According to Ignatios Souvatzis:
> 
> I think thats not equivalent to what Chris proposed. I see three time types
> here:
> 
I think, off the cuff, that you can reduce these three clocks to a 
single one, merely recording the last adjustment as a constant (I'd 
hazard multiple constants) for the other two.  But I haven't really 
thought hard about this, merely assumed that there are really clever 
people out there who can see further than I can :-)

++L