I spent the whole day trying to get these objects to build, not crash, and finally work somewhat properly on the Mac. I ended up with only minor changes and a single #ifdef, but it is still about 100,000 clip recordings behind the old objects in terms of testing. The features are well out of date on the newer object as well.

I may or may not have time this week to test on Windows.

I moved the recordStart/Stop member variables to the record base class, but that should not affect the Linux code.

cgc