Interesting untested compression Program

[Gonzalo]

Have you ever heard about "PowerArc"??? "Rank" algorithm?

Here is a solution that says to have:

• Compression ratio more than 1.5-2 times better than ZIP/RAR archiver.
• Compression speed up to 10 Mb/sec (on PIII-800).
• Easy using as TStream descendant
• Built in data integrity check
• All library code is thread-safe (can be work in multithread application)

Unfortunately, the distribution is in the form of a library and I'm not able to write a back-end yet.

PowerZIP.dll
PowerRank.dll
PowerPPM.dll

So, if you think it is worth, I would *really* appreciate the opportunity of make some tests...

Thanks in advance!

[Gonzalo]

Exports Table

  Name:            PowerRank.dll
  Characteristics: 00000000
  TimeDateStamp:   00000000
  Version:         0.00
  Ordinal Base:    00000001
  # of Functions:  0000000A (10)
  # of Names:      0000000A (10)

  Entry Pt  Ordn  Name
  --------------------
  00001AF0     1  @Calloc@CAlloc_Free$qqrpv
  00001A7C     2  @Calloc@CAlloc_Get$qqri
  00001B0C     3  @Calloc@Finalization$qqrv
  00001B3C     4  @Calloc@initialization$qqrv
  000038EC     5  Compress
  000039BC     6  CompressMem
  00003A90     7  Decompress
  00001B44     8  GetPowerArcModuleInfo
  00001B4C     9  SetOptions
  000151D6    10  ___CPPdebugHook

[Skymmer]

There is nothing interesting there. RANK is nothing more than Symbol Ranking. Its even not a compression but the method of transformation.
http://mattmahoney.net/dc/dce.html#Section_54

[Gonzalo]

There is nothing interesting there. RANK is nothing more than Symbol Ranking. Its even not a compression but the method of transformation.
http://mattmahoney.net/dc/dce.html#Section_54

Simbol ranking 2x better than Rar? I don't think so...
I think the only thing these two have in common is the name.

[Skymmer]

Just go to htmldoc\index.html and read:

Rank Some idea of this algorithm describe in article Peter Fenwick "Symbol Ranking Text Compression with Shannon Recodings". At small compression the very fast and effective implementation is possible.

[Gonzalo]

Oh damn. You're right this time. I didn't look well enough. Lost interest now.
The ratio must be achieved using ppm.