Cosmos 2008 SP3 Non-Linear Restart broken?

I've called this into my VAR but while they are checking intoit I wondered if anyone has any suggestions....

I am running Cosmosworks Advanced Professional 2008 SP3 both onXP64 and XP. At first I thought this was only an XP64 problem but Ifinally checked it this morning and it does it on XP also.

I am running Non-linear analyses and often start an analysis, runit a step or two, then stop it to check that things are going wellbefore I leave it to run completely. I have recently (since SP3)run into a problem where the analyses don't seem to remember wherethey are when I do a restart. "The save data for restart" box waschecked at the start of the run and the "Restart" box can beselected, but when I do, the start time stays at 0 and the analysisrestarts from 0 instead of from the step it was at.

This USED to work just fine. It is very annoying as I lose hours ofcomputation time whenever I can't restart.

I have tried almost so far. I've started in a fresh directory,remeshed, rebuilt the study, etc. Sometimes restart works,sometimes it doesn't. I haven't been able to figure out a patternof what causes it to fail.

Thanks.

-MikeSolidworksSimulation