From 4f39648320ccb0aa56b7645ffdda67eff4a3c8b0 Mon Sep 17 00:00:00 2001 From: h_vogt Date: Sat, 27 Jul 2013 09:19:06 +0200 Subject: [PATCH] ./BUGS, update the list of known bugs --- BUGS | 25 +++++++++++++------------ 1 file changed, 13 insertions(+), 12 deletions(-) diff --git a/BUGS b/BUGS index 6f6906e43..f8da85f85 100644 --- a/BUGS +++ b/BUGS @@ -18,20 +18,21 @@ tracking system or on the mailing lists is public. If you need to send sensitive information, send them to a developer directly but still write non sensitive data on the bug tracking system to keep track of bug history. +* please refer to the sourceforge bug tracker at + http://sourceforge.net/p/ngspice/bugs/ + ------------------------------------------------------------------------------ OPEN BUGS: ------------------------------------------------------------------------------ -* return value: +* pole-zero analysis - When ngspice is called in batch mode it returns 1 even if the simulations - are run. The problem arises when there is a .control section but no - .print, .plot or .fourier lines. Ngspice outputs: + The pole-zero (.pz) analysis, which still uses the original spice3f5 + algorithms, does not give correct results for all tested input files. + The effort to correct that issue seems to be large. - Note: No ".plot", ".print", or ".fourier" lines; no simulations run - - and exits returning 1. - - Workaround: add -r to the command line to avoid this. - -* please refer to the sourceforge bug tracker at - http://sourceforge.net/tracker/?group_id=38962&atid=423915 + +* voltage (and current) controlled switches need better timing + + The controlled switches (SW, CSW), which still stem from the original + spice3f5 implementation, sometimes lead to wrong results when on and off + ramp times differ considerably.