|Also available in Acrobat format|
Multimedia in the Teaching Space
2.4 THE COMPRESSION STANDARDS.
This section briefly discusses the principle compression standards. A more detailed discussion takes place in section 2.7.
2.4.1. Video Standards
The acronym MPEG stands for the Moving Picture Experts Group, which meets to generate standards for digital video (sequences of images in time), and audio, and for the development and defining of the conditions for compression algorithms. The standard defines a compressed bit-stream, which implicitly defines the function of the compressor. The compression algorithm, design by individual manufacturers must meet certain conditions laid down by the standard and this has led to the development of proprietary algorithms optimising the characteristics of a particular manufacturer's equipment. These proprietary algorithms have largely merged into the background and everyone produces systems that conform to the MPEG standard, and as a result compatibility between different systems has improved.
H.261 and MPEG-1:
H.261 is the standard used in most ISDN Videoconferencing systems and handles the bandwidth range form 64Kbps to 2Mbps. This range overlaps with the MPEG-1 standard which was developed for encoding video to be stored on CD-ROM. The common data rate for CD-ROM is 1.2Mbps which is higher than that used in Videoconferencing, i.e. 128 and 384Kbps. Visually the quality of video from the MPEG-1 and the H.261 standards are comparable when using the same bandwidth.
MPEG-1, MPEG-2 Motion-JPEG and MPEG-4:
Video digitisation uses the standards MPEG-1, MPEG-2, Motion-JPEG, MPEG-4 and near standards such as AVI. It is now possible to have both hardware and software encoder and decoders. This results in simpler hardware for terminal as it is not necessary to have special cards inserted in the computers to be able to handle these various standards. MPEG-1 equipment is cheaply available, but the other standards are more costly to implement. The user needs to appreciate that it is costly to encode material in the MPEG-2 standard, but the decoders are cheap because this standard has been developed with video-on-demand in mind. The Motion-JPEG requires considerable computing power because it is digitising each video frame in real-time, i.e. pictures are being encoded at 25 frames per second.
Fullscreen in real-time, is now possible, but it must be remembered that the encoding and display of information is a 2-stage process. In standards like MPEG-2 where the encoding is slower and more complicated it may be necessary to carry this out off-line
2.4.2 Audio Standards
These refer to the audio standards for the compression/decompression and transmission of audio signals. G.xxx classifications gives these standards.
There are 3 principal standards which are most commonly used, namely G.771, G.722, and G.728
G.711 A standard for compressing and decompressing audio (50 -3000Hz) into a 48, 56, or 64Kbps stream.
G.722 A standard for compressing and decompressing audio (50 - 7000Hz) into a 48, 56, or 64Kbps stream.
G.728 A standard for compressing and decompressing audio (50 - 3000 Hz) into a 16Kbps stream.
G.728 is particularly interesting as it uses on 16Kbps of the available bandwidth compared with 48Kbps in G.711 and G.722. This means that more of the bandwidth can be devoted to video and the picture quality is therefore more acceptable. The proportionate effect is greater at 128Kbps than at higher bandwidths.
This is all brought together in the Videoconferencing standard H.320 AND H.323 which will be dealt with in a separate section.
2.4.3. Still Picture Standards
This is a standardised image compression standard from the Joint Photographic Experts Group. This standard is designed to compress either a full-colour or grey-scale images of the real-world. It works well on photographs and artwork, but not so well on lettering, simple cartoons and line drawings. It only handles still images. It is a "lossy" standard in that the compressed image is not exactly the same as the original. This standard deliberately exploits characteristics of the human eye, in that it is more sensitive to changes in the brightness of colours than in changes of colour - i.e. it is design to produce compressed images to be looked at by humans.
An important aspect of JPEG is that the decoders can trade off decoding speed against image quality.
GIF can only store images with 8 bits per pixel, so its is suitable for inexpensive computer displays. As full colour hardware is becoming cheaper then GIF should be less in demand, but for certain types of image GIF is superior in image quality, and file size.
CD-ROM is a means of delivering multimedia to the desk-top, and into teaching space and is suitable for stand-alone systems. This medium is able to handle audio and video, and the disks can contain large amounts of data as well as. Technological advances mean that it is possible to write on CD-ROM using equipment little more expensive than the computer itself. Lap-top computers now have CD-ROM drives and small loudspeakers which make these system highly portable. This portability can be very valuable in teaching as the technology can be carried into the lecture theatre when it is required, which does away with the problems of keeping expensive computer hardware permanently in lecture theatres and seminar rooms. This is particularly convenient as many lecture theatres are not wired to local networks.
Browsing the World Wide Web (WWW) with Netscape and the other available web browsers, has given students the opportunity to work alone accessing very extensive sources of information. The problem now is not whether the student has access to information, but how the student can assess the quality of the information and cope with the volume of information they can so easily access. The WWW is here to stay and we are still learning new methods of using this technology, which is accessible from many desk-top computers. The CD-ROM holds selected information, very often by an expert in the field and with a mind to an academic course. The WWW is "indiscriminate" information which has been made available because someone considers it will be useful to someone else, and the use envisaged in many cases is not related to academic pursuits at all; e.g. suppliers wishing to give detailed information about their products and how they function. The relative roles of CD-ROM and the WWW will be interesting to watch in the near future.
The original single speed CD-ROM drive had transfer rates of 150 bytes per second, which corresponds with 1.2Mbps which is the speed of the MPEG-1 standard. Today's CD-ROM drives are much faster with transfer rates as high as 1.5Mbytes per second, i.e. 12Mbps. The throughput is not the only measure of playback speed in a random access system. The size of the cache will have some influence on the effective transfer rate, and the access time is measured by the seek time and the latency; the seek time measures how long it takes to find the appropriate track and the latency how long it takes for the start point to rotate under the read-head. A CD-ROM drive is a constant linear velocity device, which means the read head must speed up to read the out sector of the disk, and slow down to read the inner tracks.
CD-Recording systems (CD-R) have become cheap and common; they are invaluable because they provide the ability to prototype to see how a product runs at CD speeds, which is important with interactive systems, Also they give the ability to produce a one-off disk quickly.
The CD-formats that a user should expect CD-ROM software to support should include DOS or Windows compatible disks, HFS (Macintosh files structures), CD-audio and mixed modes, e.g. audio plus data. It is possible to create disk for Mac and PC systems, where the same dataset is to be read on Mac or PC as the directory structures are set up for each platform.
There are other formats such as CD-I (CD Interactive), and CD-ROM XA, an extended architecture which combines CD-data with interleaved compressed audio or video. Other include the Kodak Photo-CD and Video-CD. CD-R software should support multi-session writing which means the recording of the disk can be carried out putting down different tracks at different times.
Visual information is presented on a computer as pictures made up of picture elements (pixels) sufficiently small that each individual element is not easily visible. The resolution, quality or detail of a picture depends upon how many pixels are used. A television picture however is made up of lines of information and is not divided into pixels. When the computer derived image is presented upon a television monitor there is a compromise between the two technologies. It is recognised that an image derived from a certain density of pixels will appear very similar to the television line picture. The pixels are usually expressed in terms of the number in a horizontal-vertical array; for example SVGA is defined as 800 x 600 pixels, which converts into the 650 line television image with lines to spare. For this reason the SVGA image usually is cut off at the bottom when displayed on a television monitor. The aspect ratio (the ratio of height to width of the picture) of the SVGA image is different to the television picture which also explains why the images are not simply displayed.
To display this image a scan converter can be used, and now scan converters are becoming available which will transfer the images from high quality computer displays such as the Sun and Silicon Graphics to the television monitor. In these latter cases there will be some loss of resolution as information has to be thrown away to display on a television monitor.
Older equipment would have worked with VGA resolution, and this is the case with much video conferencing equipment. The higher resolution computer displays require more information and therefore more bandwidth, which is not available on ISDN and current IP systems.
In the audio-visual and communications industry, we have seen the transformation of a PC computer with the capability of doing number crunching using a spread sheet, record keeping with the data base and glorified typing using a word processor to the advent of the powerful workstation that appears limitless. We have also seen the introduction of the multimedia PC allowing the gap between computer and video to be closed further.
Equipment is available which allows computer output to appear on a single scan monitor or television. With multimedia information, which can be processed on desktop-computers, systems are required which will permit the multimedia output to be presented on large screens for use in teaching space. Only one or two people can easily observe a computer display at the same time, but in teaching situations, a significant number will need to observe the display screen, and this can be done cheaply by using television monitors or projectors.
Scan converters can be used to convert the signal from a computer for display on an NTSC or PAL consumer television/composite video monitor. If a resolution of 640x480 pixels is adequate, the MAC or VGA screen from a PC can be displayed on the monitor. There are also scan converters designed to deal with higher resolutions up to 1024x768 from a SUN and Silicon Graphics terminal at 48Khz. The NTSC or PAL Monitor must have a video input jack (line video like the output of a VCR). These devices can be situated at several feet distance from the monitor allowing their use in teaching space. In some cases the signal can be extended for a long distance with very little visible degradation. In other cases the distance can be very short.
It is possible also to link the MAC or PC to an Liquid Crystal Display (LCD) panel as well as to television monitors. As LCD panels can be portable and provide a bright image it is now increasingly common for them to be used in lecture theatres and seminar rooms.
Graphics Multimedia Virtual Environments Visualisation Contents