Re: 6702 chip

On Tue, 17 Jul 2012, William Levak wrote:

I have analyzed the logic of the chip's operation, and tried a number of 
variations.  The last number register and the flip flops are initialized
to 214 and the cannot be moved forward or backward in the logic stream. If 
they are, then subsequent output is wrong.  It's like a CRC.  Not only 
must the number be right, but the position in memory must be right also.

This would then mean that operation does not begin with the shift 
registers outputing to the flip-flops. This step is at the end of the 
cycle.  Operation would begin with the flip-flops being loaded with 214, 
which would also send 214 to the output register.

Also if we make the even/odd register a not even/odd register, then 
everything else is initialized to 0.

This is simpler. The revised chip operation is:

INITIALIZATION:

The flip-flops and last number register are set to 214.

Everything else is set to 0.

OPERATION:

1.  The flip-flops output to the output register.

2.  The chip waits for input.  If the input is odd and the not even/odd
     register is 1, continue below.  Otherwise transfer the complement
     of the low bit of the input register to the not even/odd register, and
     continue waiting for input.

3.  Exclusive or the input with the last number register.  Exclusive or
     that result with the current position of the shift registers.

4.  Shift the shift registers one position, and copy the input register to
     the last number register.

5.  Output the current position of the shift registers to the output
     register.

6.  Go to step 1. and continue indefinitely.


I have run more than 4000 randomly generated numbers through this 
sequence, and it works perfectly.

I have not been able to pin down exactly when the not even/odd register is 
set.


wlevak@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org

       Message was sent through the cbm-hackers mailing list