9/22    electronic question.  what is the standard of merit on choosing
        by-pass capacitor for power supply lines?  it seems that ppl
        always use 0.1 uF all the time.
        \_ It's based on a rule of thumb, I think the idea is to have the
           capacitor smooth out dips in power as different components
           temporarily draw additionally more.  Of course if your
           capacitor fails and becomes equivalent to a closed switch
           that can suck a battery dry real quick --oj
        \_ you want the bypasses to be as large as possible. that way,
           you can smooth out all the transients. unfortunatley, the
           bigger the cap, typically, the higher its internal resistance,
           so the worse it becomes at smoothing out high frequency current
           drains. so you settle for either a small cap or an expensive
           tantalum. the power drain also typically happens when the output
           of your chips flip, which is at some factor times your clock rate.
           if you expect current draw every 4 ticks, you want to put a low
           pass which smooths out spikes at 4*clock_freq. but like oj said,
           it's a rule of thumb. if you notice bad shit happening, you put
           in more caps. -ali.