Has anyone had any experience with the DS1302 serial time-chip ?

I have one big problem in that I can't get the oscillator to run with a
MultiComp 32.768k watch crystal.  The crystal has a load capacitance of 12.5
pf instead of the 6 pf as spec'd by Dallas, but by-pass caps don't help, nor
does a bleed resistor. Are there any funny or special crystal characteristics
that I should be aware of - the data sheet is a bit light in this regard.  It
says to just connect a crystal between X1 and X2. Well, I've done that and it
doesn't work. Is the 6 pf really that critical ?

My second problem is that I can't get the DS1302 to talk back to me. I can only
assume that this is because the oscillator isn't running. Could someone please
confirm this for me as I've been going nuts with two really nice routines I've
written to set and read the time, but can't test them properly.

I've only got one DS1302 device at the moment, so if anyone has one in a
bread-board (or some other scenario which would allow them to pop the crystal
out) and could confirm that without a crystal, X2 sits at half the rail
voltage with X1 at ground potential, I'd appreciate it.

If anyone is using this device, could they please provide me with details of a
known good crystal along with details of any by-pass caps, or bleed resistors
if required.

Thanks and Regards,

Brenton