Home > Archive > Compression > August 2005 > Re: Inventing store 1.12Gb raw video in 5Mb!
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Re: Inventing store 1.12Gb raw video in 5Mb!
|
|
| Jim Leonard 2005-07-31, 9:59 pm |
| Denis wrote:
> Now someone can say I use this compressor to compress N movie and then I
> connect
> all these movie toghethere to obtain a new incompressible movie , if I send
> this movie to the compressor what happen? The answer is the compressor does
> not
> compress it , after compression the movie has the same dimension becouse it
> is
> not a movie!
It doesn't matter if the data is a movie or not; it's still impossible
if the claim is to reproduce the movie *exactly*.
Now, if you can accept a *lossy* conversion -- the movie will look
*almost* the same -- then there is still a discussion here (although
compression ratios of 1000:1 is still extremely far-fetched).
| |
|
| >
> It doesn't matter if the data is a movie or not; it's still impossible
> if the claim is to reproduce the movie *exactly*.
>
I think it is possible becouse does not exist 2^8.000.000.000 differents
movies.
Do you agree with me ?
Denis.
| |
| Earl Colby Pottinger 2005-08-02, 9:00 am |
| "Denis" <dnsflex_cut_this_@yahoo.com> :
[color=darkred]
> I think it is possible becouse does not exist 2^8.000.000.000 differents
> movies.
> Do you agree with me ?
NO! As has already been pointed out and single movie can be cut up into
hundreds of segments and and shuffled about to produce that many diffirent
movies from just that one set of data.
Heck, the way movies are made there is so much cut out for timing reasons
that you probably could shuffle the contents and make that many 'meaningful'
movies if you wanted to. But that would still be onlt the film from one
movie - and there are pently more out there.
Earl Colby Pottinger
--
I make public email sent to me! Hydrogen Peroxide Rockets, OpenBeos,
SerialTransfer 3.0, RAMDISK, BoatBuilding, DIY TabletPC. What happened to
the time? http://webhome.idirect.com/~earlcp
| |
|
|
>
>
>
> NO! As has already been pointed out and single movie can be cut up into
> hundreds of segments and and shuffled about to produce that many diffirent
> movies from just that one set of data.
Yes , you can do something like that but you can not obtain 2^8.000.000.000
differents data .
From a set of data in the best case you can subdivide the number in
8.000.000.000 differents segment of one bit and the number of differents new
data you can obtain are 8.000.000.000! .
How can you compute all these combinations? It is impossible . In this
number there many and many configurations impossible to find , impossible to
compute .
When I make a movie a make some operations and this set of operations is a
"program" and there is not a program to build 2^8.000.000.000 combinations!
If you subdivide the data into 100 segments you have 100! combinations about
10e+157, using 1024 bits you can map 10e+307 only using 128 bytes. Only 128
bytes to map 100! movie of 1Gb .
Denis.
| |
|
|
>
> I don't think the inventors named in this posting used this approach,
> but the idea is interesting.
>
> For example if you take 10.000 commercial DVD video movies and lay all
> frames on each other, to check what combinations in every frame is used
> and not used, what shall be the result? What percentage of the
> theoretical combinations in one frame is used and not used? How quick
> are the used frame combinations increasing by every new compared movie?
> Are there significant combinations often used and others almost not
> used? Is it possible to write a video alphabet as Samuel Morse did with
> his Morse code alphabet? Are the results the same when using a matrix
> instead of frame where multiple matrixes form exactly one frame?
>
> Or turn it around, how many different frame combinations a true random
> generator can generate in for example 1 hour on a fast computer. What
> percentage of the theoretical frame combinations is generated? How
> quick new combinations are found? Is it possible to write a video
> alphabet for only the combinations not found?
>
> And if both test are done, what percentage of the random frames
> combinations found equal the combinations found in the DVD frames?
>
> Are these tests done? If yes what where the results?
>
I didn't build a compression program but I think it is possible to build it
| |
|
|
"Jim Leonard" <MobyGamer@gmail.com> ha scritto nel messaggio
news:1123091669.658348.298050@o13g2000cwo.googlegroups.com...
> Denis wrote:
about[color=darkred]
128[color=darkred]
>
> Mapping is not the same as compression. What do you map to? Where is
> the original data stored?
>
> I have 100 movies on my shelf. I have given them all a number. If I
> put a number into a text file and give it to you, do you have all of
> the digital information needed to watch the movie?
>
Yes , this is true if I have 100 Kolmogorov complexity these are
incompressible and it is impossible to compress the data .
But this is Shannon independent.
Thanks , Denis
|
|
|
|
|