Hi Timothy,

On Tuesday 15 February 2005 12:50, Timothy Miller wrote:
> I'm a little confused.  Are you using two tables to compute only one
> reciprocal?  Can you also compute another reciprocal at the same
> time?

If I may jump in here, I also plan to do that.  To be precise, I now 
have one table with sample points and the other with precomputed 
deltas.  Yes, that assumes computing two reciprocals in parallel.  As 
far as I can see, there's no good way to do the whole job with a single 
table.  This is on the understanding that a dual ported 36 bit lookup 
would kill off a multiplier.

Given that there will be two tables, it works out better to have the two 
tables be different and access in parallel than to make them identical 
and access sequentially.  In my case, this eliminates an 18 bit 
difference and obviates some special case logic.  In Lourens' case, it 
lets him explore some interesting alternate formulations.

Regards,

Daniel
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to