Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.compression > #2820 > unrolled thread

My File Compression Technique: *Some* Progress!

Started byHarry Potter <rose.joseph12@yahoo.com>
First post2015-01-26 05:03 -0800
Last post2015-02-05 08:10 -0800
Articles 6 — 2 participants

Back to article view | Back to comp.compression


Contents

  My File Compression Technique: *Some* Progress! Harry Potter <rose.joseph12@yahoo.com> - 2015-01-26 05:03 -0800
    Re: My File Compression Technique: *Some* Progress! Hans-Peter Diettrich <DrDiettrich1@aol.com> - 2015-01-26 14:42 +0100
      Re: My File Compression Technique: *Some* Progress! Harry Potter <rose.joseph12@yahoo.com> - 2015-01-26 05:55 -0800
        Re: My File Compression Technique: *Some* Progress! Harry Potter <rose.joseph12@yahoo.com> - 2015-01-26 13:39 -0800
          Re: My File Compression Technique: *Some* Progress! Harry Potter <rose.joseph12@yahoo.com> - 2015-01-27 06:51 -0800
            Re: My File Compression Technique: *Some* Progress! Harry Potter <rose.joseph12@yahoo.com> - 2015-02-05 08:10 -0800

#2820 — My File Compression Technique: *Some* Progress!

FromHarry Potter <rose.joseph12@yahoo.com>
Date2015-01-26 05:03 -0800
SubjectMy File Compression Technique: *Some* Progress!
Message-ID<5b93d8b9-1161-4aea-a73d-eae444ae7e83@googlegroups.com>
Okay.  Right now, my compression technique seems to be doing better on *some* files but worse on most.  :(  I have more work to do.

[toc] | [next] | [standalone]


#2822

FromHans-Peter Diettrich <DrDiettrich1@aol.com>
Date2015-01-26 14:42 +0100
Message-ID<cimui9Fn21vU1@mid.individual.net>
In reply to#2820
Harry Potter schrieb:
> Okay.  Right now, my compression technique seems to be doing better
> on *some* files but worse on most.  :(

That's normal, cannot be avoided. And that's why zip and 7zip support
multiple compression methods, and select the best one for every file.

DoDi

[toc] | [prev] | [next] | [standalone]


#2823

FromHarry Potter <rose.joseph12@yahoo.com>
Date2015-01-26 05:55 -0800
Message-ID<a7d5cea4-4cac-4b50-b45d-a2cbb9259205@googlegroups.com>
In reply to#2822
On Monday, January 26, 2015 at 8:42:35 AM UTC-5, Hans-Peter Diettrich wrote:
> That's normal, cannot be avoided. And that's why zip and 7zip support
> multiple compression methods, and select the best one for every file.
> 
> DoDi

You're right!  Some of the techniques I tried seem to help on some files and hurt on others.  :(

[toc] | [prev] | [next] | [standalone]


#2825

FromHarry Potter <rose.joseph12@yahoo.com>
Date2015-01-26 13:39 -0800
Message-ID<7303c867-f5f3-40b5-8467-d3bc38463d47@googlegroups.com>
In reply to#2823
On Monday, January 26, 2015 at 8:55:53 AM UTC-5, Harry Potter wrote:
> On Monday, January 26, 2015 at 8:42:35 AM UTC-5, Hans-Peter Diettrich wrote:
> > That's normal, cannot be avoided. And that's why zip and 7zip support
> > multiple compression methods, and select the best one for every file.
> > 
> > DoDi
> 
> You're right!  Some of the techniques I tried seem to help on some files and hurt on others.  :(

More news: after the last post here, I got my technique to perform better on five files and worse on two over the Deflate technique.  Not only that, but the difference was considerable: around or a little more than 10% on two files!  :D  Now, this sounds too good to be true.  I will do one last idea then debug and see for sure.

[toc] | [prev] | [next] | [standalone]


#2827

FromHarry Potter <rose.joseph12@yahoo.com>
Date2015-01-27 06:51 -0800
Message-ID<3f1ea368-3088-4b33-b257-622717d97fe2@googlegroups.com>
In reply to#2825
On Monday, January 26, 2015 at 4:39:06 PM UTC-5, Harry Potter wrote:
> On Monday, January 26, 2015 at 8:55:53 AM UTC-5, Harry Potter wrote:
> > On Monday, January 26, 2015 at 8:42:35 AM UTC-5, Hans-Peter Diettrich wrote:
> > > That's normal, cannot be avoided. And that's why zip and 7zip support
> > > multiple compression methods, and select the best one for every file.
> > > 
> > > DoDi
> > 
> > You're right!  Some of the techniques I tried seem to help on some files and hurt on others.  :(
> 
> More news: after the last post here, I got my technique to perform better on five files and worse on two over the Deflate technique.  Not only that, but the difference was considerable: around or a little more than 10% on two files!  :D  Now, this sounds too good to be true.  I will do one last idea then debug and see for sure.

I did some more tweaks this morning, and the software is working slightly better than last night.  :)

[toc] | [prev] | [next] | [standalone]


#2845

FromHarry Potter <rose.joseph12@yahoo.com>
Date2015-02-05 08:10 -0800
Message-ID<e7261a5d-a9d7-46aa-9558-fb3027a8de02@googlegroups.com>
In reply to#2827
On Tuesday, January 27, 2015 at 9:51:29 AM UTC-5, Harry Potter wrote:
> I did some more tweaks this morning, and the software is working slightly better than last night.  :)

It's working slightly better now.  I'm trying to minimize the Huffman overhead (>120 bytes on some files).  Once I do that, I need to finish and start debugging the Huffman codes and inch my way into my other techniques.

[toc] | [prev] | [standalone]


Back to top | Article view | comp.compression


csiph-web