Reference books

mycroft

I'm looking for a good post production TV technical reference book.

Nothing too detailed more of a primer and jaragon buster. Y'know detailing different types of t/c, composite v component.

Anyone have any ideas suggestions?

herbie747

AAC
Advanced Audio Coding will be the successor of AC3 audio. It is based on AC3 while adding a number of improvements in various areas. Currently player and hardware support for this upcoming audio format is still very limited.

Authentication
Before a movie can be played the player and the disc have to establish a secured communication line on which they can transfer the actual movie. Before they can establish that line they need to make sure that the right "person" is on the other side - this is done via several key exchanges, verifications, etc.

AC3
Initially known as Audio Coding 3, AC3 is a synonym for Dolby Digital these days. Dolby Digital is an advanced audio compression technology allowing to encode up to 6 separate channels at bitrates up to 448kbit/s. For more information please check out the Dolby website.

ASF
Advanced Streaming Format. Microsoft's answer to Real Media and streaming media in general.

AVI
Audio Video Interleave. The video format most commonly used on Windows PC's. It defines how video and audio are attached to each other, without specifying a codec.

BUP file
A bup file is a Back UP file of an IFO file. These files are commonly found on DVDs.

Cell (ID)
A cell is the smallest video unit on a DVD. Normally used to contain a chapter it can also be used to contain a smaller unit in case of multiangles or seamless branching titles.

Codec
COder/DECoder. A codec is a piece of software that allows you to encode something - usually audio or video - to a specific format and can decode media encoded in this specific format again. Popular Codecs: MPEG1, MPEG2, MPEG-4, Indeo, etc. AVI, ASF, etc is not a codec but a format - that can be encoded using different codecs.

Container
A container is, like the name says, a construct to contain data - in this case video and audio data and possibly subtitles and navigational information. For instance, you would like to put a soundless video stream and the audio track together in one file. To do that you need a container format. Examples of container formats are: AVI, ASF, OGM, Quicktime, VOB and MPG.

CSS
Content Scrambling System. Prioprietary scrambling system for video DVDs. Designed to stop people from making copies of DVDs, most commercial DVDs are encrypted using CSS. During playback, DVDs are then decrypted on the fly. Only parts of the DVD are encrypted (for instance all IFO and BUP files are not encrypted, and VIDEO_TS.VOB often isn't encrypted either) and the encryption scheme is rather weak and was quickly defeated. If you want to know what CSS does, insert a DVD video disc into your PC, start playing the disc using a software DVD player, then close the player. Now copy a 0.99GB VOB file from the disc to your harddisk and try to play back that VOB file in your software DVD player. You'll see a lot of funny colored blocks all over the picture making the movie unwatchable. But you'll also see parts of the movie (the parts that are not encrypted).

DAR
DAR stands for Display Aspect Ratio and indicates the dimension of a screen. Most PC screens have a DAR of 4:3, meaning that the horizontal size is 4/3 as large as the vertical size. For TVs we have a lot of old 4:3 displays and more and more 16:9 displays. As you can guess from the numbers 16:9 displays are broader than 4:3 displays having the same diagonal size. 16:9 screens are more suited to display Hollywood movies which are usually shot with an aspect ratio of 1:2.35 or 1:1.85 (meaning that the horizontal size of the picture is 1.85 times as wide as the vertical size).

Deinterlace
The process of restoring a progressive video stream out of an interlaced one is called deinterlacing.

Demultiplexing
The opposite of multiplexing. In this process a combined audio/video stream will be separated into the number of streams it consists of (a video stream, at least one audio stream and a navigational stream). Every VOB encoder demultiplexes the VOB files before encoding (FlaskMpeg, mpeg2avi, dvd2mpg, ReMpeg2) and every DVD player does the same (audio and video are being treated by different circuits, or decoded by different filters on a PC).

Descrambling
DVDs are usually CSS scrambled - imagine you decide to give a number to each letter, starting with 1 for a, etc. A sentence would become a couple of digits - that's what we call scrambled. Of course CSS is much better than that but it's still quite easy to crack. Descrambling means reversing the scrambling process, rendering our digits to a sentence again, or making our movie playable again - you can try to copy a movie to your hard disk when you've authenticated your DVD drive and play it, you'll get a garbled picture because it's still scrambled. Common CSS descramblers either use a pool of known descrambling keys (DeCSS or DODSrip - they contain a large number of keys but not all of them) or try to derive the key by a cryptographic attack (VobDec - that's why it works on most disc since it's not dependent on a pool of discs).

Digital Video
Digital video is usually compressed since it'd take Terabytes - thousands of Gigabytes or for the mathematicians among you : 10^12 Bytes) to store a movie uncompressed. Since standard lossless compression is insufficient for video, the video codecs have to get rid of unimportant information - stuff the human eye won't see or is unlikely to see. Since that is still not enough, modern compression algorithms use keyframes, I and P frames in order to save space.

DivX
Not to be confused with the now - thank God - obsolete DIVX (DIgital Video eXpress) system introduced by Circuit City '98. There are 2 flavors of DivX today: DivX is the name of the hacked Microsoft MPEG4 codecs (Windows Media Video V3). Those codecs were developed by Microsoft for use in its proprietary Windows Media architecture and initially supported encoding AVIs and ASFs but all non-beta versions included an AVI lock, making it impossible to use them to encode to the AVI format - and only a few tools support ASF today. What the makers of DivX did is remove that AVI lock making it possible to encode to AVI again, and changed the name to DivX video in order to prevent confusion of codecs, since it's possible to have both the unhacked and hacked codecs on the same computer if you use the Windows Media Encoder. The latest releases of DivX also include a hacked Windows Media Audio Codec called DivX audio - the hack of that codec is not perfect yet and its use is limited for higher bitrates. This codec is also known as DivX3.
The other DivX is a brand-new MPEG-4 video codec developed by DivXNetworks. It offers much advanced encoding controls and 2 pass encoding. Furthermore the codec can play the old DivX (DivX3) movies. The codec is commonly called DivX4.

DIVX
DIVX was basically DVD stripped of all its extra features - no extras, making-ofs, trailers, multi-language, widescreen picture - introduced by Circuit City and a bunch of greedy Hollywood lawyers in order to completely control movie distribution up to the end user again and to gain complete control over movie playback in your home. DIVX was pay-per-view and a "DIVX-enhanced" DVD player had to be hooked up to your phone line in order to dial in to the DIVX central computer to register when you play a disc and to bill your credit card. A movie was $4.50 - including a 48 hour viewing period - and $2.50 for additional viewing periods. DIVX was stopped after less than a year in operation due to lack of titles (Warner, Sony, New Line, and all the other smaller studios flatly refused to release any titles to the format - THANKS GUYS!!!) and the very negative press it got, mainly from DVD sites on the Internet which later made it into serious printed publications and TV news.
As DIVX uses triple DES encryption it's pretty safe against cryptographic attacks and unless you can crack that encryption there's no way to rip these discs. In other words your DIVX discs will probably remain coasters forever.

DRC
Dynamic Range Compression. AC3 Tracks contain a much larger dynamic range that most audio equipment can handle, therefore most standalone and software DVD player will compress the dynamic range somewhat, according to the actual dynamic range. In layman terms the volume will be augmented dynamically, e.g. explosions won't become louder or only a bit louder, whereas in normal dialogues the volume will be augmented quite a bit. Since your player will do the same this is the way to go to have augmented volume.

Elementary Stream (ES)
An elementary stream is a single (video or audio) stream without container. For instance a basic MPEG-2 video stream (.m2v or .mpv) is an MPEG-2 ES, and on the audio side we have AC3, MP2, etc files that are ES. Most DVD authoring program require ES as input.

herbie747

Field
Interlaced video streams contain fields rather than frames.

FourCC
FourCC stands for four character code and is a code that uniquely identifies a video data stream format. A movie player will look up the FourCC code then look for the codec associated to the FourCC code in order to play a certain video stream. A few examples: DIV3 = DivX Low-Motion, DIV4 = DivX Fast-Motion, DIVX = DivX4.

Frame
The basic source of a movie. One frame represents one image. A movie usually runs at 25 frames per seconds, so it has 25 different images per second. Imagine 25 images with a bird on it, in the first image it's on the left, then it gradually movies somewhat to the right. On the 25th frame the bird is on the right end of the image. Imagine these 25 images being played in sequence fast enough and it looks like a bird would fly from left to right to the human eye.

I and P Frames
Frame describing only the differences to the frame before (this is less than accurate but I think you'll get the picture that way). Say we have a keyframe with a bird before a cloudy sky. Then we can use I frames which say something like this : move the bird an inch to the left and one inch to the bottom.

iDCT
The video information inside MPEG files is stored in the frequency domain rather than in the spatial domain (the images we see). That way, the information gets compacted and that compactation can be used to compress (reduce) the amount of information you have to send over the transmission channel. MPEG uses the DCT (Discrete Cosine Transform) to translate spatial information into frequency information.
To bring back the spatial information from the MPEG stream you have to apply the iDCT, that is, the Inverse Discrete Cosine Transform, that undoes the DCT that was used during encoding.
DCT and iDCT are basically the same as DFT (discrete fourier transforms) but the results are integers rather than complex reals you get in i/DFT. For more info please refer to a university-level book about DSP, communication systems or similar.

IFO file
InFOrmation file commonly found on DVDs. Such files contain navigational information for your DVD player. For more info refer to the DVD structure article.

Interlaced
Interlaced is a video storage mode. An interlaced video stream doesn't contain frames (pictures as we know them) but fields with each field containing half of the lines of one frame (all even or all odd lines).

Interleaving
Imagine gluing together the audio and the video track at defined points, that's about it. The player will recognize the interleave points and make sure that both audio and video are played in a manner that the "glued" points match through the movie.
A more detailed explanation: Imagine we have 10 seconds of video and 10 seconds of audio. Let each second of video be represented by a V and each second of audio by an A. If you have an interleaving setting of 10 seconds the file on the disc will look like this: VVVVVVVVVVAAAAAAAAAA. Now if you have an interleaving setting of 1 second instead here's what you get: VAVAVAVAVAVAVAVAVAVA.

Inverse Telecine
The inverse of Telecine. This process is performed to extract the original 24fps of a 29.97fps source.

Keyframe
A complete frame but heavily compressed.

m1v/m2v
These two terms are used as extensions for MPEG-1 respectively MPEG-2 video data (video only, without any audio).

miniDVD
miniDVD is basically a DVD on a CD. A miniDVD can contain bitrates up to 10mbit/s (audio and video combined). Video is MPEG2 of course - preferably VBR and audio can be MPEG1 audio layer 2, raw uncompressed PCM or AC3. Video quality can be up to an actual DVD level if you accept the limited playtime of a CD. You can create DVD-like menus as well. The drawback is miniDVDs will only play on PCs and on a very limited number of standalone players.

MPEG
MPEG means Motion Picture Expert Group and it's THE resource for video formats in general. This group defines standards in digital video, among it the MPEG1 standard (used in Video CDs), the MPEG2 standard (used on DVDs and SVCDs), the MPEG4 standard, and several audio standards - among them MP3 and AAC. Files containing MPEG-1 or MPEG-2 video often use either the .mpg or .mpeg extension.

MPEG4
MPEG4 is pretty much a collection of standards defined by the MPEG Group, and it should become the next standard in digital video (mainly for picture phones, streaming media on the Internet and more). MPEG4 allows the use of different encoding methods, for instance a keyframe can be encoded using ICT or Wavelets resulting in different output qualities.

MPG
MPG can be either an abbreviation for MPEG or is used as a file extension for MPEG-1 and MPEG-2 video data. It is a container to contain MPEG-1/2 video stream and MPEG1 layer 2 audio (aka mp2 files). MPG containers are also refered to as program streams (PS).

MM4
Multiple MPEG 4: A combination of different bitrate encoded files. For instance you could take a 2000kbit/s encode, a 910kbit/s encode and combine the files together, use the lower bitrate file and replace scenes where the quality gets too bad due to a lot of action with the parts taken from the 2000kbit/s one. It also includes the use of both DivX codecs: You can combine DivX low motion and DivX high motion files (and once again you can choose different bitrates).



Multipass encoding
Before you read on: Currently true multipass encoding is available only for WM8 and MPEG-2 (SVCD & miniDVD). M4C is not true multipass encoding (and read the M4C guide to find out what it is and how it works). An encoder that supports multipass will in a first pass analyze the video stream to be encoded and write down a log about everything it encounters. Let's assume that we have a short clip which starts out in a dialog scene where we have few cuts and the camera stays static. Then it leads over to a karate fight with lots of fast cuts and a lot of action (people flying through the air, kicking, punching, etc). In regular CBR encoding every second gets more or less the same bitrate (it's hard to stay 100% CBR but that's a detail) whereas in multipass VBR mode the encoder will use the bitrate according to its knowledge about the video stream, that is the dialog part gets of the available bitrate and the fighting part gets more bitrate. The more passes the more refined will the bitrate distribution be. In single pass VBR the encoder has to base his decisions on where to use how much bitrate solely on the knowledge of the stuff it previously has encoded.

Multiplexing
Usually video and audio are encoded separately. Then you have to join both of them to make a movie that you can play (you can of course play audio and video separately in two players but to get synch would be rather hard). During multiplexing the audio and video track are combined to one audio/video stream. The audio and video stream will be like woven together and navigational information will be added so that the player can example fast forward/backward and still retain synch audio/video.

Normalizing
Normalizing consists of finding the volume peak of a soundtrack, then increase the rest of the soundtrack to the same level. This is usually done at WAV level, that is all variables are integers so the precision is not as good as in DRC. Hence DRC is to be preferred over Normalizing.

OGM
OGM stands for OGg Media which is the name of the Ogg container implementation by Tobias Waldvogel. OGM can be used as an alternative to the AVI container and it can contain Ogg Vorbis, MP3 and AC3 audio, all kinds of video formats, chapter information and subtitles.

PGC
PGC stands for ProGram Chain. It is a term often used in DVD authoring and represents basically one concurrent playback item. For instance the main movie has its own PGC, each trailer on a DVD usually has its own PGC. The studio logo that comes up when you enter the disc has its own PGC, etc. Both SmartRipper in Movie mode and DVD Decrypter in IFO mode will show you all PGCs a DVD has. For more info about PGCs visit MPUCoder's DVD information site.

Program Stream (PS)
A program stream is a combination of elementary video and audio streams (ES). An MPEG-1 program stream contains MPEG-1 video and MPEG1 layer 2 audio (mp2) whereas an MPEG-2 program stream contains MPEG-2 video and MPEG1 layer 2 audio (mp2).

Progressive
The opposite of interlaced. A video stream consisting of only full frames is considered progressive. More about progressive and interlaced can be found in video basics.

Quantizer
To truly understand this term you'd have to take a course in signal processing. In laymen's terms it means compression factor. The higher this value the more compressed an image is (and therefore a high quantizer means low quality picture and small size whereas low quantizers means high quality picture and larger size).

herbie747

rff/tff flags
RFF means repeat first frame, it's a technique used to make the necessary 29.97 frames per second out of a 24 frames per second source - the movie like it was recorded with a traditional movie camera used by Hollywood. The rff flag tells the player to repeat one field of the video stream. Tff means top field first and is also used to perform a telecine to make a 24fps movie into 29.97fps. You can read more about film to video transfer in Robshot's article on synch.

Ripping
Lots of confusion about that one. Basically ripping means copying a DVD movie to your hard disk. This includes the authentication process for the DVD Drive (try to copy a file off a DVD and you'll get a message that this operation is not supported if your drive hasn't been authenticated) and the actual CSS Descrambling. CSS (Content Scrambling System) is a copy protection scheme designed to prevent unauthorized copying of DVD movies, although many argue that it was also designed to control where DVD movies can be played since without a CSS license you essentially have to crack the encryption to play a DVD movie - and I quite agree with that. The term "ripping" is also often used (even on this site) to describe the whole process of descrambling a DVD, then convert the audio and video into another - lesser - format.

SBC
Smart Bitrate Control. A new kind of DivX encoder called Nandub can modify many internal codec parameters on the fly during compression, giving you better quality and a lot more control over the encoding session. More information can be found in the SBC guide in the DivX guides section.

Streamlist
A streamlist is an ASCII test file that contains the pathnames and filenames of your VOB files, one file on every line. Here's a small example:
e:\video_ts\vts_01_1.vob
e:\video_ts\vts_01_2.vob
e:\video_ts\vts_01_3.vob
e:\video_ts\vts_01_4.vob
e:\video_ts\vts_01_5.vob
Save this file as streamlist.txt, or streamlist.lst. Make sure that you save it as unformatted ASCII text, I suggest you use notepad to edit your streamlists, since notepad won't save in another format. Mpeg2avi needs the streamlist to have the extension lst, whereas you're free to chose any other extension for other programs that use a streamlist. However the GUI is kind of limited in its choice for input files/extensions, therefore you might have to rename your streamlist, if it doesn't show up in the file selection dialog.

SVCD
Super Video CD, mainly used in Asian countries. Uses MPEG2 Video and therefore much better image quality - LaserDisc-like and also offers High quality surround sound. Furthermore it can take advantage of hardware decoders and there are players for many operating systems. However there are only a few - mostly Asian made low-end - DVD player that can play SVCD and other than SVCD player which are not sold in the US and Europe you can only play SVCDs on your computer. Video is MPEG2 at up to 2600kbit/s and audio MPEG1 audio layer 2 up to 224kbit/s. MPEG2 multichannel audio is also possible, but most players will only output 2 channels and those that will pass through 5.1 audio still require that you have an mpeg2 multichannel capable receiver.

Telecine
A process to bring a 24fps source (usually a movie is shot at that speed) to 29.97fps or 29.97x2 interlaced fields per second. Please read this article on how the conversion actually works.

VBR
Variable BitRate. It's possible to encode both audio and video in VBR mode, which won't use the same bitrate for the whole file (as in CBR = Constant BitRate) but rather more complicated parts of the video/audio will receive more bitrate in order to look/sound better and less demanding parts will get less bitrate.

VCD
Video CD, works on many DVD players, there are software players on almost every operating systems, doesn't need a fast computer but the image is VHS-like. Video is MPEG1 at 1150kbit/s and audio MPEG1 audio layer 2 at 224kbit/s.

VKI
Variable Keyframe Interval. Basically that means that keyframes will not be inserted in regular intervals as in the regular DivX codecs but where they are needed. There are 2 ways of VKI: The first is that the encoder analyzes the compressed frame, compares it against the original and re-encodes the frame again as a keyframe if the quality difference is higher than a set threshold. This way of encoding is only possible with a certain special application: M4C. There's a command line based version and a plugin for AviUtl available (the latter is described in detail in the AviUtl guide). If you set the threshold too high you'll end up with a lot of keyframes. Then there's the 2nd way which is basically keyframe insertion at scene changes. In order to do that the encoding program or the codec will detect when there's a cut (as it's called by movie makers) occur and make the first frame of the new scene a keyframe. This can be achieved by using mpeg2aviAr (part of AviRevolution 2.1), m4c or by installing the DivX VKI codec. If you use the latter you don't have to worry about the encoder... every program that can encode to DivX will then result in files that has keyframes at scene changes. VKI, when properly used (that applies to the first way), can help you increase quality and reduce the amount of keyframes, which may lead to higher quality again because especially at lower bitrates too many keyframes will give you a worse quality.


VM2
Short version of VKI + MM4 + VBR MP3

VOB ID
VOB IDs are used to internally group cells in a PGC on a DVD.

VOB Files
All DVD movies are stored in so-called VOB files. Vob files usually contain multiplexed Dolby Digital Audio and MPEG2 video. Vob Files are called as follows: vts_XX_y.vob where XX represents the title and Y the part of the title. There can be 99 titles and 10 parts, although VTS_XX_0.VOB does never contain any video, usually just menu or navigational information. There's 2 ways to find out which files contain the main movie: First: Play the movie in any DVD player and watch the LED on a standalone or the status window on a software player. Second: The main movie is the largest number of consecutively numbered VOB files. For instance it's vts_05_1.vob, vts_05_2.vob.... vts_05_8.vob (I haven't seen movies with 9 partial files but that doesn't mean that they don't exist).

VTS
VTS stands for Video TitleSet and means a set of consecutively named VOB files with the corresponding IFO and BUP files. For instance VTS2 would be VTS_02_0.VOB (containing the menus), VTS_02_1.VOB, VTS_02_2.VOB, etc, VTS_02_0.IFO and VTS_02_0.BUP. VTS are used to group video stuff together that belongs together. For instance one VTS is usually used for the main movie (sometimes including the trailer and some studio logos), other VTS are used for extras.

Wavelets
Wavelets are an alternative basis space. There are infinitely many wavelet bases (Daubechies, Haar, Mexican Hat, "Spline", Zebra, etc), but their primary feature is that they are localized. Fourier basis functions span all space (from negative to positive infinity). Wavelets are basically individual pulses of waves (at various positions and scales).
Their value in compression stems from factors like the grouping which generally shows that a good 90% of the data is modelled by the low-pass filters, with the high-pass filters generally showing very small values that are mostly details. (of course, this is not true if the source is noisy in the first place). For images, the greatest value comes from localization of the basis, which means that we can model discontinuities (e.g. edges) VERY well with wavelets. You will NOT get those weird JPEG halos if you use wavelets.

Windows Media
Microsoft's proprietary architecture for audio and video on the PC. It's based on a collection of codecs which can be used by the WindowsMedia Player to play files encoded in any supported format. The current release of the Windows Media Player is version 7.0. WindowsMedia 7.0 offers a new set of codecs, among them a fully ISO compliant MPEG4 codec (called MS Windows Video V1), an improved MPEG-4 codec called MS Video V7 (although I did not notice any improvement compared with MS Windows Video V3 on which DivX is based), an encoder that supports Deinterlacing and Inverse Telecine.

XCD
eXtended CD is an upcoming CD format which allows your CDs to be written in mode2 form 2 mode which basically means that it contains less error correction codes thus allowing you to store more data onto a single CD. XCD allows you to store 800 MB of data on a regular 700 MB CD. But as there's less error correction on the discs, XCDs are not very scratch resistant and it's suggested that you only put data that has additional error correction in the container (for instance OGM) on such CDs. For more info refer to the XCD guide.

XviD
XviD is a word play, read it the reverse way and you might find a familiar term. XviD is an open source MPEG-4 codec which depending on whom you're asking yields even better quality than the best DivX codec. The XviD homepage can be found here.

mycroft

Not what I was looking for, I'm looking for actual explanations rather than definitions, and I'm looking for a book I can thumb through.....

herbie747

Damn.

I spent 4 hours typing that out.....



I have this book, and it's amazing - I think it's what you're after:
http://www.amazon.com/exec/obidos/tg/detail/-/0534559891/103-7875697-9865450?v=glance

mycroft

Sorry for giving you carpul tunnel syndrome...

Looks good but I'm only interested in post video stuff, something that details everything from t/c to hd....