Author: Dan Andersson
Date: 03:18:39 03/27/02
Go up one level in this thread
In the scheme you are proposing the assumption should be that all the exact values in the skip array are equally likely to be a skip value. The easy way of ensuring a compact form is to make a Huffman encoding based on the combined alphabet. Then you will have a good compression. This compression should in theory have a better efficiency than the skip index and a literal. If you are constrained by the 32 slot array, or any size for that matter. Then you construct an encoding that contain one or a few escape values. And then you make a new Huffman encoding on the values that go to those values. But it's hard to tell what scheme that is best/fastest/acceptable/doable without the actual data. It would be a nice project finding such a compression, by the way. OTOH, I could be completely wrong. Since I'm running a severe cold at the moment. MvH Dan Andersson
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.