What is the following type of compressed data file?

Answer 1

A compressed data file is a file where the data has been re-encoded in such a way as to reduce the overall length of the file (in bytes). Data compression is useful as it allows us to fit more data into a given space, which is useful for backup purposes (hence compression programs are also known as archive programs). Compression also allows us to transmit data across a network more efficiently. In addition, we can re-encode multiple files into a single compressed file known as a volume. Large volumes can also be split into a series of smaller volumes. This is useful when a single volume is too large to fit onto a DVD, for instance. Compression programs can also be integrated into the operating system shell program such that we can treat compressed volumes as easily as we can folders, allowing compression and decompression "on-the-fly".

There are two ways to compress data: with loss and without loss. Lossless compression means we can faithfully restore the original data from the re-encoded data. All data archiving programs use lossless compression. Lossy compression means the lost data can never be recovered unless we have a backup of the original source data.

Audio and video data can both make use of lossy compression (up to a point) because the human eye and ear are quite forgiving; we can remove massive amounts of redundant data before we begin to notice any difference. For instance, MP3 audio compression can reduce a 50 MB CD-quality audio file to around 5 MB without any appreciable loss in fidelity. The audiophile and "golden ear" community will always claim MP3 is inferior to CD (which it technically is), but extensive blind-testing under laboratory conditions has shown that they can't tell the difference.

Lossy compression can also benefit from lossless compression to a degree. Once the redundant data has been removed, a lossless compression algorithm can be applied to the remaining data. In the case of MP3, Huffman prefix encoding can be applied to each encoded audio frame.

Run Length Encoding, Huffman prefix encoding and arithmetic encoding are common examples of lossless encoding techniques. The more efficient lossless algorithms, such as LZW, are patented and require a licence to use in your own programs, however there are open source algorithms available, including the popular 7-Zip, although some restrictions still apply.