Behavior of management-related features under apm/apmd, i.e., the status quo before going to ACPI: - power button causes immediate power off (ouch!) The following features work pretty much as expected: - Fn+Font, Fn+CRT/LCD, Fn+{Brightness,Contrast}{Up,Down} do the expected things. - Fn+EjectCD does the expected thing but takes ~10 sec to do it. OS loses the time ('date' shows none elapsed). The following work well in my usual circumstances (i.e. have not tested heavily with CardBus devices active and such). They work in text mode or X, but all suffer from the same two glitches described below. - Fn+Suspend, lid down, or apm -z cause deep sleep, pwr button or lid up wakes. - Fn+A causes Suspend-to-Disk and power off, pwr button causes POST, Resume-from-Disk. - Fn+D turns off the display. Any key restores. - Fn+H spins down the disk. Any disk activity restores. - Fn+Setup and Fn+Bat enter the BIOS setup at the main page or the battery status page, respectively; escape returns. - Critical battery level triggers automatic suspend-to-disk. Text Glitch: text can be displayed in a screen-filling font-stretched mode, or in the natural font size with wide blank margins. I prefer the latter, but all the above operations leave the former in effect. They do this without updating some recorded state somewhere, because it always takes TWO presses of Fn+Font to get back to natural size. I.e., the state is somewhere still recorded as natural size, so the first press would select stretched (with no visible effect because the mode has already been changed to stretched) and the second press correctly restores natural size. I suspect a DSDT bug. X Glitch: in X, on return from any of the above actions, the X cursor has some random pixels attached to it. Moving the mouse over any item that causes the cursor to change corrects it.