- JD protocol timing adjusted to handle NTSC in manager
- disabled write cache default (can cause corruption on power loss)
- progress bar in manager was sometimes wrong for long directories
- set pio0 transfer setting and remember it (for some hardware)
- drive detection updates, hope it's not worse now than before
- use reset for recovery (fixes hang after interrupted transfers)
- don't spin up power saving drives when changing timers
- invalidate drive after direct write (consistency after filesystem creation)
- change geometry handling, some new drives don't like CHS at all
http://idedos.ide64.org