I'd like to hear from anybody here who has experience cracking compression schemes.
I'm working on a video decoder for Digital Pictures' Sega CD and 32X games. So far I've been able to figure out most of the compression and coding techniques they used in their videos, but I've got one or two left that I can't quite crack.
One of them uses a series of code words that =< $0f. Each code word is followed by 32-bits of compression flags. A bit value of 0 indicates a literal byte that should be copied as is. AFAIK, if the code word is 1, a bit value of 1 indicates that the byte should be copied from the decompressed output (i.e. RLE). The compression flags are followed by literal bytes (if any), and then comes the next code word, ad nauseam. There are multiple code words (a total of 16 in theory), and I'm assuming that each has a different behavior (some might be LZSS instead of RLE, for example).
Has anybody here seen this type of compression before? Does it sound like a standard type? I'd appreciate any help you can offer me.