Re: Hardware for OGR and simple code?

On 10 Mar, 2005, at 20:28, Frederic Bastien wrote:

> Hi,
>
> I'm thinking of doing a design for OGR on FPGA in my spare time. I was
> looking to simple(not optimized) code that calculate OGR from scratch. 
> I
> did find anything, only optimized code. I didn't also find the 
> algorythm
> for the tree how it is made. Do you know where I can get such code? Do 
> you
> know where if there is people that have tried to do OGR on hardware?
>
> Any pointer to information will be apreciated.

RC5 is probably much more worthwhile, if only because parallelism works 
much better. Each key is checked in some 200 steps, and finding the 
next key is also trivial.

OGR, on the other hand, depends on finding the next Golomb ruler, and 
seeing how long it is. That depends on finding the next ruler (for some 
value of 'next') and seeing if it's a Golomb ruler, and that depends on 
incrementing some length in the ruler, and somehow checking when to 
stop and then incrementing a different length. At least, that's how I 
understand it.

There does seem to be enough assembly going on that an FPGA may be 
worthwhile, though (if not only because you can fit a lot onto one 
chip).

- Purr

_______________________________________________
Hardware mailing list
Hardware-Ra3b/QYEcJ3d140v2zMXi0fjHoOT/h/[email protected]
http://lists.distributed.net/mailman/listinfo/hardware