A common FAQ is how to convert analogue footage from older cameras into DVDs, through a computer with suitable editing software. (It is possible to also use an analogue input into a stand-alone DVD recorder, but with very limited editing possibilities.)
This post aims to provide a quick insight into how it can be done. The problem
can be divided into its component parts:
1. Analogue-to-digital conversion
4. Burning the DVD
However, before we explore these components, there is a general comment that must be made:
the quality of analogue tape footage is poorer than any full screen (704 or 720 by 480 or 576 pixels)
digital format (resolution and signal-to-noise ratio). This means that we have to conserve as much as
possible of the video data without further deterioration. It also means that going to extreme lengths,
such as high definition or uncompressed video is a waste of time and resources;
in fact, it may even exaggerate the blemishes of the analogue system.
Ulead/Intervideo/Corel do not endorse any products given as examples as being suitable for any purpose.
The examples given are just a small selection among many; please use a search engine to find others.
1. Analogue-to-digital conversion
There are many ways of doing this, but we shall discuss only the three most popular methods:
a) USB converters
b) IEEE-1349 converters
c) graphic cards
1a) USB converters
These consist of either a "black box" with a cable that plugs into the USB port or a "stick"
that does the same. The video and audio inputs are usually via composite RCA or Cinch plugs
and/or S/RGB with a jack for the audio. Either type may be powered from the computer, but the external
box type is sometimes powered separately.
The output format into the computer depends on the hardware but is most often in some form of MPEG-2 or,
in the latest versions, MPEG-4. Other formats are possible.
See below for a more detailed discussion of formats.
Many users have run into great difficulties when using USB converters.
This is because of two factors: either an old computer has a USB-1 port or, if it has a USB-2 port,
it has insufficient bandwidth. Generally, if the video converter is the only
device on the port,
there should not be too great a problem. However, most computers have USB keyboards, mice, webcams,
printers, scanners, digicams, external drives, memory sticks and other devices attached to the port.
Each device attached cuts down the available bandwidth. In addition 'handshake' and 'checksum' times for
each device cuts the available bandwidth further. The result is that the device buffer fills up before it
has a chance to pass the data to the computer and either crashes or you lose frames.
If you must use a USB device, make sure that it is the only one connected to the port.
Above all, it is almost always fatal to capture via a USB device to a USB hard drive.
Always use an internal ATA or SATA hard drive.
1b) IEEE-1394 converters
These are usually external "black boxes", similar to the USB ones, but they connect to an IEEE-1394 port
("Firewire (R)" or "i-Link (R)") and they may be either powered from the computer or externally.
There are also PCI cards. There is a wide variety of types (and prices) available, mono- and bi-directional,
with or without time-base correction, level correction, audio/video locking and other features.
They have one thing in common: the analogue input signal is converted into a DV stream,
in exactly the same format as from a DV or mini-DV camera. Some of the cheaper models do not offer the same
video quality as the better makes. Notwithstanding, the DV stream offers, by far, the most hassle-free system
with the possibility of excellent quality and reliability from good converters.
Note that some DV camcorders with analogue-in and pass-through features may also be used, in place of a converter. Depending on the model, this may be more-or-less easy or tricky; in some cases, it may entail recording the DV on tape and, with some other camcorders, the pass-through will work only if a tape is installed, even though it is not used. However, a purpose-built converter is probably to be recommended.
1c) Graphics cards
This subject is opening a can of worms because of all the possibilities.
Some cards are dedicated video input/output cards, some with TV tuners, other cards of this type,
today, have the same functions on the same card as is connected to the computer monitor.
Some of them have hardware and some software conversion to the digital format. In either case,
it is done through a 'codec' (the USB and IEEE-1394 converters have codecs in their hardware,
but you don't need to know much about them!). The codec or co
ompressor is what
determines the digital format. Hardware codecs usually permit conversion only to one type of format,
such as MPEG-2 or MJPEG, while software codec systems may offer a really wide choice from uncompressed,
to lossless compressors, such as HUFFYuv, to lossy compressors such as MPEG-1. What this means is that the
quality ranges from extremely good to mediocre, as the file sizes range from enormous to small.
This is usually a compromise and MJPEG is often a good choice for subsequent editing of captured
analogue from tape, if DV is unavailable. One advantage of this system is that one can choose a wider
range of parameters with these systems, but this may appear daunting to beginners (and even to pros, at times!).
Because there are no adjustments, by far the easiest capture system is DV. The incoming DV stream is saved
directly into an AVI file, which is copied onto a hard disk. Both MediaStudio Pro
do this admirably. Because there is no format conversion this generally passes without any hitches.
These capture applications also permit transcoding into an MPEG format, but this is not recommended
because the quality may be poorer than can be obtained from an AVI file, particularly if the project is
going to be edited before encoding onto a DVD.
In order to capture with any other system, there is generally a wide choice of possibilities,
such as the format, colour depth, bitrate, codec and so on. Most beginners (and many not-so-beginners)
make the fatal mistake of setting the quality ratings too high. If it doesn't cause the computer to crash,
this can produce very large files which fill up the hard disk unnecessarily. Depending on the system,
always choose a frame size of 704/720 by 480/576 pixels at the appropriate frame rate of 29.97 fps for NTSC
or 25 fps for PAL. From there, start off with trials using a medium quality setting. As stated above,
if there is a choice of codec, the MJPEG is a good place to start. Unfortunately,
many hardware devices allow no choice, especially MPEG types.
In this case, choose MPEG-2 at the above frame size and rate, but there remains a problem;
what bitrate should be chosen? Assuming that we are capturing from camera tapes, these last rarely more
than one hour and a bit rate of 6000 kbit/s for the video is suggested. With many systems, there is no or
little choice regarding the audio format. As we shall see later the ideal format is AC-3, also known as
Dolby Digital. However, this is not always available during the capture, in which case simply choose the
default settings and we shall see how they can be modified later on.
Most digital video formats are more or less lossy. This is because the video is compressed. Uncompressed
video is possible but would be very stupid for this application because the file sizes would be simply
enormous and the input quality from an analogue tape is never very good, so nothing would be gained.
Some systems, such as DV, compress each frame individually, whereas other systems, such as MPEG,
compress only one frame in 12 or 15 and then record only the differences between these frames and the
following 11 or 14. The fully compressed frame is called an I-frame and the intermediate frames are
called P-frame or B-frame, depending on whether the difference is after or before an I-frame.
This sounds unduly technical but there is a good reason to mention it: if we subsequently make a cut or
transition some distance from an I-frame, valuable video data may be lost, causing a distinct
deterioration of quality, especially if any special effects are subsequently added.
This is why, if the choice is available, it is preferable to use a codec, such as DV or MJPEG,
with only I-frames. While on the subject, it is better to avoid complex compressors designed only for
distribution, such as WMV, DivX and XviD. These may give much smaller file sizes, but the quality may
be poorer and the encoding times to MPEG-2 will be prohibitive.
Before transferring your file to the timeline, the most essential thing is to ensure that the properties
or settings correspond exactly to those of the file itself. In other words, if, for example, your file is
DV type 1, then your settings should be DV type 1. This is even more essential when you import an MPEG file
into the timeline. Some editing software, including MediaStudio Pro
, has a feature called Smart Render.
This allows you to not have to re-render unmodified parts of the video, thus ensuring maximum quality of
the new file, but this can only work if the settings correspond exactly to those of the capture.
When working with MPEG or any other highly compressed video formats, try to keep the effects to a minimum,
in order to preserve the quality. In the less compressed video formats, such as MJPEG or DV, this does not
matter because re-rendering or encoding does not significantly affect the quality.
It is strongly recommended to create a new file in the same format as the settings, once all the editing has
been finished. At the very least, it is a safeguard against problems occurring at a later stage.
This is in despite of the fact that some systems may allow you to author DVD straight from the timeline.
4. Burning the DVD
Unless the file created in the last section is already DVD-compatible, it is necessary to create a
DVD-compliant one. This can be done either in the editor or in the authoring program.
It is an error to create a DVD-compliant MPEG-2 file at too high a bitrate, as explained in
. When creating a DVD from an analogue tape,
it is doubly ridiculous to use bitrates where the resultant quality would be significantly higher
than the original video. Anything with a video bitrate higher than 6000 kbit/s, constant bitrate,
would be inappropriate. In reality, it is probable that little or no difference would be discernible
when viewing on an ordinary television with bitrates as low as 4000 kbit/s,
when the original was a VHS or Hi-8 tape.
There is another question which must be raised, concerning the audio bitrate. Wherever possible,
it is recommended that the MPEG-2 file be created using the AC-3 (Dolby Digital) format.
This will be in stereo (2/0) format and a bitrate of 192 kbit/s is more than sufficient.
One word of warning: many of the older VHS-C and Hi-8 camcorders recorded the sound in mono, rather than stereo.
If this is the case, it may be that the playback will be on one stereo channel, usually the left.
Use an audio editor to correct this before generating the DVD-compliant file.
If the project has been in DVD-compliant MPEG-2 format from the start, then it should be kept that way.
However, some editing and authoring programs allow the audio format to be changed, for example to AC-3,
without the video having to be re-rendered.
To complete the authoring process, follow the instructions of the programme being used.
If keeping within the Ulead/InterVideo/Corel software, there is a choice of
and DVD Workshop
The latter is a very high end and versatile but is unfortunately obsolete.