Live Business Chat
2015 Sep 06, 01:55:16 pm *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
News: This is a friendly non-profit discussion group about making money. Click the "NOTIFY" button every chance you get to receive instant alerts about new information. We don't like spam. If your first posts are about something you want from us, it is probably spam, and you will be banned in forums worldwide. No soliciting! If you have problems with the forum, contact badon in the business chat.
   Home   Help Search Calendar Login Register  
Pages: [1]   Go Down
Author Topic: is RSC32 source code available?  (Read 1413 times)
0 Members and 1 Guest are viewing this topic.
Almost Nobody

Offline Offline

Gender: Male
Posts: 3


« on: 2014 Sep 08, 03:32:44 pm »

Persicum RSC32 is a fast program, but the fact that I have not found any reference for the source code have keep this program out of my Linux computer for years.

Persicum, is there any change you will release the source code for everyone to check?
Offline Offline

Posts: 10

« Reply #1 on: 2015 Mar 28, 04:18:38 pm »

I would like to see source code too, but where is persicum?  How else can he be contacted?  I emailed him a few weeks ago, but I got no reply yet.  I am not 100% sure the email address I used to contact him is correct.
Offline Offline

Posts: 57

« Reply #2 on: 2015 Apr 07, 06:22:52 am »
author about sources, but I think this just bad joke.
(his five year ago implementation still best on speed (more over 100-times better than average), but useless without at least slow decoder sources)
The last version is 3.05, it is greatly improved since 2.x. I have changed my hobby to high voltage electronics, so this 3.05 is couple years old indeed...
P.S. Unfortunately it is all sad.
if you still wants the source you could begin to learn how use disassembler and all about finite field Fast Fourier Transform and multipoint evaluation (though it's only a hypothesis: that he used this approarch, I just did not find anything fast enough and less complex).
multiplication by the inverse of a square Vandermonde matrix is known
as the interpolation problem and its complexity is O(k *
(log(k))^^2). The multiplication by a Vandermonde matrix, known as
the multipoint evaluation problem, requires O((n-k) * log(k)) by
using Fast Fourier Transform, as explained in [GO94

Gohberg, I. and V. Olshevsky, "Fast algorithms with
preprocessing for matrix-vector multiplication problems",
Journal of Complexity, pp. 411-427, vol. 10, 1994.

]. The total
complexity of this encoding algorithm is then O((k/(n-k)) *
(log(k))^^2 + log(k)) operations per repair element.
With the classical
Gauss-Jordan algorithm, the matrix inversion requires O(k^^3)
operations and the vector-matrix multiplication is performed in
O(k^^2) operations.
This complexity can be improved by considering that the received
submatrix of GM is the product between the inverse of a Vandermonde
matrix (V_(k,k)^^-1) and another Vandermonde matrix (denoted by V’,
which is a submatrix of V_(k,n)). The decoding can be done by
multiplying the received vector by V’^^-1 (interpolation problem with
complexity O( k * (log(k))^^2) ) then by V_{k,k} (multipoint
evaluation with complexity O(k * log(k))). The global decoding
complexity is then O((log(k))^^2) operations per source element.

you need a lot of luck on this way Smiley.
« Last Edit: 2015 Apr 07, 07:07:09 am by ddmk » Logged
Offline Offline

Posts: 57

« Reply #3 on: 2015 Apr 07, 07:38:31 am »

However rsc32.exe is using library rsc32_fftw3.dll

which looks like ?modified? GPL FFTW library

so it's license REQUIRED to open source code of any program which using the library.

Offline Offline

Posts: 57

« Reply #4 on: 2015 Apr 09, 03:47:34 am »

First version info is here
(1.96 (2009) can be still loaded , and already uses FFT and FLI(Fast Lagrange Interpolation), file Benchmark.pdf is of particular interest)
had no decencies on GPL FFTW (so can be published with author license)
that's in case if someone wants to investigate this tool through history before persicum came here.

Offline Offline

Posts: 57

« Reply #5 on: 2015 Apr 09, 04:07:40 am »

ok I found why he using non modified(or not so much) fftw
FFT using FFTW I am calling on a key -tm10. Advantages are:
1) a prime number can be anything, not necessarily k.2 ^ n + 1, in my case FFFFFF79
2) the number of points can be anything, not necessarily the 2 ^ n, I select nearest from the top to the N, which is divisible by 2,3,5,7 - but this is a purely arbitrary
3) FFTW is certainly using complex numbers, but there is a transformation to form result of purely real number form. Then from the result residue is taken by mod ffffff79, and that's all. This (mod calc) could be done at the very end.
4) in order to achieve this, you need a good understanding of the issues fft, try ...
(forum is offline now I get it though google cache
, for simplicity,
but citation is from page 14 which is not in archive)
it can be seen as developer diary, so if someone wants to spend a lot of time to reread it he could reimplement algorithm.
there is no actual links to articles which author used, but enough information to find them.
А реализовать это дело - это как сыграть по нотам, задача вроде техническая и второстепенная, но требует ведь заметных усилий
To realize this thing - it's like to play on the notes, the problem seems minor and technical, but also requires significant efforts.
« Last Edit: 2015 Apr 09, 04:31:10 am by ddmk » Logged
Pages: [1]   Go Up
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.18 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!
Page created in 0.029 seconds with 17 queries.