Bj_rn wrote (2022-01-23):

 O>> Why reboot? Isn't the lost memory freed after binkd has been
 O>> terminated?

 BF>     Well, if that was the case, I wouldn't regard it as leakage.

On a modern mainstream OS, the mess is cleaned up after the application has been terminated, but it's still called memory leak.

 BF> No, it isn't. Can it be some cache grabbing that isn't done correctly?

 BF>     I've always thought it was related to the badly written WinAPI calls
 BF> for cache usage, but then I saw someone else mentioning it, so I thought
 BF> I'd bring it up. As I said -- no big deal.

What do I know about Windows ... ;).

The support hotline would suggest to upgrade to the newest version and try again.

2:203/2's binkd is 7 years old.