[SI-LIST] Re: Problem in converting IBIS2XTK

On Friday 19 October 2001 11:45 am, Muranyi, Arpad wrote:
> Good point, Al.  I just wonder whether that is really the
> reason XTK (or other tools) does it.

The only one I know well is Hyperlynx.  I cannot speak for the others.

Dropping points complicates the algorithm, so there must be a good 
reason to do it, otherwise they would just take the easy way, which 
would seem to be to take all points.

There are two significant reasons for dropping points.  One is to 
smooth out noise in the derivative so the reflections make sense.  
The other is to improve convergence.  If you analyze it, you will see 
that these are really the same problem.

I don't know how aggressive the various implementations are at 
dropping points.  The one I did for Hyperlynx only drops points where 
there is a problem.  If anything, it may not be aggressive enough.

As far as I know, most simulators use linear interpolation, which 
implies piecewise constant derivatives.  It is tempting to say to use 
splines, but this introduces a whole new set of problems.
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field

or to administer your membership from a web page, go to:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List archives are viewable at:     
                http://www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages 
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu

Other related posts: