Video Capture, Transcoding, and Authoring/Capture (Hardware)
ATTN: Very outdated information! This book section needed to be updated.
Analog Standard Definition Capture[edit | edit source]
This section describes ways to capture standard definition video from an analog source. This includes component and S-Video connections, and also television tuners.
Using a DVD or DVD-RAM Recorder[edit | edit source]
One straightforward and reliable way to capture video is via a consumer DVD or DVD-RAM recorder. DVD recorders are inexpensive (US$200 or less) and easy to use. Connect the recorder to your signal (composite video, S-Video, or TV input), and record onto blank media. When you are done, you have DVD media with your content on it in MPEG-2 format.
Perhaps the most significant advantage of using a DVD recorder is that the files you obtain will likely have perfect A/V sync. Also, in some cases, if your desired output is an MPEG-2 stream, you can obtain satisfactory results by editing the output of the recorder and authoring with that, without ever transcoding.
Another advantage is that the MPEG-2 video from the recorder does not take a large amount of space. High-quality MPEG-2 encoded video is 1/5 or 1/10 the size of HuffYUV encoded video. A couple of hours of source material will fit in 5-10GB rather than 50GB.
The MPEG-2 stream on the recorded media will be interlaced. If the source material was a telecined film, the recording will be a hard telecine; it will not be inverse telecined into a progressive source by the recorder. This may be perfectly acceptable to a viewer; after all, it is exactly what a viewer sees when a film is played over a standard television.
In cases where you want to do filtering or extensive editing, or change the format of the video (including inverse telecine), transcoding will be required. There is a potential loss in quality as a result, because the video will pass through a lossy MPEG-2 encoding/decoding phase.
- Signal → MPEG-2 encoding (DVD-R) → MPEG-2 decoding (computer)
→ Editing (computer) → MPEG-2 encoding (computer) → Result
Depending on the signal and the eventual use of the result, the loss in quality due to an additional transcoding step may or may not be significant. Clean analog signals tend to look best (e.g. a LaserDisc). Signals that have already been through one transcoding phase (e.g. the composite output from a digital satellite tuner) generally look worse. DVD recorders allow a variety of quality settings, up to an 8000+ kilobit/s high quality mode that records one hour of video on a single 4.7 GB DVD. However, because DVD recorders encode MPEG in real time, the encoding is not as good as a careful 2-pass encoding at the same bit rate done off-line on a computer.
Certain types of filters, applied judiciously during transcoding, can improve the results slightly. Some video noise removal (both static and temporal), deblocking, and/or "smart" sharpening can be particularly helpful.
If you have a DVD recorder, it is worthwhile to give it a try. The ease of use and the reliable A/V sync may outweigh the disadvantages of the MPEG-2 encoding of the output.
Using a Format Converter[edit | edit source]
A format converter is a device that converts video signals of one type to another in real time. Typically this includes analog to digital (A/D) and digital to analog (D/A) conversion. These are generally professional-level devices that are moderately priced. An example is the Datavideo DAC-15, which at a 2007 price of US$750 can convert between the following standard definition signals:
- Composite video
- Component video (Y, R-Y, B-Y)
- Y/C (S-Video)
- Unbalanced stereo audio
- DV (IEEE-1394 aka FireWire)
This can allow transcoding of an analog signal to DV for direct capture by a computer. This is generally superior to direct capture of an analog signal by computer, because less CPU power is required, and A/V sync is likely to be good.
Some digital camcorders that accept analog inputs can effectively function as real time format converters. However, consumer-level camcorders do not accept component video input.
High definition format converters are also available, but are expensive. See below.