MAGIC Magic Mailing List
 
 
From: R. Timothy Edwards (tim AT stravinsky DOT jhuapl.edu)
Date: Thu Aug 23 2001 - 09:44:13 EDT

  • Next message: Jeff W. Sondeen: "Re: DRC rules" 

    > yes, you're right, while it DOES give an error with SCN6M_SUBM.10.tech27
> both files have the same rule specified:

> spacing ndiff,anres,rnd,nfet,ndc/a,ndm12c/a \
> pdiff,apres,rpd,pfet,pdc/a,pdm12c/a 12 touching_illegal \
> "N-Diffusion spacing to P-Diffusion < 12 (Mosis #2.3+2.3)"

> but as there is a rule 'optimizer' going on maybe it's getting deleted.

In my experience writing tech files, this sort of thing is rarely if
ever due to "rule optimization".  Most of the time it's due to the rule
being preempted by another rule.

I've always wanted to write a "rule compiler" to simplify the problem of
untangling all the interactions between rules, and to hide the very
non-intuitive edge4way rule structure behind something graphical and
intuitively obvious.  But it seems like such a Herculean task. . .

						---Tim

    •  
     
    Questions? Contact Rajit Manohar     		cornell logo