the
page : make a reference to the current page instead. For other uses, please
get in touch with brava_mgt@ina.fr.
Problems due to film technology, handling, scratching, ageing, storage conditions, are numerous, and some of them well know (dirt and sparkle, scratches, unsteadiness...). Some others are less documented such as flicker, dirty splices, fading...
2. Film recording (kinescope) impairments
It is worth noting that the use of VTRs for producing TV programmes took off much later than the actual broadcast start. To keep memory of the programmes directly broadcast from studios or from events, The only solution was to use a film recording device : kinescope was more ore less a film camera aimed at a video monitor. In addition to the usual film impairments, some problems are specific of film recording of TV pictures : kinescope moiré, suppressed field problems...
3. Video problems related to the VTR technology
The Video Tape Recording Technology has evolved considerably from the introduction of the 2 Inch quad in 1956, to Digital Betacam, in 1993, and is still evolving at fast pace. The heritage is a constellation of different standards, and playing back the tapes is often difficult, when not impossible. In addition to the impairments due to the standard itself (bandwidth reduction, drop-outs...), other problems may be caused by poor settings during the original recording, equipment obsolescence, know-how losses, tape ageing and storage conditions. The most recurrent problem is the drop-outs problem (transient signal loss due to loss of contact between head and tape), but may other problems can arise, such as head-clogging or synchronisation problems.
4. Video problems related to the composite colour systems
The technical history of Television is paved of compromises between different constraints. Among others, the requirement to transmit colour in addition to luminance, still keeping the same bandwidth, has led the engineer to pack the colour information too near from the luminance, causing in some cases undesired leakage between the two, called cross-chroma and cross-luminance. Other problems can also arise, detailed in the list.
5. Video problems related to video transmission
Analogue video transmission is supposed to be transparent. But, cables, RF links, amplifiers, filters, multiple paths, interferences, result in several different problems referenced in the list : bandwidth problems, echoes, overshoot, streaking, sound in picture...
6. Problems related to video shooting
It can be questioned whether impairments resulting from the original camera should be considered for restoration. Nevertheless these impairments do exist and are worthwhile mentioning, such as colour convergence problems, comet tails...
7. Problems related to video edition and post-production
As for the previous item, it is sometimes difficult to decide whether an impairment should be corrected or kept because accepted and desired by the director. However the analogue technology for post-production very often led to impairments one would like to correct now : early chroma-keying, PAL shifts...
8. Other video problems related to VTRs, transmission, and shooting
Some problems whose origin are multiple, or cannot be determined strictly are classified here : they include continuous and impulsive noise, line jitter, bearding, off-locks...
9. Problems related to standard conversion
Early Television standards used a different number of lines per picture : 405, 819, instead of the current 525 and 625 lines. Even now, two different frame rates co-exist : 30 and 25 Hz. Conversion devices were used (and are still in use) to convert from one standard to another. These conversion systems could be very primitive, and generate their own problems.
10. Other video-related problems
This category contains other problems related to video, that cannot be classified in the previous categories.
This category contains other problems related any kind of media (video or film), that cannot be classified in the previous categories.
Although there may be more comprehensive lists of audio impairments, here have been referenced the most common television programmes soundtrack problems. Some are common with audio-only technology, some are more specific of television programmes, such as synchronisation problems or vinegar syndrome.
|
|
|
|
|
|
| F1 | Film Grain | Grain Film |
|
Overlaying
of film grains (processed silver halide crystals) at different depths in
emulsion. random patterning, loss of focus, density errors.
|
| F2 | Dirt | Salissures |
|
Surface dirt, ingrained dirt. Can be printed/duped in copying stock from original. |
| F3 | Neg. sparkle | Salissures sur Negatif |
|
Fine specs of dirt, dust printed through from original onto print stock. |
| F4 | Gelatin Loss | Eclats de gélatine |
|
Gelatin
was used as vehicle for carrying silver halide in film emulsion.
Image content removed when scratched. |
| F5 | Severe Emulsion Loss | Pertes importantes de gélatine |
|
Extreme cases of gelatin losses : the information is wiped off the film support on a large area of the image, for long periods. |
| F6 | Base scratches | Rayures coté support |
|
Black in appearance in print stock, commonly seen as tramlines. White in appearance if printed through from negative stock. |
| F7 | Emulsion scratches | Rayures
coté émulsion |
|
As above except it can cause discoloration, dependent on depth of scratch. |
| F8 | Film Hair | Poussières |
|
Duped/printed in copying stock. original camera fault, hair in gate camera or printer. Fraying of film due to physical damage. |
| F9 | Film scuffing | Egratignures multiples |
|
Small type scratches due to poor loose wind on film. Usually found at head and tail of material. |
| F10 | Visible Scotch-tape cut edition or repair | Collant Scotch visible |
|
Scotch-tape edition is often visible : the scotch tape is dirty, or coloured. Trapped in dirt, racking problems, frame bar errors, clicking on optical tracks. |
| F11 | Ageing Scotch tape joins, dirty splice | Collant Scotch Détérioré |
|
As above except old joins tend to seep glue causing film material to stick damaging emulsion. |
| F12 | Black Images | Images Noires |
|
Some images may be black, due to a problem in the camera shutter, or lab printer problems. |
| F13 | Colour Grading problems | Défaut d’Etalonnage |
|
Each
scene has to be corrected differently. Poor grading causes unbalanced colour
changes in scenes and density / contrast problems.
It may happen that colour grading is not applied exactly at the cut : a few frames will show different grading. |
| F14 | Black and White Grading | Défaut d’Etalonnage noir et blanc |
|
Printer light setting errors at printing stage causing density contrast problems . |
| F15 | Colour fading | Virage Colorimétrique |
|
One or several colour layers have changed or disappeared |
| F16 | Varying colour | Instabilité couleur |
|
Same problem, but the change is not uniform in time . Individual colour layers fading at different rates. Positive cut material from different origins. Duped in material from different sources. |
| F17 | Non-Uniform colour | Couleur non uniforme |
|
Same problem, but the change is not uniform in space (Film Edges) |
| F18 | Scene cut Image Instability | Saute de changement de plan |
|
Scotch-tape edition may not be perfectly realised, or may have moved : the perforations are not correctly aligned or damaged, hence a jerk around the scene cut on some telecines. |
| F19 | Film unsteadiness | Instabilité film |
|
Film
transport-generated Image unsteadiness : the picking mechanism on perforations
was not reliable enough, or the perforations were damaged.
Also know as 'Position Flicker' : 'Unsteadiness' is preferable. |
| F20 | Planarity problems | Défaut de planéité |
|
The film is not kept planar : the images are non-uniformly distorted and unfocused. |
| F21 | Veil ? | Voile |
|
Accidental exposition of the film to light before or during development, resulting in clearer or darker images (usually at reel ends) |
| F22 | Mixed Geometry | Changement de Géométrie |
|
Mixed geometry in positive cut and duped material. |
| F23 | Vinegar Syndrome | Syndrome du Vinaigre |
|
Breakdown of gelatine due to poor ambient storage conditions. In extreme cases loss of signal. |
| F24 | Flicker (Luminance Flicker) | Pompage, Battements, Flicker |
|
Variations of intensity, possibly due to uneven chemical degradation of the film pack, or uneven shutter operation. |
| F25 | Twin-Lens Telecine Flicker | Battements de Telecine à double trajet optique |
|
In Twin-Lens Telecines, poor setup or dust on the two optical paths results in a different intensity for even and odd fields. The visual effect is of a fast permanent flicker (at frame rate). Subtle occurrences are met, but very few extreme cases : the problem is easily solved by proper setup of the telecine. Furthermore, Twin-lens telecines are less and less exploited. |
| F26 | Twin-Lens Telecine Unsteadiness | Instabilité Telecine à double trajet optique |
|
In Twin-Lens Telecines, poor setup of the electronic or physical registration between the two optical paths results in a different position, scaling, and angle for even and odd fields. The visual effect is of a fast permanent judder (at frame rate). Subtle occurrences are met, but very few extreme cases : the problem is easily solved by proper setup of the telecine. Furthermore, Twin-lens telecines are less and less exploited. |
2. Film Recording (Kinescope)
Impairments
|
|
|
|
|
|
| K1 | Setup Error of glycerine ND filter | Défaut de réglage du filtre à glycérine |
|
Due
to incorrect setup of glycerine ND (?) filter used to correct luminance
decay between stored and active picture, there will be a luminance error
approximately 1/3 of the way down the frame. |
| K2 | Severe Phasing | Défaut de Phase Kinescope |
|
Loss
of synchronisation between camera and display may result in several peculiar
effects including two non matching fields being recorded on the same frame
of film making a rather strange half mix.
Very common in early recordings. |
| K3 | Suppressed field pixelation | Défaut de Kinescope à Trame Unique |
|
In order to allow non fast pull down cameras one field of the image is omitted. This leads to loss of vertical resolution - visible line structure and pixelation. |
| K4 | Spot wobble resolution impairment |
|
To
eliminate suppressed filed pixelation the scanning spot was wobbled vertically.
this produced an alternative impairment in which the image has astigmatism
with very soft focus in the vertical axis
used on both telecines and film recorders. |
|
| K5 | Vertical instability due to fast pull down camera | Instabilité verticale due a une Caméra a avance rapide |
|
Use
of non suppressed field display requires extremely rapid film movement which
leads to excessive vertical tearing and blurring |
| K6 | Defocused recording | Défaut de mise au point sur Kinescope |
|
An alternative to spot wobble was to simply defocus the camera lens so that line structure was not resolved. The problem is that everything else is also made very soft. |
| K7 | Motion blur due to long persistence tube | Défaut de persistance Kinescope |
|
Extra long persistence tubes were used for 35mm non suppressed field Kinescopes. This yields maximum resolution on stills but degenerates on motion due to tube memory. |
| K8 | Dirty window effect (kinescope) | Effet de fenêtre sale (kinescope) |
|
Dirt and fluff attracted to the CRT in the film recorder appears as though the scene were viewed through a dirty window. Most noticeable on pans and tracking shots. |
| K9 | Non linearity of scans | Défaut de linéarité Kinescope |
|
Due to poor linearity of original CRT display aspect ratio and/or vertical or horizontal linearity of image may be distorted |
| K10 | Kinescope MoirŽ (Line scan Aliasing & moirŽ) | Moiré Kinescope |
|
Occurs
when a previously film recorded item is the telecine transferred. Visible
line structure causes a beat pattern with telecine scanning, with a nearly
periodic phase variation.
The BBC flying spot Telecines were able to overcome this problem by modulating the beam careful line-up required to eliminate streaking. Can be eliminated depending on the type of TK used. |
| K11 | Kinescope Film impairment | Défauts Film Kinescope |
|
Nearly
every film impairment can be seen on kinescope documents.
Plus electronic problems such as streaking . Flare was another problem |
3. Video Problems related
to the VTR Technology
|
|
|
|
|
|
| V1 | 2 Inch Scratches | Rayures 2 Pouces |
|
Scoring
and damage causing a white quincunx-organised grid of dots to appear superimposed
on the video. Often mistaken for drop-out and switcher phase error.
It can be argued that this is the single most commonly encountered fault appearing on material originated from 2 Inch - as most of the others should have been eliminated at replay - whereas this - along with more general physical errors cannot be. |
| V2 | 2 Inch Tip Penetration Error (Geometrical Error : 2 Inch Venitian Blind or Jogs) | Défaut de Pénétration 2 Pouces (Défaut de Géométrie) : Stores vénitiens, Dents de Scie |
|
Vertical
lines and image edges exhibit stable 15-16 lines saw-tooth distortion due
to an incorrect setting of the vacuum guide projection. Usually corrected
by the 2 Inch internal TBCs. See "Broadcast Video Tape Recording Technology",
Todorovic, p 48
This should be corrected at replay - TBC on most VTRs have a wide window of tolerance |
| V3 | 2 Inch Height Error (Geometrical Error) : Scallop Error | Défaut de Hauteur 2 Pouces (Défaut de Géométrie) : Défaut de Feston |
|
Vertical
lines and image edges exhibit stable 15-16 lines scallop distortion due
to an incorrect setting of the guide height. Note the velocity error associated
to this problem. See "Broadcast Video Tape Recording Technology", Todorovic,
p 48
This should be corrected at replay - TBC on most VTRs have a wide window of tolerance |
| V4 | 2 Inch Quadrature Errors (Geometrical Error) | Défaut de Quadrature 2 Pouces (Défaut de Géométrie) |
|
Every
4th group of 15-16 lines is shifted due to an incorrect setting of the 90¡
angle between heads. Usually corrected by the 2 Inch internal TBCs if not
too important. See "Broadcast Video Tape Recording Technology", Todorovic,
p 46
Very rarely seen, as should be corrected at replay. |
| V5 | 2 Inch Velocity Errors | Erreur de vitesse 2 Pouces |
|
Synonym for scallop+tip Error. |
| V6 | 1 Inch B Tension Errors : venitian blind | Défaut de tension 1Pouce B : Stores vénitiens |
|
Same saw-tooth pattern as for 2 Inch, but with 52/53 blocks. Usually corrected by internal TBCs. |
| V7 | Flagging, hooking, 1 Inch C Tension Errors | Drapeau, Défaut de tension 1 Pouce C: |
|
In some formats such as 1"C the tape speed would not be held as tight as one might desire, or the Time base or Velocity error is larger than the TBC can compensate for, or there will be other problems (including guide problems as already mentioned) that will cause flagging. Visually this will be seen at the top of active video, the top lines will appear skewed in horizontal position and sometimes fold down on top of other lines. This can also be caused by head switching in the wrong area in vertical interval, but the important point here is that the top portion of the picture, as much as 10-20% can be skewed. It generally does not remain constant (that is why it is called flagging) but is extremely bothersome to watch. |
| V8 | Flagging, Hooking, 3/4" Skew Problems | Drapeau, Hooking, défaut de skew 3/4 " |
|
As for 1"C, and some other formats, the image is hook-shaped due to an incorrect tension of the tape. |
| V9 | 2 Inch Drop-Outs | Drop-Outs 2 Pouces |
|
The
duration of drop-outs on two Inch is usually of 1 to 10 usec, the number
of lines affected varies from 1 to 4 (SECAM). When DOCs are in use, the
line or part of line affected are usually replaced by one of the preceding
lines.
Can be worse on a VR 2000. |
| V10 | 1"B Drop-Outs | Drop-Outs 1 Pouce B |
|
The duration of drop-outs on 1"B is usually of 1 to 10 usec, the number of lines affected varies from 1 to 4 (SECAM). When DOCs are in use, the whole line (in SECAM), or part of line affected, are usually replaced by one of the preceding lines. |
| V11 | 1"C Drop-Outs | Drop-Outs 1 Pouce C |
|
The
duration of drop-outs on 1"C is usually of 1 to 10 lines. When DOCs are
in use (always ?), the line (in SECAM), or part of line affected, are usually
replaced by one of the preceding lines.
1"C dropout compensation is usually quite good. |
| V12 | 3/4 " Drop-outs | Drop-outs 3/4 |
|
The duration of drop-outs on 3/4 Inch VTRs is usually of 1 to 10 lines. When DOCs are in use (always ?), the line (in SECAM), or part of line affected, are usually replaced by one of the preceding lines. |
| V13 | Beta Drop-outs | Drop-outs Beta |
|
The duration of drop-outs on Beta and Beta SP is usually of 5 to 100 lines. The line, or part of line affected, are always replaced by the preceding line. This is a very common problem. |
| V14 | 2 Inch Head Switcher Phase Error | Erreur de commutations de têtes 2 Pouces |
|
The
commutation between heads during playback occurs late, which results in
the beginning of each 15-16 lines group being switched to black
Also known as head switcher phase - a std adjustment. Fault should not be seen in practice, but happens to be out-of-range if incorrectly recorded. Worse on VR 2000. |
| V15 | 2 Inch Tracking Error | Erreur de Tracking 2 Pouces |
|
The
heads are not aligned with the tracks, reduces the SNR (moirŽs). It is usually
due to an incorrect position of the control track during recording. If not
out-of range, it may be corrected by a manual tuning of the tracking parameter.
Should not occur only when tracking error is out-of-range of the manual setup. Can happen on an edit made with tach(ymeter ?). phase out and tracking error during line-up |
| V16 | 2 Inch Head Axial Error (Head-Banding) | Erreur de positionnement axial |
|
One
of the heads is not aligned with the tracks, which reduces the SNR of every
4th 15-16 lines group. See "Broadcast Video Tape recording Technology",
Todorovic, p 47.
very rare |
| V17 | 2 Inch Head azimut Error (Head-Banding) | Erreur d'azimut sur une tête 2 Pouces |
|
One
of the heads is not oriented correctly, which reduces the SNR of every 4th
15-16 lines group. See "Broadcast Video Tape recording Technology", Todorovic,
p 46.
very rare |
| V18 | 2 Inch Head-Banding (Equalisation) | Défaut d’égalisation 2 Pouces |
|
On
2 Inch, there are four head that alternatively read 15 or 16 lines blocks
; in case of incorrect equalisation, the signal will be more noisy, or band-limited,
or show moirŽ pattern, on 16-line blocks.
should be optimised at replay - residual errors are usually small |
| V19 | 2 Inch Head-clogging (Dirt) | Encrassement têtes 2 Pouces |
|
Different
kind of head-banding, due to dirt on one head, either during playback, or
during recording.
Should try to correct at replay because resultant increase in noise and loss of resolution will be quite severe. But if clogging occurred during recording, cannot be corrected during replay. |
| V20 | 2 Inch edge of track-Banding | Bruit de bord de bande 2 Pouces |
|
Due to the impact of the heads on the tape, the first line of each block could exhibit a lower SNR - can also result in hue and saturation errors. |
| V21 | 1"B Head-Banding (Equalisation) | Défaut d’égalisation têtes 1 Pouce |
|
On 1 Inch B, there are two head that alternatively read 56 lines blocks ; in case of incorrect equalisation, the signal will be more noisy, or band-limited, or show moirŽ pattern, on 16-line blocks. |
| V22 | 1"B Head-Banding (clogging) | Encrassement têtes 1 Pouce B |
|
Head-banding on 1 Inch B can be caused by head-clogging. |
| V23 | 2 Inch FM MoirŽ | Moiré FM 2 Pouces |
|
Specific
case of spectrum folding of the FM signal (Briefly described in "Broadcast
Video Tape recording Technology", Todorovic, p 22).
Much more noticeable on highly saturated areas |
| V24 | 1 Inch MoirŽ FM | Moiré FM 1 Pouce |
|
Spectrum folding, different from 2 Inch because of the different FM frequencies |
| V25 | 2 Inch Mechanical Editing Shift | Collant Mécanique 2 Pouces |
|
Early mechanical editing on 2 Inch give horizontal and/or vertical instabilities, or various kinds of drop-outs. |
| V26 | 2 Inch Tracking breaks | Perte de tracking 2 Pouces |
|
Temporary
deterioration of the control track give horizontal and/or vertical instabilities.
Worse when control track returns with a phase change. Can occur on assemble edits |
| V27 | 2 Inch Tape Problems | Problèmes de Bande 2 Pouces |
|
Extensive
scoring oxide shedding, white powder, sticky oxide, curling, and other physical
tape problems . Very large set of picture impairments. Also the most common
underlying cause of other problems listed above.
Very serious and usually virtually impossible to cure - examples are difficult to produce as one has to look inside the box- at the actual tape material to see the problem. The effect can render the tape unplayable. |
| V28 | 1 Inch B Tracking Error | Défaut de Tracking 1 Pouce B |
|
FM Noise due to tracking errors -> MoirŽs |
| V29 | 1 Inch C Tracking Error | Défaut de Tracking 1 Pouce C |
|
FM
Noise due to tracking errors (different from above) -> moirŽs
Error would probably remain constant |
| V30 | 1"C Equalisation Errors | Défaut d'égalisation têtes 1 Pouce |
|
Difference
of reaction between the equalisation of the two playback heads. Could result
in a difference between odd and even fields.
Usually not visible to the eye |
| V31 | 1"C Head-clogging | Encrassement têtes 1 Pouce |
|
Due
to dirt on one head, A different SNR between odd and even field appears,
possibly combined with field beating |
| V32 | 3/4" Equalisation Problems | Problème d'égalisation 3/4 |
|
Different SNR between odd and even fields, or field beating. There are several formats for 3/4, they all have different characteristics |
| V33 | Beta Equalisation Problems | Problèmes d'égalisation Beta |
|
Different SNR between odd and even fields, or field beating, for luminance or chrominance (2 heads for luminance, 2 heads for chrominance) |
| V34 | Field switching inside Image | Commutation Trame dans Image |
|
The switching between odd and even head occurs during the active period of the image : the head or the bottom of the image is lost. |
| V35 | Betacam chroma head clogging | Encrassement tête couleur Betacam |
|
Dirt on chroma head : only one field has chroma |
| V36 | Betacam Luminance head clogging | Encrassement tête luminance Betacam |
|
Dirt on one luminance head : only one field has luminance |
| V37 | Betacam saturation (bearding) | Saturation Betacam (flammèches) |
|
When picture is complex, with high frequencies areas, these areas may be followed by white flames. |
| V38 | other obsolete VTR Problems | Problèmes liés aux autres formats périmés de magnétoscopes |
|
Too many to enumerate them |
| V39 | Consumer VTR Problems | Problèmes liés aux magnétoscopes grand public |
|
For memory only ... |
4. Video Problems related
to the Composite colour systems
|
|
|
|
|
|
| C1 | PAL Cross-Colour | Cross-Color PAL |
|
Luminance information around the subcarrier centre frequency (4.43 MHz) is incorrectly interpreted as colour (Coloured MoirŽ). Inconvenience depends on the content of the picture. |
| C2 | PAL Delay-Line decoded | Décodage PAL avec Delay-Line |
|
Slight
vertical colour blurring, colour error on diagonal and Horizontal edges,
colour slightly shifted downwards.
Not Applicable in a broadcast environment |
| C3 | PAL Subcarrier Phase Error (with DL) | Erreur de Phase PAL (avec DL) |
|
A
subcarrier Phase Error decoded with a delay-line decoder results in Colour
Desaturation of uniform areas, colour errors on diagonal and horizontal
edges
Important to correct this before decoding |
| C4 | PAL Subcarrier Phase Error (without DL) : Venitian Blind | Persiennes (Erreur de Phase PAL (sans DL) |
|
A subcarrier Phase Error decoded with a simple PAL decoder results in Hue Errors alternating each line and each field. |
| C5 | PAL Cross-Luminance | Cross-Luminance PAL |
|
Residual Colour Subcarrier in Luminance Signal. Uniform areas and vertical edges. |
| C6 | SECAM Cross-Colour | Cross-Color SECAM |
|
Luminance information between 3.9 MHz and 4.75 MHz is incorrectly interpreted as colour (Coloured MoirŽ). More disturbing than in PAL and NTSC because "illegal colours" are generated. |
| C7 | SECAM Cross-Luminance | Cross-Luminance SECAM |
|
Residual Colour Subcarrier in Luminance Signal. Visible both in uniform areas and vertical edges. |
| C8 | SECAM Erroneous Horizontal. & Diagonal Edges | Erreur de couleur sur les Transitions Horizontales |
|
Errors on diagonals and horizontal edges, due to the use of last line's previous DR or DB information. |
| C9 | SECAM Colour Antiphase | Antiphase Couleur |
|
Permutation of DR and DB components (from a few lines to a whole field). May be due to loss or attenuation of field identification signals, loss, attenuation, or incorrect timing of colour burst. |
| C10 | SECAM Antiphase Edition | Montage en Antiphase Couleur |
|
The four-field sequence (DR,DB,DB,DR) may not be respected when editing early SECAM programs, decoding these programs results in a permutation of DR and DB components (from a few lines to a whole field) on the first field after the cut. |
| C11 | NTSC Cross-Colour | Cross-Color NTSC |
|
Luminance information around 3.58 MHz is incorrectly interpreted as colour (Coloured MoirŽ) |
| C12 | NTSC Subcarrier Phase Error | Erreur de Phase NTSC |
|
An phase error in transmission of subcarrier or burst will result in a rotation of colour space in NTSC |
| C13 | NTSC Cross-Luminance | Cross-Luminance NTSC |
|
Residual Colour Subcarrier in Luminance Signal, visible in uniform areas and vertical edges. |
| C14 | Loss of Luminance detail | Perte de détail en Luminance |
|
Due to the principles of composite coding, the luminance signal, in addition to cross-luminance, suffers a band-cut or low-pass filtering around the chrominance subcarrier frequency. |
| C15 | Loss of Chrominance Detail | Perte de détail en Chrominance |
|
Due to the principles of composite coding, in addition to cross-colour, the chrominance signal suffers a low-pass filtering below half the subcarrier frequency. |
| C16 | Short Colour Loss | Perte de Couleur Courte |
|
the "Colour-Killer" system triggers when the Colour Burst is not recognised on time. It can happen on several lines, or fields. |
| C17 | Temporary Colour Loss | Perte de couleur Temporaire |
|
the "Colour-Killer" system triggers when the Colour Burst is not recognised on time. It can last for a few seconds. |
| C18 | Complete Colour Loss | Perte de couleur permanente |
|
For some reason, the Colour Information is lost for a duration of more than a few seconds. |
| C19 | Undecoded composite stored as Component | Signal Composite Non décodé |
|
Due to an error in cabling, a failure in the colour system identification, or made on purpose. The Luminance signal is then identical to the original composite signal, but the colour burst may be lost. |
| C20 | SECAM incorrectly recorded as PAL, or vice-versa | SECAM enregistré en PAL, et réciproquement |
|
Due
to an error in cabling, or an incorrect composite standard selection on
VTRs, the colour bursts and identification sequences will be damaged or
lost, resulting in distorted signals. |
5. Video Problems related
to Video Transmission
|
|
|
|
|
|
| T1 | High Frequency Loss, Bandwidth limitation. | Perte de Bande Passante |
|
This is a general problem : when high frequencies are not filtered out by VTRs or other equipments, they are filtered out by colour coding and decoding processes. This problem is limited to horizontal frequencies. |
| T2 | Non-uniform frequency response | Réponse en fréquence non uniforme |
|
Incorrect cable compensation or pre-post accentuation filters may boost high frequencies, and different frequency bands may not be uniformly amplified. High frequency boosts are especially noticeable on faces. |
| T3 | Echoes | Echos |
|
Often due to incorrect daisy-chaining or poor radio reception, a distorted and delayed signal copy was added to the signal. Echoes may be positive or negative, Multiple echoes are very common. |
| T4 | Overshoot, Over-oscillations, | Suroscillations |
|
Overshoot with oscillations visible on abrupt transitions (vertical edges). Can be caused by rapid cut-off bandwidth limitation, echoes, excessive aperture correction. |
| T5 | Undershoot | Sous-oscillations |
|
>1 MHz band amplification error, abrupt transitions undergo an exponential distortion, (decay time up to 1µsec) making objects followed by a short trail. |
| T6 | Delay-frequency errors. (Non-uniform group delay ?) | Inégalité de temps de groupe |
|
Could cause overshoots |
| T7 | Streaking (Trail ?) | Traînage |
|
<1 MHz band amplification error, abrupt transitions undergo an exponential distortion, (decay time from 1µsec to several lines) making objects followed (on the right hand side) by a trail, and look flat and transparent. See L.E. Weaver, p 103. |
| T8 | Colour Streaking (Trail ?) | Traînage Couleur |
|
Same as above, but chrominance being more affected. |
| T9 | Luminance Chrominance Delay | Délai luminance Chrominance |
|
Due to different signal paths, the chrominance and luminance may not be delayed equally. This is especially visible on vertical edges. |
| T10 | Luminance Non-linearity | Non-Linearité en Luminance |
|
Non-linear relationship between the input luminance signal and the output. The eye is usually very tolerant to this problem, but simple corrective measures can be taken, once the desired correction is estimated. See L.E. Weaver, p 104 |
| T11 | Non-linearity transmission of composite signal | Non-Linearité dans la transmission d'un signal composite |
|
Same as above, but the different paths between the luminance and the chrominance signals usually generate saturation errors. See L.E. Weaver, p 104 |
| T12 | Differential Phase | Phase Différentielle |
|
Phase
of the modulated chrominance changes according to the amplitude of the luminance
signal. In NTSC, it appears as a variation of hues in the picture according
to brightness and saturation. In PAL, it is in part corrected with DL. See
L.E. Weaver, p 105
SECAM is usually not affected, but complex phenomena may appear along the transitions. |
| T13 | Differential Gain | Gain Différentiel |
|
Amplitude of the modulated Chrominance signal varies according to the amplitude of the Luminance signal. In PAL and NTSC, saturations are affected, especially yellows and reds. SECAM is not affected. See L.E. Weaver, p 105 |
| T14 | Luminance Chrominance Intermodulation | Intermodulation luminance Chrominance |
|
The luminance signal is affected by the amplitude of the modulated chrominance signal. |
| T15 | Signal Bounce (Low-Freq.) | Pompage Longue durée |
|
Oscillatory very low frequency (~1 sec) behaviour of signal level following abrupt changes of luminance level. Usually eliminated by black level clamp, but, at higher levels, can generate periodic signal losses. See L.E. Weaver, p 105 |
| T16 | DC Wander | Variation de la composante continue. |
|
Very low frequency additive noise whose consequence is a slow variation of the intensity of the signal. Especially noticeable when overloading results. |
| T17 | Crosstalk (Phantom Images ? ) | Diaphotie, Images Fantômes |
|
Superimposition of distorted images from other channels. Impairment is worse if the two signals are not synchronous. See L.E. Weaver, p 107. |
| T18 | 50-60 Hz Noise, Mains Hum | Ronflette Secteur (ronflement d'alimentation) |
|
Superimposed 50/60 Hz parasite signal from electromagnetic devices. Subjective impairment is worse when the parasite signal is not locked to the field frequency. See L.E. Weaver, p 60. |
| T19 | Periodic Noise (Interferences) | Bruit Périodique (Interférences) |
|
Superimposed quasi-periodic parasite signal from electromagnetic devices. Visibility and disturbance depends on the frequency and of the stability of its pattern. See L.E. Weaver, p 106 |
| T20 | Inverter Noise | Bruit d'inverseur |
|
Cited By L.E. Weaver, Breakthrough of inverter switching pulses, narrow bandwidth random noise. |
| T21 | Sound in picture | Son dans l’image |
|
Superimposed parasite signal from sound |
| T22 | Quantification Noise | Bruit de Quantification |
|
May
be seen when images are converted to low word size digital media. Further
degradation if these images are converted back to analogue, and digitised
again, or if successive operations during post-production results in several
rounding stages.
Early digital TBCs (3/4 Inch external TBCs) used internal 6 or 7 bits formats. D1's and many digital equipment truncate to 8 bits. |
| T23 | Blocking due to digital compression | Blocking du a la compression |
|
Block effect due to digital compression |
| T24 | Fading | Evanouissements |
|
Generally appears on terrestrial Hertzian transmissions : SNR decreases during small periods (seconds) |
|
|
|
|
|
|
| S1 | Colour convergence defects Misconvergence | Défauts de convergence couleur |
|
RGB images are not coincident. (rotation, focus, concentration, h&v linearity, h&v amplitude, h&v position) |
| S2 | Excessive Aperture Correction | Correction de contour excessive |
|
Early aperture correctors were monodirectionnal. Recent ones are bi-directional. Too much aperture correction leads to a hi-frequency boosting (cf. Overshoot). |
| S3 | Poor concentration (Blur) | Défaut de concentration (Flou) |
|
An incorrect focusing of the beam in the pickup tubes will make images blurred. |
| S4 | Poor Focus Blur | Flou de Mise au point |
|
May be due to an incorrect lens setting. But in common video broadcast documents, focus blur is made on purpose. |
| S5 | Motion Blur | Flou de Bougé |
|
Due to a long exposition time, motion blur is usually beneficial to the quality, but in some cases (fast panoramic movements), it may be inconvenient. |
| S6 | Blooming | Eblouissement |
|
Saturation of pickup devices together with a dispersion of charges around the saturated point, usually caused by a spotlight in direct view of the camera. (limited by "Automatic Beam Optimizer" for pickup tubes). |
| S7 | Comet | Comète, Queue de Comète |
|
Memory effect associated with blooming and motion. |
| S8 | Lag | Rémanence |
|
Memory effect associated with dark areas and motion |
| S9 | Flare | Halo (Flare?) |
|
Variation of black level dependent on global illumination, due to multiple diffusions and reflections inside the lens and pickup cavity. This variation is different for each colour channel. |
| S10 | CCD Halo | Halo CCD |
|
Halo around light sources, due to diffusion of light inside the CCD surface. Different on each colour channel (Red channel is usually affected first). |
| S11 | Smearing | Smearing |
|
Saturation of CCDs resulting in charges flowing between adjacent pixels or registers. The consequence is a saturation of one colour channel extending along rows (Red channel is usually affected first). |
| S12 | CCD Aliasing | Aliasing CCD |
|
Aliasing between high-frequencies of images and the CCD pixel pattern. Especially visible when the focusing is very precise, and when the useful area of the CCD pixels is lower than 80%. |
| S13 | Cushion, barrel ? | Coussin, Barillet |
|
Deformation of images due to radial optical distortion in lenses. Sometimes corrected in pickup tubes cameras, not corrected in CCD cameras (but the lenses have been improved accordingly) . |
| S14 | Dynamic Shearing | Cisaillement Dynamique |
|
In pickup tubes cameras, the pickup of the images is continuous, i.e. the upper part of a field is sampled before the lower part. Usually not perceptible , but in some cases, such as very fast motion of objects or camera, the shape distortion generated may become visible. |
| S15 | Over-Exposition | Sur-Exposition |
|
Sometimes made on purpose, may become problematic if the SNR is low |
| S16 | Under-Exposition | Sous-Exposition |
|
Sometimes made on purpose, may become problematic if the SNR is low |
| S17 | Tube noise | Bruit de Tube |
|
Tube cameras exhibit a characteristic noise pattern which is particularly noticeable in low light. |
| S18 | Scan Ringing | ? |
|
Aliasing of left hand edge of frame. Appears as vertical stripes down picture. |
| S19 | Lens
Cut-off
|
Vignettage |
|
Darker picture corners, due to lenses. |
| English Name | French Name |
|
|
||||
| P1 | Staircase Edges | Crénelage |
|
Staircase
edges : sometimes visible on computer-generated images and early chroma-keys
|
|||
| P2 | Early Chroma-key artefacts | Défauts de Chroma-key |
|
Early chroma-key systems generated blue, crawling, and hard-shaped silhouettes, which are now less acceptable. | |||
| P3 | Lost Texture | Perte de texture |
|
Some noise reduction devices making use of median filters suppress texture information | |||
| P4 | Dirty Window (video) | Effet de Fenêtre Sale (vidéo) |
|
Abusive frame-averaging Noise-Filtering makes often noise look like affixed on a window through which the document is seen. | |||
| P5 | Editing irrespectful of PAL colour sequence | Non respect de la séquence couleur en PAL |
|
Edition of PAL documents irrespectful of the colour sequence will often be compensated by VTRs by a horizontal shift of images of half a subcarrier period. This is mainly noticeable when the edition is made from two excerpts from the same shot. | |||
| P6 | Crystallisation | "Cristallisation" |
|
This phenomenon appears when analogue video passes successively through varied digital equipment (like external TBCs) with multiple A-D and D-conversions, or when excessive noise reduction is applied. | |||
| P7 | Multigenerations | Multigénérations |
|
Several generation of processing on analogue devices usually lower the SNR, and attenuate the high frequencies. | |||
| P8 | Colour Grading Timing Errors | Décalage Temporel des Corrections Colorimétriques |
|
Early video or film colour grading systems were not temporally accurate, for instance these systems start the correction of a shot two images before or after the beginning of the shot. This results in a few frames with different and incorrect colour grading before or after cuts. Restoring affected programmes consumes considerable time to adjust the grading of the affected images. | |||
8. Other Video Problems related to VTRs, Transmission,
and Shooting
|
|
|
|
|
|
| O1 | Continuous Random Noise | Bruit |
|
Very large number of possible causes, with different spectral distributions : approx. flat for amplifiers, pickup tubes, CCDs, triangular for FM demodulated signals. In practice, most continuous random noises lie between the two. |
| O2 | Impulsive Noise | Bruit Impulsif |
|
Erratic spikes or pulses of different origins, generally they impair seriously the picture. Narrow and high amplitude pulses are very frequent. |
| O3 | Continuous Random Chrominance Noise | Bruit Continu de Chrominance |
|
When decoding noisy composite signals, the possibly high level of high frequency noise is translated to low frequencies. The demodulated chrominance signal can hence be considerably more noisy than luminance. |
| O4 | Line Jitter | Jitter Ligne |
|
Random varying shifts of the lines of the picture, various origins, among others hum and TBC failures. |
| O5 | Full-Field Drop-out | Drop-out Trame |
|
Many problems may result in a full field being completely distorted. |
| O6 | Colour Bearding (Colour Flames ?) | Flammèches couleur |
|
High-intensity areas show coloured tails. |
| O7 | VTR saturation | Saturation VTR |
|
Visible on recording as black "strikes" when the Rec. level of the RF is too misaligned or when recording was made with used heads. Known at least on Betacam, betacam SP, 3/4 U-Matic and BVU. |
| O8 | Off Locks due to sync disturbance. | Décrochements à l'enregistrement |
|
Various picture and sound disturbances which are introduced at the time of recording and hence become "embedded" in the signal |
9. Problems Related to
Standard Conversion
|
|
|
|
|
|
| O9 | 819 lines Moiré | Moiré 819 Lignes |
|
Old scan-conversions from 819 lines to 625 lines using a video monitor and a camera, will exhibit moirŽ patterns similar to kinescopes ones |
| O10 | 525-625 Transcoded | Transcodage 525-625 |
|
NTSC Transcoded footage usually exhibit serious artefacts of very different types : jaggies, blur, tooth-comb moving edges, poor colour, 5-10 Hz judder |
| O11 | 625-525 Transcoded | Transcodage 625-525 |
|
|
| O12 | 4.43 to SECAM Moiré | Moiré 4.43 vers SECAM |
|
Aliasing between the 4.43 Mhz colour subcarrier and SECAM subcarrier. Happens on documents edited in PAL and broadcast in SECAM. |
| O13 | Other Colour Transcoding Problems | Autres Problèmes de Transcodage Couleur |
|
10. Other Video-related
Problems
|
|
|
|
|
|
| O14 | Saturations (Crushing ?) | Saturations |
|
When the video signals exceeds the limits of one of the components of the chain it passes trough, it is crushed, with sometimes numerous non-linear distortions, such as strikes. saturated areas are generally not recoverable. |
| O15 | Under-Black ? | Infra-Noirs |
|
Forbidden luminance levels under black level over the maximum allowed range. |
| O16 | Super-White ? | Sur-Niveau |
|
Forbidden luminance levels under black level generated by abrupt transitions such as titles from a character generator or computer-generated images. Usually not noticeable, but can generate off-locks later in he transmission chain. |
| O17 | Stroboscopic Effect | Effet Stroboscopique |
|
Wheels running backwards ... |
| O18 | Interlacement-related aliasing | Aliasing dû a l'entrelacement |
|
Flicking horizontal edges, moving staircase on diagonals, flicker and moirŽ on horizontal stripes |
|
|
|
|
|
|
| O19 | Several Frames damaged beyond repair | Plusieurs Images irrécupérables |
|
Several Frames damaged beyond repair, due to mechanical tape Problems, damaged film, film breaks, off-locks... |
| 020 | Full-Frame impairment | Image en erreur |
|
One frame damaged beyond repair, due to mechanical tape Problems, damaged film, film breaks, off-locks... |
| 021 | Framing Errors | Erreur de position Image, Errreur de cadrage. |
|
The sides of the picture may not be at the expected position : a black bar will be visible on top, bottom, right, or left, or even inside the picture. |
| 022 | Framing unsteadiness | Instabilité Image |
|
In addition to Film Unsteadiness, unsteadiness may also be caused by unsteady video cameras, or mains hum. |
| O23 | Saturation | Saturation |
|
Any kind of saturation. |
| O24 | Temporal Aliasing | Aliasing Temporel |
|
A
special case of temporal Aliasing is very often seen on film-generated footage
transferred to video : as the video 625 lines frequency is twice the film
frequency, fast movements may look jaggy.
This is usually not noticed by most viewers.. |
|
|
|
|
|
|
| A1 | Noise | Bruit, Souffle |
|
An unwanted part of a processed signal that is not related to the original signal Audio noise usually sounds like a continuous hiss. |
| A2 | Sporadic noise | Bruit transitoire |
|
Same as above, but with fluctuations in the noise level. |
| A3 | Synchronisation problems Mis-Sync | Défaut de synchronisation |
|
Sound
is not precisely in phase with the picture content. It occurs mainly on
the original sound track (bad post synchronisation, bad transfers during
the original mix process, film breaks repair, poor processing during telecine
transfer…)
Poor " lip sync "
is noticeable at -1 or-+1/2 frames. Sync of audio can also fluctuate along
the programme. |
| A4 | Interferences (Hum ?) | Fréquences parasites |
|
This
default is generally due to unwanted frequencies in the sound spectrum.
Some examples of interference are: Head wheel rotation frequency of VTRs and their harmonics, hum, buzz, rumble, noises generated by cameras players or recorders and electronic stages and captured by microphones.(for example mechanical noise generated by the mechanism of recorders of cameras and captures by microphones, rumble of LP turntables..) Generally can be removed
or dimmed by narrow filters. |
| A5 | Echo | Echo |
|
A
reverberation artefact so spread out that the reflected signal is perceived
as a distinct sound |
| A6 | Print-Through | Enregistrement entre spires |
|
Specific
case of echo due to accidental partial copy from one layer of magnetic tape
to another, often happening immediately after recording. Very often pre-echo
rather than post-echo, as for other types of echo. |
| A7 | Cross talk | Diaphonie |
|
The
leakage of a signal from one channel of a system to another. A system with
low cross talk has a good separation between channels |
| A8 | Phasiness | ?? |
|
On some stereo recordings, the left and right channels will appear to be in phase while the centre will appear slightly out of phase, attenuated and ill defined |
| A9 | Bandwidth losses | Pertes de bande passante |
|
Most
losses in bandwidth occur in the high frequencies of the spectrum Sound
has lower levels for trebles than lows
This is generally due to bad magnetic recordings or copies with mis aligned analogue recorders. Sometimes losses may occur in medium or low frequencies. |
| A10 | Level variations | Fluctuations de niveau |
|
Global
level may be too high or too low.
When the global level is too high, the sound may be saturate. When the overall level is too low, noise is more present and tiny sounds cannot be heard. Level may also fluctuate along the programme. This can occur when the tape is damaged (for example edges or magnetic coat or wounding tape) |
| A11 | Sound
Holes (Gaps ?)
Dropouts |
Trous de son, Pertes de son |
|
Lack
of sound on the audio tracks. This lack may have a very short duration or
last several frames.
It may occur on a damaged original tape or may correspond to a mute on digital recordings (it occurs when the concealment rate is too high.) Short sound holes may be considered as drop outs. A drop pout is the result of a coating defect or a dirt deposit on the tape creating a momentary discontinuity in the playback signal |
| A12 | Long Sound Holes (Gaps ?) | Trous de son, Pertes de son importantes |
|
Long
sound holes may last from several frames to seconds.
Often the lack of sound is due to a technical problem during the shooting or an undesired erasure of the recording. |
| A13 | Wow | Pleurage |
|
Wow
is a pseudo-periodic fluctuation of the sound caused by cyclic variations
of speed of the audio tape, causing a "taffy like" sound quality It occurs
mainly during recording or playback on analogue tape recorders. Wow is mainly
caused by misaligned capstan servo or damaged capstan and pinch roller.
Wow is a slow speed variation. |
| A14 | Flutter | Scintillement |
|
As Wow, but Flutter is a rapid speed variation. |
| A15 | Saturation | Saturation |
|
A
magnetic recording term used to describe a condition when an audio tape
becomes fully magnetised and an increase in signal input does not produce
a corresponding increase in recorded levels.
It can also occur when a tape head cannot carry all the signals to handle. Replayed sound is affected by distortion and part of information in the sound is lost. Saturation may also be caused by the distortion of a microphone cap. There is currently no way today to recover a good sound from a saturated sound. The only thing that can be done is to apply filters to improve a little the perception of the sound.
|
| A16 | Clipping | Ecrètage |
|
Specific
case of Saturation. In audio the result of an analogue signal being over
driven to the extend that its peak levels cannot be accommodated and are
"clipped" off the audible signal.
Typically it is the most audible electronic distortion occurring in small amplifiers
|
| A17 | Clicks and Pops, crackle | Clics et pops, grattements |
|
Impulsive noise (not continuous) may appear ounce (sporadic)or may appear all along the program like crackles of the optical sound on film |
| A18 | Azimut Error | Erreur d'azimut |
|
Azimut is the angle between the magnetic gap of the tape head and the direction of the travel ideally 90°.Azimut varies from one deck to another and impacts the response on high frequencies |
| A19 | Audio Vinegar Syndrome | Syndrome du Vinaigre (audio) |
|
Autocatalytic hydrolysis of acetate separate magnetic audio16mm film, and magnetic tracts of films, which results in different kings of audio impairments, up to the point the source material is no longer playable. |
Some of the terms used in the previous lists are documented hereafter.
| CCD | Charge Coupled Device : current semi-conductor imaging technology in cameras. |
| CRT | Cathode Ray Tube. Still in use technology for television displays. |
| DOC | Drop Out Compensator : the video tape technology is prone to short losses of signal due to dust or tape problems (drop-outs). To make these faults less visible to the eye, the drop-out concealment devices replaced the lost signal by the signal from the previous line, or the 2nd previous line (SECAM). |
| Kinescope or film recordings | Early TV programs could not be recorded on VTRs, because these devices did not exist, or because of the price of videotapes. The only way of keeping an archive of the broadcast programmes was to record them on film, using a CRT and a film camera. |
| Reversal | Positive film, used directly for news gathering, or TV productions |
| Segmented VTRs | A VTR format is segmented when one
full field is not recorded on the tape by one single head. Instead, the
recording is made using several heads on different tracks, each track storing
a group of lines.
Two VTR formats that have been in wide use : 2 Inch (15 to 16 lines per track), and 1 Inch B (approx. 56 lines per track). |
| TBC | Time-Base Corrector : devices used to correct time-shifts. In the case of the VTRs, as is very difficult to read the tracks synchronously, most of the VTRs use TBCs to compensate for time-shifts. |
| TK, Telecine | Telecinema : Device used for obtaining a real-time video output from a film programme. |
| VR2000 | A 2 Inch VTR model. |
| VTR | Video Tape Recorder |