"In general, it is likely that end users should only experience and notice TSC drift on single-processor dual-core platforms which do not expose HPET and which are running an older operating system that is using TSC on that platform."
... single-processor dual-core platforms... OH! the X2 again. BTW, the only s939 supporting chipset (that I am aware of) that has support for HPET is the ULi 1689. No PCI-X or SataII though.
"Until TSC becomes invariant, AMD recommends that operating system developers avoid TSC as a fast timer source on affected systems."
So we wait till Q4 of 2006 and pray Vista doesn't rely on TSC? Cuz /usepmtimer is still not working 100% in everything else MS OS ... like Win 2000, XP Pro, Xp Home, Server 2003, XP 64-bit. And what about the programs that access the TSC directly? Enter *ntel to the rescue.
"When an operating system can not avoid using TSC in the short-term, the operating system will need to either re-synchronize the TSC of the halted core when exiting halt or disable C1-clock ramping.
Lets see what else the MS Dual core patch does:
http://support.microsoft.com/?id=896256
Under 'Correct TSC synchronization' it says:
" ... This change makes sure that the multiple-processor HAL continues to correctly synchronize the TSCs across all processors on a computer ..."
Does the MS patch muffle the problem? Yes. Does it completely solve the problem? No, cuz the TSC is still drifting, hence the stutter/desync.