I''ve read all 17 pages of this, and I think you guys need to realize that this kieren_j guy is lying. You guys are all probably disappointed; but don''t despair! I''ve come up with a new algorithm that actually works!

My algorithm has several advantages. First, it can compress ANY file (even random data) at a 1/10 ratio. Second, although you need to use two different programs to compress it, you DON''T need a decompressor! Here''s a description of how it works.

The first program "prepares" the file. It adds random data at the end of the file until it is 10 times as long.

The second program gets rid of the random data. The user looks at the file size of the old file, and the file size of the compressed version, and is satisfied to know that it is 1/10th of the size.

And best of all, the user doesn''t need a decompressor program to decompress it!

I called the US Patent Office to apply for a patent. However, the guy on the phone said something about there being a crack in my head, or something to that effect, then hung up. What a fool.

Please post comments about my awesome compression algorithm!

- Lava

I dont think kieren_j is LIENG, I think he just figured out his compression wont work exactly how he thought or something. What would be the reason to LIE about it? To look like an idiot, I dont think anyone wants that, just think about, hes not LIEING, just think, he might be wrong though, but I think I understand his Bit(or did he do byte?) rearranging and how that could definitly help ALOT in compression.

to all who still believe in kieren_j !

i contacted him. He said there will be a demo soon, and it will rock !

stay tuned !

quote: Original post by Anonymous Poster

people who have been in orbit (say, on a space station) are technically "time travelers". They are a few milliseconds younger than if they would have stayed on earth.

They''ve also took 2 very accurate clocks, flown one around the world very fast, and left one on earth. The one on earth was a (very small) fraction of a second faster, its true.

Im not going to say anything about compression, sorry. its just not my area to talk (heh like time travel is). none the less I''ve found this thread interesting....I hope it doesnt get killed.....

People in orbit are not accelerated because acceleration is relative to the gravitational field. We are accelerated on earth with 9.81 m/s^2. During the start and the landing time goes slower in the space ship, but while it''s in an orbit time goes faster than on earth.

Lack: I''m examining an arj (counting bit patterns) on my old computer with one of my old programs. And I''ll tell you if there are big differences to random data (I don''t exspect).

Visit our homepage: www.rarebyte.de.st

GA

*Sigh* since this thread refuses to die and no-one will post in my "compression" thread, I might as well post here.

Irritable - Wavelets are not exactly fractal compression. I''ve seen a bit of theory about them, and they are very interesting. Wavelets are a form of encoding, and it doesn''t have to be lossy at all. It reorganises data in a way so that its detail is hierarchical, up to the total detail available.

I should dig up my papers on the Haar basis and show you what I mean, but I''m lazy. The trick to wavelet stuff is, that using the reorganisation, you could get rid of some of the higher "frequencies" of data, thereby reducing file size, but not losing too much of the original data density.

Wavelets are useful for a whole bunch of other things too though.


#pragma DWIM // Do What I Mean!
~ Mad Keith ~
**I use Software Mode**

quote: Original post by Vetinari
I thought this was well explained. It''s not ''random'' as such, but it might as well be. By saying it''s random we say that it does not contain enough patterns to be compressable. This is contained in the same proof I mentioned earlier about Huffman encoding being optimal. Let me make sure there is no misunderstanding this: There are not enough patterns in a compresssed file to compress it further. This is not my ''opinion''.

But then:

quote:
I agree that there is an off chance that there could be more patterns in a specific bit window than another. I don''t agree that it makes any difference as to how well it would be compressed.

So you say, looking at the file with a different byte/window size can create an imbalance in patterns. But just above you said they didn''t have enough patterns to compress. And if you have extra repetitions of something, how do you not agree that it makes a difference in how well it can be compressed?

quote:

Well, you did and continue to. Your attitude seems to be ''I have never worked with compression before, but I will talk like I have.''

Sorry, I don''t mean to come across that way.

quote:

And you agreed earlier that no one algorithm can compress all files of length n, so your only hope now is this ''data massasging'', which I still have yet to hear any way of it to be implemented. That is, unless data massaging and windowing are the same thing.

Of course not. But one can be used inside the other, also.

quote:

I said you could find the proof in any standard algorithm textbook, and that it is too long and complicated to put here. You even said you would like to study it.

Fair enough, but for now (since it hasn''t been proven to me otherwise), I think it''s possible with data massage and such.

quote:
I am awaiting the proof behind those words. I stand by my statement.

Of what words? You don''t think I do as the program would?

quote:
I cannot believe that you still believe in kieren now. All he implemented was Huffman, which is nothing new.

There seems to be a misunderstanding regarding his demo. Correct me if I''m wrong, but wasn''t his original demo using six steps and data massage and all that, and this other demo he''s just putting out to show the 4-16-4 concept?

I thought the Huffman was to prove a point, and the other one he was saving.

quote:

Here you go again taking your opinion on something you obviously don''t understand. I don''t mind you expressing your opinion or challenging statements, but please don''t comment like you know what your talking about. Implement, then talk.

I shouldn''t have said anything at all, but I did indicate my words were empty with " (?)". That the symbol for "aw, shutup(/nevermind)," doncha know. I just meant, not in the sense that it makes it possible for you to go visit your great-grandson by stepping into a time machine.


Lack

Christianity, Creation, metric, Dvorak, and BeOS for all!

Oh yeah, can someone please post a 96-bit stream of ''random data'', by your definition?


Lack

Christianity, Creation, metric, Dvorak, and BeOS for all!

I''m not sure if this is what you mean, but i just wrote c program to use the rand() function 96 times. here''s the result:

11100101 11001101 10101101 00110011 11101101 01101001 00000110 00010010 10010010 10011100 11110110 11100000

96 bits of random data.

this is probably gibberish but i tried

MENTAL

ga wrote:

You are incorrect. Being in orbit means you are constantly accelerating (i.e. gravity always sucks). Your earth center acceleration vector is balanced by your inertia and, thusly, you remain in orbit. This is what the classic time relativity experiment of having one twin orbit a black hole while the other stays on Earth is all about. Gravity is acceleration - everywhere.

Mike Roberts
aka milo
mlbobs@telocity.com

ga wrote:


People in orbit are constantly being accelerated. Gravity is always an acceleration. You are correct that time is moving slower for them since we (on the surface) are deeper in the gravity well.

Mike Roberts
aka milo
mlbobs@telocity.com