Can someone describe the difference between these compression models?

[MAME]

I have been looking online but they both seem do to the same thing based on the limited descriptions I've seen of both.

Does anyone know for sure about how they differ in implementation?

 

[KLin]

http://forums.anandtech.com/categories.aspx?catid=50&flcache=2131421&entercat=y

or

http://forums.anandtech.com/categories.aspx?catid=33&flcache=2131421&entercat=y

 

[MAME]

Highly Technical
Check your egos at the door. This forum is for those who are serious about hardware. ON TOPIC discussions about PC architecture, technology and in-depth looks at what we all know and love. NO FLAME WARS. Keep the discussions truly technical.

and it's not software, it's just an alogrithm

 

[Goosemaster]

One is based on type c and the other is based on markov's type d

 

[Goosemaster]

Appearantly Markhov's isn't good with long blocks of data

 

[MAME]

Originally posted by: Goosemaster
One is based on type c and the other is based on markov's type d

ok, but anything more specific than that?

for example, I read that ppmd increments the escape symbol for every symbol not found in that context

however, I also read that ppmc increments the esccape symbol for every other symbol already in the context

these end up with the exact same ith-order context tables, and thus I'm confused about any real difference.

I'm guessing they simply didn't go in depth enough to show the differences

 

[MAME]

w00t!
http://www1.cs.columbia.edu/~rambow/mas...n/PDF-ONECOLUMN/05-JinHuHuang-1col.pdf