I don't know the answer - I'd normally just try it to find out, which presumably you can do fairly easily? (I'm sat at home, in Windows, and can't easily test, and there's little point me testing something you can just as easily test).

Regards,

Andrew.

I decided to save indysjl a little effort, and tried it myself.

Negative delay values do work.

Geoffrey - it's not quite the same as your suggestion, because with a positive delay, the pulse does not start until after the delay has happened - so the behaviour is different at the beginning of the simulation. If negative delay weren't supported though, it would be a reasonable alternative.

I've attached a picture showing no delay, a negative (-5ns) delay, and Geoffrey's suggestion (25ns delay, since the period is 30ns).

Regards,

Andrew.