.VQA

From Wikipedia, the free encyclopedia

Vector Quantized Animation
File name extension	`.vqa`
Type code	'WVQA'
Developed by	Westwood Studios
Container for	Audio, Video

Vector Quantized Animation, known by its acronym VQA is a file format originally developed by Westwood Studios for video encoding in their game The Legend of Kyrandia.

1 Description
2 Versions
3 Encoding / Decoding
4 External links

[edit] Description

The VQA format was used in every Westwood game released between 1995 and 2000. Each VQA file is comprised of a series of chunks aligned in a tree structure, beginning with the chunk FORM, which is the parent of every other chunk in the file. Every data is in little endian. Audio in VQA files are in the .AUD file format.

[edit] Versions

[edit] Version 1

Version 1 was used in the Kyrandia series. It allowed the use of 256 colors and a maximum resolution of 640*480 (but most videos were encoded in 320*156). Version 1 VQAs must always use 8-bit sound. A frame rate set to higher than 10 renders the VQA unplayable to Legends of Kyrandia. Setting any value in the header chunk to zero makes the video use the default value.

[edit] Version 2 (C&C VQA)

Version 2 was used in Command & Conquer, it's prequel Command & Conquer: Red Alert, Dune 2000 and Lands of Lore 2. The main changes were the introduction of 16-bit sound, frame rates higher than 15 and the removal of default values (using 0 in the header chunk results in unplayable video).

[edit] Version 3 (15 bit Hi Color VQA)

The biggest addition of Version 3 is the support of videos with 32,768 colors. The maximum value of the colors entry in the header chunk is still 256 but setting it to 0 will default to HiColor mode. Another modification is that if the first byte of the chunk is set to 0×00 then the frame is not encoded using the standard Format80 algorithm.

[edit] Encoding / Decoding

There is only one known encoder for VQA files except the encoder used by Westwood.

The video codec VQA files use is very asymmetrical meaning the encoding process is very time consuming whereas the decoding algorithm is among the simplest and fastest out there.