Date: Wed Oct 17 2007 - 00:33:51 PDT
Hi Paul, My understanding is that if you (as a modeller) have more information about the nature of the discontinuity you can provide that information to the solver by means of the numerical argument. For instance, if you make triangular waveform you can use a $discontinuity(1). However, if you do not know the nature of the discontinuity, you just do not provide the argument. The idea is to speed up the "recovery" from a discontinuity if you can provide the information. If you cannot provide that information, the solver will be more "carefull" and hence slower, while if you provide the wrong information (say $discontinuity(1), while it is actually $discontinuity(0)) will possibly slow down the solver even more as it may first try the wrong approach, fail and have to try the more carefull approach. Note - I'm not an implementer, so I could be completely wrong about this... However, I do agree with you that specific information is missing in the LRM and it would be good to add it. Cheers, Marq To verilog-ams <verilog-ams@server.eda-stds.org> Paul Floyd cc <Paul_Floyd@mentor.c om> Subject Default discontinuity Sent by: Classification owner-verilog-ams@se rver.eda.org 16-10-2007 16:40 Hi Is "$discontinuity;" (no argument) the equivalent of "discontinuity(0);"? I can't see anything explicitly stating this either in the 2.2 LRM or the 2.3 draft. Regards Paul Floyd -- Dr Paul Floyd Mentor Graphics Corporation -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
