Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.compression > #366
| Newsgroups | comp.compression |
|---|---|
| From | stan <smoore@exis.net> |
| Subject | Re: I want to replace the logic |
| References | (2 earlier) <915bc4fa-6d0e-429e-a648-87127a59eab3@j25g2000vbr.googlegroups.com> <05fdd6b2-fef5-49cc-9342-309c17b5904b@v10g2000yqn.googlegroups.com> <4e070c49$0$307$14726298@news.sunsite.dk> <iuavd8$mmp$1@news.albasani.net> <65ddfae3-97ca-40bc-a7fc-ab544b08447f@gv8g2000vbb.googlegroups.com> |
| Message-ID | <rj2se8-bo9.ln1@invalid.net> (permalink) |
| Organization | TeraNews.com |
| Date | 2011-07-11 16:21 -0400 |
jules Gilbert wrote: <snip> > compressed file, block it into buffers, compress the buffers, and > *now* emit a single memory vector, in fact a byte-wide vector. Which > becomes the target of the subsequent pass. Out of curiosity, are you a native English speaker? What exactly do you mean by "single memory vector"? Those are all valid words individually, but put them together into a phrase and you get meaningless jargon. Are you talking about an undetermined length of contiguous bytes stored in memory? Possibly an array to use terminology from many programming languages and a well known term? > Now this approach has problems, For those who insist Jules doesn't know what he's talking about, I offer the above as counterpoint. <snipping story of Jules mythological magic recursive machine> > But it does take a large file, crunch it, and produce a file below 8k > bytes. > > Since everyone always asks "Can you reconstruct?" You find this question surprising? > I'll say (I think it's obvious from the above,) that no, I don't have > a completed program that runs and re-processes the memory vector. Let me get this straight. Here you say no I can't reverse the process. > Although I have written the decompressor for this method and it > works, so far this is only reconstructing from a malloc'ed block. And here you say you can "decompress" stuff stored in memory. Where exactly does the "single memory vector" exist? > (And let me tell you -- when you first see those pairs of values, > well, I think people here know what I'm talking about.) If the betting windows are open, I'd like to put buttloads on "nobody knows what you are talking about" > And if you're new here what I'm saying is this: All files, every > single file, including packages of files (such as a .tar file > contains,) can be represented in 8k bytes. > > And yes, though I've never done it, I'm pretty sure I can compress and > reconstruct more than 2 ^ 64k files. So you've repeated; very, very often. Do you still claim this doesn't violate the counting argument? If I give you a single file of bytes, how many different files will your program generate?
Back to comp.compression | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
I want to replace the logic jules Gilbert <jules.stocks@gmail.com> - 2011-06-23 16:13 -0700
Re: I want to replace the logic Noob <root@127.0.0.1> - 2011-06-24 10:40 +0200
Re: I want to replace the logic jules Gilbert <jules.stocks@gmail.com> - 2011-06-24 06:04 -0700
Re: I want to replace the logic Earl_Colby_Pottinger <earlcolby.pottinger@sympatico.ca> - 2011-06-24 21:32 -0700
Re: I want to replace the logic Mark Nelson <snorkelman@gmail.com> - 2011-06-25 11:44 -0700
Re: I want to replace the logic Captain Obvious <alex.mizrahi@gmail.com> - 2011-06-26 13:39 +0300
Choosing between GPL and LGPL Noob <root@127.0.0.1> - 2011-06-27 11:31 +0200
Re: I want to replace the logic BGB <cr88192@hotmail.com> - 2011-06-27 15:11 -0700
Re: I want to replace the logic jules Gilbert <jules.stocks@gmail.com> - 2011-07-10 14:22 -0700
Re: I want to replace the logic Sebastian <s.gesemann@gmail.com> - 2011-07-10 23:39 -0700
Re: I want to replace the logic stan <smoore@exis.net> - 2011-07-11 16:21 -0400
Re: I want to replace the logic Jim Leonard <mobygamer@gmail.com> - 2011-07-12 08:22 -0700
Re: I want to replace the logic jules Gilbert <jules.stocks@gmail.com> - 2011-06-26 10:38 -0700
csiph-web