Re: [stella] TIA Audio Polynomials

Subject: Re: [stella] TIA Audio Polynomials
From: Adam Wozniak <adam@xxxxxxxxxxxxxxxx>
Date: Fri, 21 Nov 2003 00:11:01 -0800 (PST)
On Thu, 20 Nov 2003, Frank Palazzolo wrote:
> There is an algorithm which, given a bit sequence, will generate the 
> smallest LFSR which can be used to generate it.  It is called the 
> Berlekamp-Massey algorithm, and you can probably find an 
> implementation on the net.

Thanks for the pointer, but I'm still a little fuddled...

I've found two or three implementations of this algorithm.
When I give them the bit stream "101011" I always get the same answer,
which I interpret as [3 2] in my earlier notation.

However, when I feed this into my LFSR as the polynomial, I do NOT
get "101011" out.

However, when I use "100110101111000", I do get x^4+x^3+1 which I
interpret as [4 3] which is what I expect.  Very confusing.

And of course, it blows up completely with longer bit strings.

Will code for food.
"The dinosaurs are not around today because they did not have a space program."
  -- Arthur C. Clarke

Archives (includes files) at
Unsub & more at

Current Thread