Re: MDDs vs. BDDs

From: Fabio Somenzi (Fabio@colorado.edu)
Date: Wed Jun 21 2000 - 11:17:59 MDT

  • Next message: Yuji Kukimoto: "Re: MDDs vs. BDDs"

    >>>>> "RD" == Rainer Dorsch <rainer@rai16.informatik.uni-stuttgart.de> writes:

     RD> The VIS user manual says that VIS is using MDDs and BDDs. Are
     RD> there any recent papers about MDDs, which compare them to BDDs
     RD> and analyse in which areas BDDs are more appropriate and in which
     RD> areas MDDs should be preferer?

    The one paper I'm aware of is:

    @INPROCEEDINGS { Sriniv90 ,
            ADDRESS = "Santa Clara, CA" ,
            AUTHOR = "A. Srinivasan and T. Kam and S. Malik and R. K. Brayton" ,
            BOOKTITLE = iccad ,
            MONTH = nov ,
            PAGES = "92-95" ,
            TITLE = "Algorithms for Discrete Function Manipulation" ,
            YEAR = "1990"
            }

    MDDs are implemented as a layer on top of BDDs. A direct
    implementation is obviously possible, but the paper explains why it is
    not advisable. MDDs provide a convenient way to abstract some detail
    when dealing with multi-valued variables.

     RD> Does anybody know T. Kam's PhD thesis helps here and if yes, if it is
     RD> available online?

    I could not find it with a quick search.

    Fabio



    This archive was generated by hypermail 2b29 : Wed Jun 21 2000 - 11:20:52 MDT