|
|
Atari coding BBS
| Re: LZ78 packer |
|---|
|
Posted by: ray
|
Feb,23.2006-13:20
|
to RA: ...ok, that's cool to hear, so i will try to release it soon.
to ggn: unlike in entropy coders, that try to shorten word length the LZ group of algorithms perform well on repetetive string patterns (as those will be encoded by a reference to a fixed wordbook which is build during the decode - (LZ78) using variable field lengths or a reference to yet decoded data (LZ77) using a sliding window of a constant size, so compression will occur).
as leonard already pointed out, combined packers like RLE+Huffman, LZxx+Huffman etc. will have a better ratio in general while they are much slower in depacking data (Huffman basically needs to treat the input stream bitwise, unless implemented with a smart solution which explains that).
regards,
ray
|
[All messages in this thread] [Start new thread]
|
Topic
|
Posted by
|
Date
|
|
LZ78 packer
|
ray
|
Feb,22.2006-18:24
|
|
Re: LZ78 packer
|
ggn/KUA
|
Feb,22.2006-22:37
|
|
Re: LZ78 packer
|
Leonard
|
Feb,23.2006-02:44
|
|
Re: LZ78 packer
|
RA/paradox
|
Feb,23.2006-00:19
|
|
Re: LZ78 packer
|
ray
|
Feb,23.2006-13:20
|
|
Re: LZ78 packer
|
tobe
|
Feb,23.2006-14:33
|
|
Re: LZ78 packer
|
Simon Sunnyboy / Paradize
|
Feb,23.2006-17:19
|
|
Re: LZ78 packer
|
ray
|
Feb,23.2006-18:55
|
|
Re: LZ78 packer
|
Simon Sunnyboy / Paradize
|
Feb,24.2006-10:10
|
|
Re: LZ78 packer
|
ggn/KUA
|
Feb,24.2006-20:12
|
What's the anti-troll code?
That's your personal code to be able to add comments and messages on the dhs.nu site.
Don't have a code or forgot it? Fix it here.
|
