which is in line with Jim Otteson's statement in the Markman transcript about the ring oscillator VCO in PLLs operating within a range. The important portion of the section below is that the ring oscillator in our patent does demonstrate a "free running" ring oscillator, among other ideas. Within PLLs over short time intervals, the VCOs act like free running ring oscilators. <<<<>>>>"An excellent discussion on the ef- fect of supply and substrate noise on PLL jitter, caused by dig- ital switching circuits, is given in [13]. In the following sections, the effect of VCO noise on the time-domain behavior of PLLs is analyzed using the above results in conjunction with those given in Section II. It is reasonable to expect that the PLL be- haves like the reference oscillator over long measurement inter- vals, and like the (free-running) VCO over short measurement intervals, such that time-domain results for oscillators (given in Section III) can provide useful approximations and insights."
http://www.berkeley-da.com/prod/datasheets/PLLJitterDavid%20Lee_IEEE_2002.pdf