2 Inside video

 

1. What is digital technology?

To some people the so-called DV revolution spells the end of a certain kind of movie: the end of celluloid and all the associations that we have with it — projectors, rolls of film and the romance of Hollywood. To others, though, it is more than a revolution or a change in the industry — it's year zero. To filmmakers who could only dream of committing their ideas to moving images, the DV age is the start of a career, not just a change in one, as the affordability of cameras and editing equipment makes it possible to get involved.

DV is here to stay, is reaching far wider into the mainstream of filmmaking than people thought possible and, in a few years, has affected every part of the video and film industries from Hollywood down to indie features, and finally — and most dramatically — the low-budget emerging filmmaker. In most revolutions of the technological sort (and maybe the political) it is those at the top that seem to benefit most financially, but the DV revolution is one that bucks the trend; big money producers agree DV makes a difference to the costs of a film, but to the filmmaker at the bottom it is the deciding factor in being able to make one at all. But before we run off with our winnings, it is worth looking more closely at this gift horse and trying to understand why it works so well and how it offers us what it does.

Digital video

Digital is called digital because it records information by the use of numbers: ones and zeros, which correspond to ‘on’ or ‘off’ commands. It has no variables as does the wave of analog; a signal is either one or the other, black or white, yes or no. This means that when the tape signal degrades after copying or playing — which happens however hard you try to avoid it — it alters only the strength of the yes or no, the on or off, the one or zero. It still gets read as one or the other, regardless of the strength of the signal. This is why digital is a better method of storing and reproducing information.

It is the reproduction of it that is crucial, since the ability to edit and distribute without loss of quality is to remain true to a director's original intentions.

Filmmakers used to rely on 16 mm film for their first forays into movies, and VHS tape never really caught on as an acceptable replacement. Filmmakers care about the way a picture looks and digital offers what many are looking for at the price they can afford, while VHS was neither quality nor did it have a specific ‘look’ as did 16 mm film. Why was VHS analog so bad? If you take a picture and photocopy it you produce a version of the original. But if you want to re-copy it you lose some information and the resulting copy is less clear than the original. Repeat this process several times and you end up with a muddy, unclear image. With digital, you are approaching the picture wholly differently: imagine breaking the tonal values and colours of that photo down into numerical values, in turn represented by ones and zeros. You have then got a set of instructions for the make-up of that picture and can send these instructions anywhere. All that it requires is that the receiver has the same information as the sender in order to be able to reassemble the image from the numbers it is given. With the right decoding knowledge it will reconstruct the image. Furthermore, because it exists in terms of numbers, it can be manipulated more easily, so that a picture can be turned black and white by exchanging one set of numbers for another, while keeping the rest unchanged.

Figure 2.1 The digital signal on the right works by attaching a quantity to the wave. Each step of the curve is recorded as a number and this can then be reproduced more accurately than the analog signal on the left.

Why choose DV?

Picture resolution

In general, the resolution of a picture on DV is about 25 per cent better than a comparable image on S-VHS — the best consumer-level tape for analog — and has about twice the horizontal resolution of a VHS recording. Low-priced mini-DV offers about the same quality as Betacam, previously the highest level analog format, which is a remarkable leap forward for the low-budget filmmaker.

Colour rendition

This refers to how well the format reproduces colour accurately. Analog is prone to vision ‘noise’ or interference, which smears and blurs colours. With DV, colour rendition is more true to what the digital camera first recorded. Edges are sharper and shapes more defined. But, as with analog, a lack of adequate light in DV filming quickly reduces quality. When you go up to better quality cameras — threechip camcorders — things improve as the range of colour that can be recorded is larger, with each chip accommodating one of three colours rather than one chip having to cover all three.

Copying

DV offers greater quality control when copying. Since digital loses virtually no information each time it is copied, images can be transmitted or stored and the image that is read by the player is as good as the one on the master copy. The image is limited only by the quality of the playing instrument. Storing a film on any digital format guarantees greater longevity too. Magnetic tape is vulnerable to all sorts of environmental change, including heat, humidity and other magnetic fields, but digital tapes are very hard-wearing compared to previous formats.

Compatibility

Digital is ‘backwards compatible’; in other words, all digital is compatible with itself, whereas other formats such as analog were not so interchangeable.

Broadcast standards

Most countries have very specific guidelines about what kind of picture quality is acceptable for broadcast. Pictures have to measure up to a set of standards and failure to meet these consistently may lead to the licence to broadcast being revoked, so broadcasters reject material from filmmakers which don't meet these standards. Some digital camcorders meet these specifications, but at the lower end there is an important difference. But broadcasters will accept material, in fact whole programmes and movies, which have been shot on cameras that are not actually broadcast standard. The Sony PD170 range, for instance, is a favourite with filmmakers, particularly documentary makers, and yet it should not really be good enough for TV in terms of its technical characteristics. Yet it is close enough to do the job, while other cams such as the Canon XM2 are well regarded by TV commissioners across the board.

However, beware of the term ‘broadcast quality’ if you see it on the shop window at your video store. The term has changed considerably since DV became a consumer item; in terms of the consumer it has become a loose term, often simply referring to a quality far above that offered by analog but not necessarily meeting broadcast technical quality guidelines. Whatever the publicity blurb says, check with an independent advisor first (don't even trust a manufacturer) before committing yourself to an expensive camera that claims to be broadcast standard but stops your programmes being accepted by television companies.

Audio

The CD format in music offers superior quality to analog and almost ‘lossless’ copying. As you would expect, then, audio channels on DV video are comparable to CD and when recorded on 16-bit stereo offer unparalleled quality.

A brief history of digital technology

DV has grown in the last decade from a corner of the video industry into a mainstay of the consumer leisure industry. Developments have been rapid in this early stage of the DV revolution, so it is easy to confuse the passing fads from the long-lasting major changes to the industry. But you could chart the advance of digital video in two simple words: Apple and compression.

A brief history of digital video

It was the arrival of the Apple Computer in 1984 that heralded the start of the multimedia age ‘for the rest of us’, as the company logo read. The big leap forward was the ‘graphical user interface’; in other words, a more user-friendly way of dealing with computers, by having windows instead of glowing green letters on a black screen. This windows layout is the only feasible way for creative programs to work because you need to be able to work with several different kinds of commands at once — palettes of colours, video timelines and so on. The use of Microsoft Windows on the vast majority of the world's computers is a direct result of the pioneering work done by Apple's founders and by companies such as Lotus, Adobe and Digital Research Inc. Apple supremo Steve Jobs crops up again later in our short history lesson as the pioneer of another big step forward.

The important breakthrough in thinking was the view by Apple that computers could be creative tools, as Apple's early Macintosh home computers came equipped with painting software as well as spreadsheets. Rather than adopt a passive, consumer ethos for its products, Apple foresaw the Mac as something owned by makers, users and creators.

We used to edit in a very long-winded way, piecing together bits of film with tape or glue. Video was not much of a leap forward as you simply had two VCRs hooked up and you copied your good stuff from one tape onto your blank tape on the next VCR. It made sense later to call this ‘linear’ editing, since you laid down the bits of good footage one after the other and, crucially, there was no going back. Once you had copied them to tape you couldn't revisit an earlier section and re-edit it without having to redo all the following clips too. On the plus side, editors had to be more decisive than today.

So, when Avid, a leading multimedia company, unveiled its non-linear editing system it was the biggest shake-up in editing since Melies played around with time and sequence in the early 1900s.

We called it ‘non-linear’ editing in the professional arena and desktop editing for the consumer. Working on a computer, clips could be endlessly switched about and played until the perfect sequence was found. You could work on various bits of the movie all at once, building it up in a more organic fashion. Actually, it has more in common with the old scissors and glue method of celluloid, since you used to go back and add bits of film and work on any part whenever you wanted. Certainly, the kinds of visual devices common to us today, such as montage or flashback, are a result of being able to access any part of a film.

1994 saw the next big step, with the patent of Apple's FireWire cable — patented as IEEE 1394, or the Institute of Electrical Engineers' 13th patent in 1994. The big problem with video is its sheer size and here was a cable which didn't baulk at the huge amounts of data it had to conduct. Transferring at a huge 400 kb/s it avoids drop-outs of frames when you transfer footage from camera to computer. To use the analogy of traffic on a road, FireWire is like a freeway/motorway, while other cables like USB 1.0 are like quiet back roads; send too much traffic and it gets clogged up. The need for FireWire arose because new digital cameras were hitting the market which could send data direct to a computer, whereas analog cameras had to first translate their analog data into digital code. Like translating a document from English to another language, this process added time and expense.

With manufacturers putting cameras into the market which recorded in digital, FireWire made it possible to make the most of them on the home PC. The biggest problem next became what to do with the movies once you had edited them. The available storage disks were insufficient — floppy disks — and the modems of early Internet users were not nearly capable of downloading video files.

MPEG, Pixar and projection

Which brings us to the development of clever ways to squash down video into manageable files. Once again, Apple stepped in with QuickTime technology in the early 1990s, again placing it firmly in the consumer market rather than expecting it to have purely professional uses. A group of international experts called MPEG had been working for some time to standardize the issues around compression and made it easier for the expanding consumer PC market to agree on the kinds of software needed. The Motion Picture Experts Group developed the standards MPEG-1, used on much early compression; MPEG-2, the standard for DVD compression; and later MPEG-4, used for web-based video and gaming.

Once you could edit digitally, it was a short step away to manipulate each of those digital frames using other software. In 1986, former Apple boss Steve Jobs bought Lucasfilms computer graphics division for $10 million. Renamed Pixar, their early films went one stage further by creating images wholly on computers, and this process of computer-generated images (CGI) has since made its way into almost every kind of movie today, whether it is to paint out an errant catering truck in the corner of a frame or the wires holding up an actor, or creating entire objects to interact with live action, such as hurricanes (Twister) or aliens (The Abyss). By 2004, films such as Sky Captain and The Polar Express pointed to a whole new method of making films, where real action is captured on computer and used to animate objects and people.

In 1992, a theatre was hired by Sony Pictures and used as an experiment with Pacific Bell to give the world a first view of digital cinema to an audience of several hundred. Rather than having to make a celluloid version of a digital movie so it could be shown on celluloid projectors, a digital projector introduced the idea that films could be digitally shot, digitally edited and finally digitally exhibited.

The web

In the mid-1990s, the idea of connecting PCs together took off when the Internet, formerly a system for hitching up university computers to share research between institutions, became the must-have addition for home PC users.

Advances in compression had been swift and by 1998 the compact disc was being used to hold entire feature films, using enhanced dual-layer discs and MPEG-2 compression — more of which in Chapter 3:2.

By the turn of the millennium, digital cinemas were becoming a reality for audiences in North America. On 19 June 1999, Fox and LucasFilms staged Star Wars: Episode 1 — The Phantom Menace in four theatres using digital projectors. A few years before, a low-budget movie called The Last Broadcast became, with no irony, the first to be broadcast by satellite to a theatre in a move that could yet pose the biggest threat to the mighty distributors.

High definition

Into the millennium, high definition (HD) video has pushed the boundaries of what filmmakers can use, doubling the lines of recorded video and kick-starting the possibility of HD-TV for consumers.

But filmmakers have found out that recording HD is one thing but getting it to squeeze onto tapes is quite another. Video compression may have to take another huge leap forward before consumers and low-budget filmmakers can use this new format. It also asks filmmakers to consider the question of whether picture clarity is the holy grail after all.

All this talk of technology can lose sight of what it was that these pioneers in DV were trying to achieve. Ultimately, it was about the art and craft of filmmaking and of storytelling. Like the Gutenberg printing press, which placed the printed spread of information in the hands of the masses, DV technology has enabled more to be produced with less, and by more people.

For all its being showered with new toys and gadgets each year, filmmaking remains an art form, concerned with images and story.

The Crunch

• Become technical — most artists have to work at this bit, but it is worth it because you are in control
• Get to know about broadcast standards for your country or the place you might want to show your work
• Analog is long gone — digital is going to give you better results
• FireWire is good — you need it.

2. How the camcorder works

Before we look at the inner workings of the video camera, we need first to define what we are looking at. Not all cameras can record images; some simply pass them on to another device, which commits them to tape. In television, for example, cameras in the studio relay images elsewhere and so are not, technically, camcorders, just cameras. For simplicity, then, we will talk throughout this section about the kind that records images as well as seeing them, from entry-level consumer models through to those in the upper range used in filmmaking.

How images get onto tape

To understand how the image gets onto the digital tape, we could start at the beginning and follow the whole process. It basically runs like this:

• image through the lens
• image is sensed by a CCD chip
• gets translated into binary code
• then gets transferred to tape.

Surrounding this process are a range of features that help modify the image, making it clearer or more stable, if needed. Audio signals recorded simultaneously go straight from the microphone, getting translated into digital information and onto the tape, and are matched with the images accompanying them.

What happens when the image hits the lens

Lens

The lens is the point at which the image crosses into the camcorder, using an iris which functions much the same as the human eye. Lens quality is crucial; the camcorder may have a perfect method of recording what it sees, but to make the most of it must see perfectly to begin with. Consumer-level camcorders will not usually have the lens quality of those in the semi-professional and professional range levels, although some manufacturers try to improve the lenses on cheaper models by teaming up with established photographic lens makers.

Iris

The iris in the lens controls the amount of light entering the camcorder and will quickly shrink if too much is entering or dilate if there is too little. The reason it must do this is to satisfy the amount of information required by the CCD chip situated near the lens. If there is too little light and the required pixels do not each receive instructions, then the chip steps in and starts adding its own default information. This is known as ‘noise’ and is read as tiny white dots all over the screen, which is why you get a grainy image if you shoot in low light conditions.

Shutter speed

The shutter and the iris overlap in their roles to some extent. When light passes through the lens it hits a shutter that is opening and closing 25 times each second (in PAL standard cameras for the UK), or 29.9 (NTSC) for North American cameras. Each of these images that gets recorded is called a frame. The camera can alter the speed at which it opens its shutter, even though it still opens it the same number of times each second. The minimum speed for many cameras is 1/50th of a second, with the maximum in consumer level cameras around 1/8000th of a second.

The difference between these is most evident when you record fast-moving objects. In 1/50th of a second, the length of time a shutter is open, the object (let's say a car) travels a long way in front of the camera in one frame (let's say 20 feet), and therefore blurs all the movement within that distance onto one frame. If you up the shutter speed to 1/1000th of a second, then the shutter is open for only a fraction of that and records only a fraction of the movement, say one foot, resulting in no blurring.

However, there is a knock-on effect in reducing the amount of time the shutter is open; less light can enter the lens. So it is kind of doing what the iris does in keeping some of the light out. If your camera has manual controls then it will be straightforward to adjust both the shutter and the iris to make sure that enough light gets in to give a full, clear picture. You can also use shutter speed and iris settings to give a darker and more jagged effect to a movie. In 28 Days Later (Danny Boyle), a high shutter speed in this digitally shot film gave it a spooky look, in keeping with the post-apocalyptic scenario and helping the large number of action shots by preventing blur.

So far so good, but the iris and shutter can play other roles too. If you reduce the hole in the iris then the range of objects that are in focus is altered too — this is called ‘depth of field’. If you think of a line of trees in an avenue with you standing with your camera in the middle of the road, not all of them are going to be in focus all the time. If you were to adjust the iris to let less light in, making the hole at the front smaller, then there would be more of the line of trees in focus. If you were to open the iris, then less of them would be clear.

It makes sense, then, to think about these three aspects of a shot — focus, iris and shutter — as being linked and needing a joined-up approach to dealing with them. When shooting, if you adjust one of them, think about how it affects the other two.

Zoom

Another function of the lens is zoom. Digital camcorders zoom in one of two ways: optical or digital zoom. Optical zoom is by far the better option, as it magnifies the image using two glass lenses and reads it more accurately. Digital zoom, however, is almost without merit, as it simply enlarges the image it saw, drastically reducing resolution.

How the camera deals with the image

When the image makes it through the lens it gets recorded in various ways by a small chip located near the lens, called the charge-coupled device (CCD). There are two sorts, determined by which kind of camera you have:

• Single-chip. These dominate the lower, domestic end of the market. It collects over 300 000 pixel bits of information, which sounds a lot but is considerably less than the three-chip.
• Three-chip. This one is found in any of the better camcorders in the mid and upper ranges. The big difference here is that there are three chips, each gathering 300 000 bits of information, but this time each chip gathers just one colour each: red, green or blue. The single-chip gathers all colours together and so has to squeeze all three groups of information into one chip. The additional quality of the three-chip is worth about an extra 20 per cent in terms of picture and colour resolution.

If you want to make programmes for broadcast then you need to get a three-chip. Most people will not notice the difference between single-chip and three-chip, but broadcasters will notice and do not accept certain quality of pictures. But if you intend to show only on the web, this difference is going to be reduced later, so it may not be useful.

At this stage the camera sorts out how much information can be stored and how much has to be left behind. Mini-DV tapes do not have the capacity to store all the information that a camera could record but are significantly better than VHS. In terms of a ratio of information kept to that thrown away, mini-DV is 5:1, compared to 7:1 for analog Hi8 and 15:1 for VHS. How well a camera squashes down its information is crucial. The new generation of HD cameras have suffered from a poor compression method. It remains a problem for manufacturers to convince the market about HD because the recorded picture is so reduced from what it should be, given the huge size of each frame of HD video.

What happens when the image gets recorded onto tape

Digital tape is much more rugged than that used for analog S-VHS or VHS video. It is made of an advanced form of metal evaporated (ME) tape. It consists of a double-coated magnetic layer, which is in turn coated in tough carbon. This enables stored digital information to be played back with minimal loss of quality and less picture noise, or interference. ‘Bleeding’ from the audio track to the video tracks is also unlikely, further reducing interference.

Solid-state cameras record onto discs but have to use a sophisticated form of compression to enable it to do so. In consumer models images are recorded onto a fixed hard drive or onto DVDs, though the latter have failed to take off in the consumer sector in the way manufacturers would have liked. Cameras using hard drive recording have become more widely used, possibly a result of the iPod effect, where consumers understand and trust mini hard drives. A further development has been in removable hard drives. For filmmakers using cameras such as the Canon XZ2 or Sony PD170, the favourites of many low-budgeteers, hard drive recording was not possible until large removable 80GB drives such as Holdan's Firestore made it to industry shows.

Figure 2.2 Compression rates of camera formats.

Recording standards

There are several other factors that are worth mentioning in relation to recording. If you intend to make movies or programmes for selling in other countries, you need to know about the differences in broadcast standards around the world. Ultimately, this affects anyone who makes films because few people can afford to ignore the potentially lucrative returns in foreign markets. None of these problems are insurmountable, but being aware of them before you start a project can increase your chances of selling your work to a wide range of markets.

PAL, NTSC and SECAM

Each country has its own specification for how many lines a television signal must produce and how many frames per second it should have. In Europe and Australia, the standard is PAL (Phase Alternating Line), which operates on a high level of quality in terms of pixels on screen but slightly lower frame rate. In America, the standard is called NTSC (National Television Systems Committee) and has a lower number of pixels on the screen, but operates on a faster frame rate of 29.9 (often rounded up to 30). A different system called SECAM (sequential couleur avec memoire, or sequential colour with memory) is used in parts of eastern Europe, some African countries, parts of South America and the former states of the USSR. SECAM is closer to PAL than NTSC, using the same bandwidth as PAL but transmitting colour information sequentially. Countries using SECAM can play PAL videos or DVDs more easily than NTSC.

To complicate matters, there is also a version of PAL known as PAL 60, leaving the old PAL to be known as PAL 50. The new version is intended to help address the problems of NTSC users viewing PAL movies. NTSC uses a frame rate of 30 frames per second while PAL uses 25, but the new PAL also uses 30.

Figure 2.3 Table of international TV standards.

Pixels

The number of pixels, or dots on the screen, is determined by how many vertical and horizontal lines the signal has. The NTSC ratio of 640 × 480 has roughly 307 000 pixels, whereas PAL has 720 × 560, giving it the edge with over 400 000. This will lead to greater clarity and sharpness in PAL, but it suffers when it comes to frame rate.

Frame rate

Frame rate (the number of individual pictures that have to make up a single second of tape) affects smoothness of movement. A film shot on fewer frames per second (fps) will look more jerky than one on higher fps. There are ways of easily transferring a recording so that a programme made for American TV (NTSC) will play on British or Australian TV (PAL), even though they use different frame rates.

Figure 2.4 Video formats explained.

Timecode

Another, more complicated factor in frame rate is timecode. In terms of frame rate, timecode is more of a headache with NTSC than with PAL. PAL, with its nice, easily divisible numbers (25 goes into 100 neatly), presents no problem for accurate timecode. The 29.9 fps of NTSC, however, needs to operate with something called ‘drop-frame timecode’, in which certain frame numbers — not your actual hardwon frames themselves from the footage — are dropped out to keep everything in round numbers. It gets rid of two frames every minute but skips every tenth minute. Thankfully, you no longer have to worry about drop-frame timecode as many edit programs offer you the chance to work in this method. But it is essential if you intend to work in productions of 60 minutes or more, as you would end up with an error rate of 3.6 seconds every hour. If you are intending to sell broadcast material to zones with other broadcast standards, check what they use and edit the movie with this in mind.

The Crunch

• You need manual override on your camera
• Get a three-CCD-chip camera if you can
• Understand the way cameras record so you can use them creatively
• OK, maybe you don't want to know all this technical stuff, but make sure someone in your next crew does.

3. Operating the camcorder

Every model of camcorder is different, but there are certain ways of handling camcorders that make the best of what you've got. From the point of view of the filmmaker, as opposed to the tourist or home movie maker, some features need to be looked at a little closer.

Camera support

Digital cameras are so lightweight that the problem of camera shake is more pronounced than with previous models and the need to support the camera in some way is crucial. Even hand-held work is going to be vastly different for digital cameras than for film, so whatever your preference in films you will need to make some plans for reducing camera shake. Check that your camera has a universal tripod fitting on its base. Many tripods offer a quick release plate that locks the camera onto the tripod without having to screw the camera onto the tripod each time.

Zoom control

Using zoom on a subject does not always produce good results; zoom controls tend to move at a constant rate and this can look quite unnatural. It is considered to be a slightly redundant way of recording, except in non-fiction work, unless used as a self-conscious device. Use zoom only when a tracking shot towards the subject is not an option.

Focusing

Automatic focusing is useful for tourist movies, but in productions where artistic considerations are high, this is going to stop you from getting good compositions. Be prepared to use the manual override option to use your own focus on certain shots. In automatic mode, the camera senses the distance between itself and the nearest object, through the use of infrared beams bounced off its target. The problem with this is that it can't focus on everything within the frame, so it chooses only that which is in the middle of the viewfinder. This renders any slightly adventurous compositions out of focus. When interviewing a subject, for example, it may look more interesting to place the head to one side of the screen, improving depth. You will need to focus manually if you are to avoid getting the wall next to the figure in focus and the head blurred. To do this, use automatic focus on the subject but place it in the middle of the screen, let the auto focus find its optimum point after a few seconds and then switch to manual mode, freezing the focus at this. Then pan the camera to the right composition. Moving the camera closer or nearer to the subject will again render it out of focus and you must go through the process again. Deal with moving, unpredictable subjects like flames and running water in a similar way, but pan the camera to the side first so that you let the auto focus settle on a solid object the same distance from the camera.

Aperture and depth of field

Aperture refers to the opening or iris which allows varied amounts of light through the lens. Aperture affects both the light entering the camera and the range of the frame that is in focus, known as depth of field. Attaching other lenses also affects the range of your focus. For example, wide-angle lenses allow for a far greater range of focus, from objects far away to those near the camera, while telephoto lenses have very little depth of field.

Altering the iris

As we saw in the earlier section on how cameras record, the iris can be altered for creative uses, using manual controls:

• Closing the iris so that it is very small will let less light in and also enable objects in the foreground and background to be in focus — something that may look unnatural or could be just the effect you want
• Closing the iris can also result in well-lit sets appearing to have extreme contrast — the dark areas are unreadable and the white areas are strong and bright
• Opening up the iris will result in a narrow band of focus, so that objects or people moving towards or away from the camera are easily thrown out of focus
• Opening up the iris will also make normal lighting conditions seem over-lit — even an average lamp will seem to glow unnaturally.

If you manually operate the iris, beware of distortion at the edges of the frame.

Light and the camera

The automatic features of your camera are going to affect how you shoot. In auto setting, the camcorder thinks it is doing you a favour by removing some of the light on a bright scene or opening itself up to lighten a dark interior. This automatic light meter causes problems as it is going to stop you from using creative lighting effects. Video, in general, does not like to be deprived of light and will react badly if you try to work in conditions that differ from the norm. Just about all filmmaking involves using lighting in some creative way and your auto functions will do their best to cancel out any effects you set up.

To make sure you get a good image, light the scene well but try removing the lit areas and increasing shadow. You can learn a lot about lighting by trying to create shadow rather than create light. Any household lamps can be used for this as DV requires far less light to make an effective picture than VHS; 300 W lamps will be sufficient rather than the bigger 800 W common in many VHS low-budget shoots.

When you have the light looking good — creating the atmosphere that is right for your scene — avoid any fluctuation of the aperture during recording. For example, a tracking shot of a figure in movement between a desk with a bright lamp to a darker corner of the room will present problems because of changes in the level of light available. Moving between these two places, the camera is designed to step in and open up the iris as it gets to the darker corner. But that loses the whole effect, as the two places will appear to the viewer to be lit equally, with the uncomfortable moment in between the two places as the iris shudders around to get the aperture right. Again, switching off the auto controls will avoid this problem, but to find the right level of aperture, look through the camera at both the dark and light areas of the scene with auto switched on for a moment. Let the camera settle in one part between the two and find an aperture that is right for both areas.

White balance

White balance is a tool — usually automatic in most cameras — which assesses the temperature of light entering the lens and compensates to alter any inaccuracies in colour, to something approaching the mid-range in degrees kelvin. Colours seen in natural daylight are more true than colours seen under artificial light, due to the effect the different colour temperatures have on natural colours.

The automatic white balance feature filters out some of these colours to arrive at a more realistic light and should be used for each new take. As with other functions, you can alter it manually and use it as a creative tool so that, for instance, daylight looks bluer and colder than it should — by setting the white balance for indoor, household light it would assume there is too much orange and remove some from the picture. Get to know how your camera functions when you set the colour balance incorrectly. Don't forget to avoid altering the white balance setting while you are in the middle of shooting a scene.

Power

Batteries are expensive and don't last long, but cables are a hindrance. Think about how you can spread the use of your batteries, or hitch up to the mains as often as possible. Loss of power creeps up on you and almost always happens when you are about to shoot a crucial scene that can't wait. Some cameras in the upper range tell you when you are low on power or have a power meter to keep you informed constantly. Check whether your camera warns you when your battery is running low. If you are on location and have no access to mains supply, conserve power by restricting the amount you rehearse on camera or play with different camera set-ups and do not use the LCD display if your camera has one fitted. Try using a dummy lens, such as a still camera, to try out different compositions or framing, so that when you use the video camera you use less power looking for shots. If you do get stuck without power suddenly, hold the weak battery under your armpit for a short time; sometimes body warmth is all it needs to give you that last bit of power to finish the shoot.

Audio inputs

These will allow you to use external microphones for recording, or the use of audio mixing boards (on some models). Using external mics is strongly recommended.

Listening to audio

Most cameras have built-in speakers so that you can hear what is being recorded. These are no substitute for earphones, though. Make sure you always use earphones every time you shoot. Interference from power cables and mobile phones will affect the quality of your sound and if you have headphones you will hear this straight away.

Timecode

Most consumer-level models do not refer to timecode as such, but do have functions to allow you to start recording right after the last frame of the last take you shot, thereby ensuring that you don't break timecode. Models above this level will all allow you to record with timecode. Always ‘stripe’ your tapes before shooting — record continuously throughout the whole tape with the lens cap still on. This lays down a constant timecode track before you use the tape.

The Crunch

• Be in control of the camera
• Use timecode
• Learn how to move with it
• Get a tripod but don't overuse it
• Zooming in or out does not look great
• Always get the image in focus
• Always use an external mic.

Figure 2.5 Camcorder buying guide: features needed in DV filmmaking.

What this project is for: to learn about manual settings in shooting

Time: allow about three hours

What this project is for

The aim of this film is to encourage you to use the camera with complete control over every aspect of what it is doing. Cameras are easy: they do what you want, when you want, but they let you down when it comes to helping you be creative. Indeed, the aim of every camera manufacturer is to make life easier for you. But you don't want an easy life; that's why you are trying to make films, so we need to tell the camera to listen and do it your way.

To achieve this, we will make a short film that is made with the camera on manual settings throughout. If you have to resort to using automatic settings, it is so that you can deceive the camera, but more on this later. As with other films in this book, we need to make it short, lasting less than four minutes. The film is going to made in-camera, so you will be shooting each shot as you need them in the order they appear.

Stage 1

In this film, the subject matter is not the most critical element; it is no more than a vehicle for the actions and movement of the camera. The theme running through this film (it is too open-ended to call it a story) is titled A Day in the Life.

The film is going to follow you as you go through a typical day, encountering various situations and people. Since you have to keep the movie short, it might be useful to divide the film into sections, looking at different parts of your day. It may sound like a simple film, but many films are extremely neat in their ideas. Consider Kevin Smith's stunning debut feature, Clerks (1994), which was filmed entirely in the New Jersey shop where he worked. Smith manages to make one location and very little else go a long way, turning in a film which catapulted him into the big time. Richard Linklater's Before Sunrise achieves a similar feat in a few streets in Paris.

Stage 2

Spend some time getting to know what the manual features on your camera are there for. Look at the guide to camera features elsewhere in this chapter and look to see what yours has. Check:

• Where are the manual and auto buttons (if yours has manual override)?
• Have you got a manual focus ring?
• Have you got manual iris control (aperture)?
• Have you got manual white balance?
• How about manual shutter speed?

Stage 3

Choose a day and start filming. Shoot some shots of, for instance, what your house looks like in the morning, about who eats breakfast in your kitchen, where you have to go to work or study, who you meet along the way, where you eat lunch, who you meet in the evening and where you go, what it's like at night in your town. In all these situations, you will have to encounter environments that are very different to each other in terms of what the camera sees. Conditions will vary from place to place and from day to night, providing you with opportunities to encounter a wide range of situations.

Before we go further, let's take a look at how to handle sound in-camera. This project can operate on two levels: a basic level that looks at what the camera sees only, and a more advanced level for those who want to start using an external mic. Take the basic for now, if you prefer, and then come back to this project later when you have got to grips with sound recording in Chapter 5:3. But if you want to try using mics, go for a directional mic, which records wherever you point it. Try the built-in mic at least once so that you know how poor the quality is.

The aim of this film is not particularly to produce a great-looking film at the end, but rather to encounter problems. You won't be able to solve all of them and you may need to get some help from the automatic button on the camera, but at least you will know the limits of what you can achieve manually and what you need some help for.

Let's look at some of the problem areas you may encounter:

Light

Usually, you expect your camera to take care of ‘exposure’ when you enter a new lighting situation. Exposure refers to the amount of light that enters the lens: too much and it is too bright to see anything clearly, too little and it is too dim to see clearly. Let's suppose you start filming in your kitchen in the early morning. You may have a couple of fluorescent lights in the room and some semblance of daylight entering from the window. You have two problems to deal with here: the first involving the intensity of the light within the room and the second concerning the colour temperature of the light. A lack of intensity of light is easy to deal with on set simply by adding more light, but at home you are going to have to get the iris control and try various settings until you have what looks like a good image on the LCD monitor. Open the iris up a little if the image looks too dark, or close it down slightly if it is too bright. Go through all the manual options it gives you until you are satisfied with the image, then start filming.

Colour temperature (measured in degrees kelvin) affects the colour of the light we see. For example, a candle gives off a very orange, warm cast of colour, while daylight on a late afternoon in winter would give off a bluish cast. If you have standard household bulbs in the kitchen, they may give an orange tint to the film, while fluorescent strip lighting can give a greenish cast. However, none of this would be picked up as the camera removes some of the extra colour. But what happens if you want extra orange? How does the picture look? For this, you may have to resort to the automatic white balance if no other option is available, but at this point run through the manual options it gives you (if any). Some cameras suggest a few settings for the most common lighting situations: indoor, outdoor, bright day, cloudy day and so on. Try out some of these settings incorrectly. If you try, for instance, cloudy outdoor settings for shooting in your kitchen it will assume that there is a dominance of blue in the scene, which would occur in this kind of light, and balance it with taking away some of the blue. The net effect is a kitchen more orange than usual, as what little blue there was is turned down even further.

Shutter speed

If your camera has options for changing the shutter speed you can alter these to take into account different situations. Shutter speed refers to the amount of time it takes the camera to open and close the shutter. A fast car, at close range, will appear blurred unless a higher (or faster) shutter speed is used (for example, from 1/125 to 1/400). However, faster shutter speeds need higher light levels, so altering these settings can produce some effects. Try shooting a fast car on a slow shutter speed and perhaps you might enjoy the blurring of the image, particularly if a stationary figure is in the foreground to contrast against it. Then you might want to play around with how a fast shutter speed — requiring more light — records in a low-light environment.

Focus

Focus rings (a ring around the lens that you turn to alter the focus manually) are rare on most camcorders. Without any manual means to alter focus you will have to rely on using (or abusing) the automatic focus feature. Auto focus works by projecting an infrared beam at the nearest object in the middle of the frame and assessing from the speed of its return how far away it is. So far so good. But if you want to stop the camera from doing this constantly you will have to let it find one particular setting and stay there. For example, if you meet a few friends and decide to put them in this film, you can shoot straight at them by putting them in the middle of the frame, which is going to look dull, or you can get a more interesting shot. Home movies and tourist movies put people in the middle, but filmmakers want more in their shots and so may move these people slightly to the side so that we see a few other elements of the scene. So, set up your friends so that the arrangement looks less predictable; then set the focus automatically; then turn it off and see what happens.

The result is that your friends are going to be in focus most of the time. If they move a great deal within the frame that's fine, as long as they move across the frame rather than close to or further away from you. But if you had left the auto focus switched on continually your friends would be switching in and out of focus every time they vacated the very middle of the screen. So, in this case, you use auto to get what you want, then get rid of it. Now you are in control of the camera.

Evaluation

This project is unusual compared to the others in this book in that it relies very much on improvisation and on making mistakes. At each step of the way, you are wresting control from the automatic settings of the camera and seeing what happens when you shoot against the way it suggests. After all, the settings it gives are simply designed to give you what is considered to be the closest representation of reality. But in most filming you are trying to create illusions, to improve reality in some way, and so it becomes crucial to defeat these realist tendencies in the camera. You know where they are if you need them.

I want something more challenging

To stretch your skills further, try combining two or more manual controls — for instance, combine the iris with shutter speed, or focus with aperture. These combinations enable you to create new effects, such as the way increasing the shutter speed reduces light entering the lens in much the same way as aperture.

The theme in this movie is designed to offer an easy route, but you could try combining this with another project so that you make a film from elsewhere in the book but solely using manual controls.

---

4. Video safety and good practice

Treat your camcorder well

Water

A camcorder can be irreparable if damaged with water. Protect it from rain and wipe off moisture from the casing. Salt water is especially harmful, and can do damage to the outer casing as well as the inner workings.

Condensation

Some camcorders warn you if condensation is building up inside the camera and many will switch off automatically. To remedy this, take out the cassette, if the camera will let you. If it is not possible, switch off and wait two or three hours. Once you have removed the cassette, leave the cassette compartment open for two or three hours, to dry, in a non-dusty room at average temperature. Condensation is particularly problematic if you film in extremely cold places and the water freezes. If this happens, let the camera thaw naturally at room temperature and then go through the same process as for condensation.

Magnetized equipment

Any magnetic field can adversely affect recording. Television monitors, video games and loudspeakers generate fields that distort picture quality. No lasting damage is done to the camera itself but your footage could be ruined. Test how the camera performs if you are in any doubt and play back the results on a monitor. Occasionally, the actual mechanism of the camera can be affected, temporarily, but this is often solved by unplugging your power supply or removing the battery and then reconnecting.

Sunlight

Pointing the viewfinder at sunlight can do great damage to inner working parts. Worse, it can permanently damage your eyesight too.

Radio transmitters and power lines

If you have ever listened to radio near an overhead power line you will have heard an amount of interference. This noise badly affects your video recording, so avoid shooting near pylons or other highvoltage lines. Radio transmitters will have a similar effect on the sound and image. Mobile phones also cause problems with sound interference while recording.

Excessive use

Many camcorders are not designed to be used for excessively long periods, as for example in surveillance filming; the inside temperature of the camera can cause malfunctions.

Dust and sand

Take care when inserting the cassette that no fine dust or sand enters the camera. If you are using the camera in a potentially dusty environment, use a protective cover. If you don't have a cover, you can improvise with a black bin-liner, wrapped around the casing. But beware that the camera could overheat over a period of time.

Cleaning

Some camcorders react badly to being cleaned with benzine or thinners. The body casing can be deformed as a result. Use a soft cloth with mild, dilute detergent mixed with water. Wring out the cloth until nearly dry before use. Never clean the camera while the battery or mains supply is attached.

Clean lenses rarely, and with great care. Avoid getting the lens dirty. Try fitting a protective clear UV filter (semi-)permanently to the front of the lens if the camera has a filter thread fitted. You can always buy another cheap filter when it gets damaged.

Cleaning the video recording heads is advised, but with care. Sometimes, after a lot of use, dirt and tape particles build up on the video heads, obstructing perfect quality recording. Some cameras tell you when this is happening, with a display on the monitor. To clean the heads, get hold of a cleaning cassette, which cleans as it plays. But don't do this too often — the cleaning cassette is an abrasive cleaner.

Power

AC adapter

• When you charge your battery, make sure that the temperature of the battery is not excessively high or low. Charging when the battery is outside the right temperature will not adversely affect the charger — it will simply not charge. Some camcorder adapters will start to charge automatically once the right temperature is reached.
• When the battery is warm, charging can take longer.
• The AC adapter may affect radio reception, so don't charge near your radio antennae. One metre away from the charger is sufficient.
• Keep the terminals of the charger and the battery clean to enable maximum charging power.

Battery care

• Batteries for many camcorders work by generating electricity through a chemical reaction, using lithium. The reaction is easily affected by temperature and humidity, and impedes the amount of power you get from it. At very cold temperatures, a battery may have its life-span cut down to just five minutes, while high temperatures may cause the battery to switch off for some time.
• Protect the battery terminals from moisture, as this can cause rust to develop.
• If metal objects touch both battery terminals at the same time they may cause it to short-circuit and generate serious amounts of heat, even starting a fire. If you pick up a battery that is short-circuiting, you may receive burns.

Do not leave the battery attached to the camera for long periods when it is not in use, as this can damage the voltage level of the battery and affect its ability to recharge.

Using the LCD monitor

• Temperature will affect picture quality on an LCD monitor. When cold, the picture is darker than usual, even in reasonable climates. After a while the ambient heat of the camera is enough to rectify this, but bear it in mind if you use the monitor in cold environments. Over 100 000 pixels are used on these monitors but less than 1 per cent will be inactive, sometimes affecting picture quality. Don't worry about this as it does not affect recorded picture quality.
• Remember that using the LCD monitor will run down the power in the battery much faster than if you use only the viewfinder; a one-hour life-span is reduced to 30 minutes when using the monitor.
• If you have a touch-sensitive LCD monitor, avoid sharp objects touching it. Also avoid touching it if you have been using cleaning chemicals or other potentially corrosive substances.

Videotape care

Even though DV tapes are much more sturdy than VHS or S-VHS, they are still prone to problems from poor handling.

Storage

• Always store tape vertically, with the tape rewound, in its case.
• Store away from magnetic fields, direct sunlight or excessive moisture or dust.
• Avoid touching the tape.
• Avoid dropping the tape, or causing any other shock or impact to it.
• Cars are very bad for tapes; the temperature inside the vehicle rises and falls dramatically when not in use.
• If cassettes are excessively cold, let them warm up to average room temperature for a couple of hours.
• Do not leave the cassette in the camcorder for long periods.
• Rewind the tapes after use; tape stretching can occur.

Always label your tapes the moment you take them out of the camera, before you reach for the next tape. Every filmmaker has stories of having to trawl through numerous tapes looking for a particular lost piece of footage.

5. Video compression

Now that there are so many different ways of watching video — phones, the web, DVD, CD — it was inevitable that at some point we would have to find a way of reducing these movies to a size that our mobile devices can cope with. Video data is huge and it is inconceivable that it could be sent to any new format just as it appears on your master tape. To reduce this data, we use a process called ‘compression’.

Over the last decade there have been big advances in the way we squash movies. There are now very sophisticated methods which present to the viewer a version of the movie that they would find hard to detect next to an uncompressed version. Much of this work is down to a group called the Motion Picture Experts Group (MPEG).

In this section we will take a tour through the basics of compression, as it is used by filmmakers on a day-to-day level. Compression is one of those subjects that gets rather involved and more than a little technical, but we will focus instead on the ways it affects a movie and how you can lessen these effects, utilizing the new mobile and web formats to the full.

How it works

In essence, compression looks at individual frames and decides which parts of them we would not notice if they were discarded. For instance, a blue sky that is the same over five frames doesn't need a new bit of information for each of those frames, so the whole group is assigned a code.

At its most sophisticated, compression works by doing three jobs: first to analyse the video signal, then to sort out which parts of each frame the viewer wouldn't miss if they were discarded, and finally by encoding this process. The best way of doing this is called discrete cosine transform (DCT), which samples a piece of video at regular intervals and figures out what an image can lose that the human eye would not detect. It is a complex algorithm that is behind the main methods of squashing down video that we use every day, including DVDs, Internet gaming, web broadcasts, CD-ROMs and cell phone movies.

The key to compression is that if you are going to squash something down, you have to be able to unsquash it again later. We could think of this as similar to sending a large poster to a friend via the postal system. The poster has to be rolled up so that the mail service can transport it efficiently. But at the other end, when it reaches your friend, it needs to be able to be unrolled again. We could call this process ‘compression-decompression’, shortened to ‘codec’, to describe the various ways we have of doing it. Many companies produce codecs for use with video and each use the standards that MPEG established. Some are good for one type of movie, others better for other types.

MPEG

The Emmy Award-winning MPEG group worked out several methods of compressing data for moving images and came up with a few different standards. Let's take a look at the main ones:

MPEG-1

This is a popular method used for encoding video for CD-ROM and for VCD, the format widely used in Asia. It is the main method used for web video with the suffix ‘.mpg’ and also gave rise to the standard known as MPEG-1, layer 3, or MP3, which is the most widely used method for compressing music files.

MPEG-2

This method is a development of MPEG-1 which grew into the way we compress video for set-top digital TV boxes and for the DVD. It is probably the most widely used compression in the world, with the spread of the consumer DVD outstripping VHS, and the imminent digitization of TV broadcasting. At the moment there is no need for an MPEG-3 because this method is seen to perform well with high definition television, which will become more widely used in the near future.

MPEG-4

The need for this standard came about as we started to use moving images on the web and in mobile web devices. The format handles multimedia items very well and is ideal for gaming, broadcasting via the web and pages that contain media items. The best aspect of this method is that you can decide how compressed you want the video to be. This is called enhanced scaleability, and is popular with media providers because various options can be given for the user, depending on what sort of connection they have. They can opt for high compression for small bandwidth (a smaller file for 56k) or low compression (a bigger file for ADSL/broadband).

MPEG-7

This is another multimedia standard that enables the user to personalize and filter what they receive. It is ideal for web-based video, where it interacts with other media content.

How MPEG-4 and -2 operate

Let's now look at the most advanced way of compressing video to see what the most useful method does for your film.

To begin with, the codec looks at the information on several frames at a time, a group of pictures (GOP), in a process known as ‘interframe compression’. It does this by breaking up a sequence into I, B and P frames. I frames are ‘index’ frames and contain the entire information in a frame. The codec uses these as a reference to compare with the next few frames.

Even here the codec performs different levels of compression on different parts of the frame, so that the centre of a frame is compressed less than the outer, permitting about 15 per cent reduction in the size. Next, the B frames — ‘bidirectional frames’ — are noted. An analysis works out the differences between a B frame and the next or the one before. It discards any data that is repeated, keeping a code of where it was. It is like compressing a numerical sequence that runs 22222333366666 by expressing it as 253465 (the digit ‘2’ is followed by a digit representing the number of times ‘2’ appears, and so on). In this way, a 14-digit sequence has been reduced by half.

Finally, P frames are labelled as ‘predicted’ frames and reduce data by predicting where certain objects are going to move across a screen. Once there is significant movement or change — after a half second or so, perhaps — the whole process starts again with a new sequence being labelled I frames at the start, then B frames and so on. Thankfully you don't really need to know how this all works, but it does prove useful later when we look at DIY encoding.

The good codec guide

H.263 — use for video conferencing

This codec is a good choice for the business community. It's ideal for a video conference where you do not need high-quality pictures and where audio is going to take priority. It is best with low movement films, so a more or less stationary talking head is going to be just perfect. The data rate for this would be small, but much quality is lost. Perhaps the only use of this codec is if you wanted to send a quick version of a film for viewing by a co-worker; you may not worry too much about the way it looks as you just want quick feedback.

Cinepak — use for CD-ROMs

This is a good, well-established system which works best with small image sizes. This codec is outperformed by many others, even though its small picture size, at 120 × 90 pixels, is now getting bigger as computers improve. It is better used on low-end machines but is not usually the first option for web film, being more suited to CD-ROMs. A big advantage with this codec is that it allows the sender to customize settings throughout a film, so you can apply heavier compression to places where there is not much movement but lighter compression to where you need more detail on screen, perhaps during an action sequence. This process avoids ‘data spikes’, where sudden increases in data cause a movie to stop playing on the user's PC because too much data is needed.

RealG2 — good for web work, not for other uses

This codec is widely used on the web. It uses ‘temporal scaleability’, which means that the result for the user is smoother than others even on a wide range of computers or devices. This means movies encoded with it play at a high frame rate for fast processors and low for slow ones. This codec is hard to beat in terms of the number of users who may have access to it on the web and the ease with which the rest can download it (for free).

Sorenson — good all-rounder

This is a really good, high-quality codec and looks better than most at a screen size of 320 × 240. It's a good solution for movies that are going to be viewed over broadband connections, but some editions (notably the Developer Edition) cater for the other users by enabling scaleable streaming. One aspect which puts many people off using this codec is the length of time it takes to encode a movie in this way, but it remains the method of choice for most short movies, for the web and CD.

Intel Indeo 4 and 5 — good, but mainly for high power

With this method of compression you get a good result with high picture quality, but it is only viable for high-powered PCs or Macs. Version 5 allows for progressive downloads. Intel's codec is generally better than Cinepak, but cannot match the picture quality of the Sorenson.

Formats

There are various ways of decoding the information once it has been coded for sending over the Internet. Computer manufacturers have adopted different formats for codecs.

Video for Windows — short films and CD-ROM

This kind of file will have an ‘.avi’ suffix at the end. The advantage of this one is that it comes as standard with Windows operating systems. This format is primarily aimed at the CD-ROM movie, for short films stored and viewed on disc. Reliability is an issue here, as this format is notorious for putting audio and vision out of sync. The preferred codec for this format is Cinepak, which has widespread use but is poor in picture quality, but for better quality it can use Intel's Indeo system, version 4 or 5.

Microsoft Windows Media — good streaming tool, flexible

This kind of file has an ‘.ask’ after it and is designed to cope with almost any user, from the snail's pace 14.4 modem to the motorway that is ADSL. This format supports best the H.263, MPEG-4 and Intel's Indeo codecs such as Netshow. Netshow servers are designed to maintain speed of data delivery even if demand increases — what is known as true streaming. But it does so by reducing the quality of the images seen by the user, first visuals and then audio.

QuickTime — flexible, highly rated, free

This format is seen with an ‘.mov’ or ‘.qt’ after its files and is Apple Computer's answer to the format race. It is a flexible format, ideal as much for CD-ROMs as for DVDs or the Internet. Unlike earlier Windows formats, this also performs on Macs and best supports Sorenson, Indeo, Cinepak and MPEG-1. QuickTime uses RTSP, or Real Time Streaming Protocol, which means that it delivers the movie in real time, as does RealG2 and Netshow.

RealPlayer — free and very widely used

Real Networks Real Media (with an ‘.rm’ at the end of the file) format is by far the most popular on the market and has earned this degree of use by being accessible — it's free to download, as is QuickTime — but mostly by being able to alter its streaming level to suit the machine it is being viewed on. This means you don't have to create multiple versions of a movie for the user to choose from.

RealPlayer is a flexible format that copes well with content streamed live as part of a webcast or through a progressive download. The only negative point is that on low bit rates the film can start to look decidedly blurry and featureless, but compressing to this degree is to be avoided anyway.

Figure 2.6 How video streaming works.

DIY compression

Most filmmakers who use the web regularly for showing movies tend to customize their compression settings. Co-ordinators of online movie cinemas will compress what they show themselves so that they can apply the right codec for the movie. As we have seen above, certain codecs are better for some movies than others. It is clear that a filmed interview with someone talking to the camera will have different compression needs to a sports film where there is constant movement, but where sound is less crucial. Customizing your movie yourself is not as difficult as it sounds, and there are software programs that will use the popular ‘wizard’ process to take you slowly through the decisions you need to make.

Decisions

The first task in customizing your movie involves asking which aspect of the movie is most important. Is it:

• sound quality
• picture clarity
• picture smoothness.

Let's see how this affects a few different films. As we saw above, a sports movie would prioritize picture smoothness first, then clarity, with sound lowest of all. A music video would probably want all three, but would have to settle for sound first, then clarity, unless there was significant movement in the movie. A film trailer may go for picture clarity (so you can read what it is and when it is released), then smoothness (for the action) and then sound.

Compression is available either in pre-set forms, whereby you simply choose from a selection on the edit program and in a few moments have performed the task, or you can use an additional piece of compression software to do it for you, which is slower but more exacting and more responsive to individual needs. If you choose the latter you need first to know something about the process to see how to get the most for your films. In practice, there are occasions when you will be able to send a movie straight to the web, with no complications and little work involved on your part, but it is more likely that you may need a separate method for each film you make, depending on content, style and where you want to send it. This gives you more control over quality and how people view the film.

Encoding for the web

A good starting point is to use a Sorenson codec (part of Cleaner EZ) at 10 fps (frames per second), with a picture size of 160 × 120 pixels, with 8- or 16-bit sound.

If you are creating a copy on your website that is specifically for low-bandwidth users, aim to compress at smaller than the average dial-up modem of 56 kbps (kilobits per second) by going instead for around 35–40 kbps.

About data rates

We have talked a lot about the data rate, as it is one of the most important elements in compression, but there are some important facts to be aware of when talking about data. This is basic computer science stuff but has a potentially huge impact if you use the wrong terms. To begin with, there is a difference between a KiloByte (with K and B in capitals, represented as KB) and a kilobit (lower case, represented as k and kb). Multimedia and video tend to go with KiloBytes, where 1K is equal to 1024 bytes, but telecommunications tend to use bits, in which there are 1000 to the kilo. The byte consists of 8 bits, so you must adjust the way you calculate the downloading ability of a modem. For example, a 28.8k modem (notice the lower case) will provide 28.8 kilobits per second (kbps), but if we measure it in bytes we see that it is 3.5 KB per second when we divide it by 8 (the number of bits in a byte).

Working out how small a file needs to be

A useful method of working out what compression you need for delivery onto CD or DVD is:

Disc space (measured in KiloBytes (KB), not MB) divided by movie length (in seconds) = KB of final movie.

For example, a film which is 10 minutes long, squeezed onto a CD-ROM with less than the full disc used up, say, 450 MB: 450MB = 450 000 KB, divided by 600 seconds, equals 750 KB per second. This means that your data rate needs to be 750 KB per second to run safely on the CD.

The Crunch

• Compression is ideal for viewing a DV movie on the web or putting it onto CD or DVD
• Know what kind of movie each codec is best suited to
• Customize your own compression to suit your movies — use compression software
• View your movie when compressed to check quality
• You don't have to get involved with all the computer/data/technical stuff, just work out what aspect of the film is most important
• Compression is complicated — agreed — but it can help send your movie around more places.