tj22

It won't switch to GPS flightplan track unless you have Nav1 set to a NON-ILS freq. I usually load the ILS into Nav1 standby and something obviously not ILS (like 108.00) into active. To transition from GPS to ILS, I set AP HDG mode to make the turn to intercept ILS GS (or just hand fly it), then swap Nav1 freqs. Once I verify the ILS is active, then AP NAV mode to capture it.

Probably the Tu-114. Much more limited service than the L-188, but respectable numbers for a prop, even by today's standards.

Too much nose-up trim. I used to fly that quite a bit and as I recall, the best TO trim setting is actually negative (nose-down). Then it behaves fairly reasonably. Probably fixable via the cfg file, but I never got around to investigating. Also the thrust centerline is offset from the pitch axis, so power change has a tendency to cause pitch change. It's most noticeable (and I suspect unrealistic) when you pull off power just before TD.

I think some here must be confusing thread with process. FS runs in a single process, but with multiple threads. And yes threads on a single process can run on multiple processors, as determined by the process affinity mask. To insist FS is confined to a single core, much less a a single thread is just plain wrong.

It was written in the single core era, but it is ~not~ single threaded. (As to how the workload is distributed over the 30 or so threads - hard to say. Task Manager doesn't dig that deep.) In any case when you spread the affinity, the CPU's internal architecture does the heavy lifting to make it happen, so results are like so many things with FS - hardware-dependent.

I would not call it pointless, but I've seen no benefit (this is on a Corei7) giving FS more than 2 physical processors (4 cores). What is more useful is being able to keep it off the 0 physical processor (cores 0-1) just because by default so much other crap will run there. I also shuffle them from one Fs session to the next. What I've found most important with my i7 is to assign entire physical processors to FS, that is cores in pairs starting with even core#s. Making it play more hopscotch than necessary across physical processors has a definite negative impact on rates, smoothness, and stability. For example (core# affinity settings): 2,3,6,7, or 4,5,10,11 = good 1,3,5,7 or 4,6,8,10 = not as good

Thanks for confirming. 3 weeks sounds about right. On closer examination there are significant differences in the .wx file binary data from one DL to the next, so at least ~something~ is being updated, but definitely not winds & temps. I just hope it's temporary. and not some harbinger of the death of FS RW weather.

I've seen no change in downloaded FS RW weather data for days now. Anybody else seeing this?