Well put, cuts to the core of the problem

Tried the same mission twice from a savegame from a few minutes before the crash. This time no crashes so I haven't been able to reproduce.

I was looking downwards from the hill to the north of the town. Had the town that is getting pounded by mortar shells in my field of vision. Since it was a PhysX crash, I wonder if some of the mortar shell effects down below could be related ?

Now that the usual dump files do not contain anything helpful, perhaps it could be possible for the physx code to write debug info to another file which could be sent to you in case of physx crashes.

NordKindchen, voted up (of course)!

Just wanted to say that to me, the two pictures you link to in this ticket does not convey the before/after situation as much as the pictures in your forum thread does. That way BI maybe has to go into the thread to understand the actual benefit. Perhaps link to the the before/after shots you have in the thread (the ones where you look through the scope)?

Ok, did test the mission once more and this time it ran fine. I may have picked a slightly different route into the water though.

But I ran some memtest using and it finds something. Has to change the RAM blocks. So no reason to investigate further for now.

I'll retest on my new system in about a week or so and let you know if it is stable.

I tried the mission twice and crashed at the exact same spot.

I went into the water right next to the two units that are on the beach and swam outwards along the sea bottom. After about 15 seconds the problem happens in both cases.

I suspect I have a bad RAM block. Would the message in the RPT about tbb4malloc_bi.dll support that theory ?