oh, also. the "bad" drive refuses to sleep now, without having been manually coerced e.g. "hdparm -y /dev/sdg". the other drive in it's raid0 pair sleeps without an issue. thinking i'd made a *visible* change to the config, i made a freedos usb and ran hdat2 v60b5 to dump all possible drive/dco/controller/etc. settings into a file to compare. the only differences were in the io addresses and serial/world wide name. fun factoid: hdat2 shows the IEEE OUI as hitachi. but alas, all settings are otherwise identical between the good and bad drive