[comp.binaries.ibm.pc.d] maximal compression utility sought

sophist@brainiac.raidernet.com (sophist) (12/21/90)

I have searched through various archive indices and have not seen
anything resembling this:

What I am after is a utility which offers MAXIMAL compression (i.e.,
significant improvement over 'compress', 'zip', or 'lzh') while
operating at a much slower speed.  Of course a fast utility like this
would be nice, but the point is that speed is not an issue.  Maximal
compression is.  Does such a program exist?  Is it even possible to
compress files much more than zip or lzh?  What I'd like to compress is
binary executables for long-term storage (archival purposes).  Perhaps
some text as well.  Any ideas?

Regards,

davidsen@sixhub.UUCP (Wm E. Davidsen Jr) (12/24/90)

In article <8DNHu3w163w@brainiac.raidernet.com> sophist@brainiac.raidernet.com (sophist) writes:

|                                                              Maximal
| compression is.  Does such a program exist?  Is it even possible to
| compress files much more than zip or lzh?  What I'd like to compress is
| binary executables for long-term storage (archival purposes).  Perhaps
| some text as well.  Any ideas?

  I have not seem any indication that such a program exists. The beta
test version of lharc does a little better (and is a little faster,
too), I haven't tried the new compressor from simtel, although I will
over the next week (I don't think it runs in unix, anyway), and the next
release of zoo will also be a little better. I don't see anything which
would be enough better to justify changing now, although I did stuble on
an idea at a seminar which I'm going to test, which might be 20-30%
smaller.

  The programs currently in use are doing a pretty good job, although I
would like something better for images. I have some 24 bit 600 dpi
images which I want to store without any loss of data, and existing
compression just doesn't cut it.
-- 
bill davidsen - davidsen@sixhub.uucp (uunet!crdgw1!sixhub!davidsen)
    sysop *IX BBS and Public Access UNIX
    moderator of comp.binaries.ibm.pc and 80386 mailing list
"Stupidity, like virtue, is its own reward" -me

ts@uwasa.fi (Timo Salmi) (12/24/90)

In article <8DNHu3w163w@brainiac.raidernet.com> sophist@brainiac.raidernet.com (sophist) writes:
>I have searched through various archive indices and have not seen
>anything resembling this:
>
>What I am after is a utility which offers MAXIMAL compression (i.e.,
>significant improvement over 'compress', 'zip', or 'lzh') while
>operating at a much slower speed.  Of course a fast utility like this
>would be nice, but the point is that speed is not an issue.  Maximal
>compression is.  Does such a program exist?  Is it even possible to
>compress files much more than zip or lzh?  What I'd like to compress is
>binary executables for long-term storage (archival purposes).  Perhaps
>some text as well.  Any ideas?

I don't think that the differences between the best archivers are
not significant nor quite consistent enough for the time being for
this idea.  Rather it is the other issues that are significant at
the moment in selecting a compressor.  Since this subject has been
discussed so often, I am not going to try to recount what has
already been said.  
   The only idea that comes to mind is that there is a new
compressor around which _claims_ better copmpression that zip and
lharc, the current leaders in that respect.  It is called
/pc/pd2/arj015a.exe still at a beta test phase.  You might take a
look.  You can obtain it by anonymous ftp from uwasa.fi archives. 
   I haven't tested arj enough to judge if its claim is warranted. 
I'll leave that to others who have a better motivation to look for
the best compression ratios in archivers.

...................................................................
Prof. Timo Salmi        (Moderating at anon. ftp site 128.214.12.3)
School of Business Studies, University of Vaasa, SF-65101, Finland
Internet: ts@chyde.uwasa.fi Funet: gado::salmi Bitnet: salmi@finfun

Peter_Gutmann@kcbbs.gen.nz (Peter Gutmann) (12/27/90)

I'm currently working on an archiver which which is supposed to outperform 
but ZIP and LHARC (and it does for all test data I can get my hands 
on - about 10MB of stuff).  If you aren't worried about SLOW compression 
you'll LOVE this program - it crawls :-).  Actually its not *that* bad 
(1/2 the speed of LHARC, 1/3 the speed of ZIP) and I hope to speed it 
up somewhat by rewriting the compressor in assembly language.  As to 
its compression, I won't make any claims here but it's not that much 
better than ZIP and LHARC.  In fact as far as compressors go ZIP and 
LHARC are pretty well as good as you can get - there are better compression 
schemes but they tend to need a Cray to run on.  In short you can (as 
fasr as I know) get slightly better compression than ZIP/LHARC at the 
cost of slow compression/decompression.

It is possible to do slightly better than ZIP and LHARC but if you want
significantly better performance you'll probably have to implement the code
yourself, and run it on a LARGE system......
 
Peter.

steven@ozonebbs.UUCP (Steven Rubin) (12/29/90)

Peter_Gutmann@kcbbs.gen.nz (Peter Gutmann) writes:

> I'm currently working on an archiver which which is supposed to outperform 
> but ZIP and LHARC (and it does for all test data I can get my hands 
> on - about 10MB of stuff).  If you aren't worried about SLOW compression 
> you'll LOVE this program - it crawls :-).  Actually its not *that* bad 
> (1/2 the speed of LHARC, 1/3 the speed of ZIP) and I hope to speed it 
> up somewhat by rewriting the compressor in assembly language.  As to 
> its compression, I won't make any claims here but it's not that much 
> better than ZIP and LHARC.  In fact as far as compressors go ZIP and 
> LHARC are pretty well as good as you can get - there are better compression 
> schemes but they tend to need a Cray to run on.  In short you can (as 
> fasr as I know) get slightly better compression than ZIP/LHARC at the 
> cost of slow compression/decompression.
> 
> It is possible to do slightly better than ZIP and LHARC but if you want
> significantly better performance you'll probably have to implement the code
> yourself, and run it on a LARGE system......
>  
> Peter.

As a side note, I received the new ARC+ from Sea a while ago.  Suprisingly, 
it does create smaller files then PKZIP, But ARC+ isnt Shareware, and with 
SEA's good atitude, I wouldn't Buy anything from them!  



---
Steven Rubin                                           @      @
{netcom, crash!nusdecs}!nusjecs!ozonebbs!steven           oo
Disclaimer: I don't even speak for myself!           \__________/