In default weather winds can shift from for example: 180 and 30kts deg to 270 and 90kts instantly, thats why they are interrupting autopilot and are causing aircraft to be unstable. Registered FSUIPC does the job of smoothing those shifts, and with ASE DWC you don't even need any smoothing :)

Thanks, I know that ;) (if that was for me) ASE DWC works very well, haven't tried registerred FSUIPC, but I'm sure it does the same job. ASE DWC smoothes 2knots/2deg I think...?My point was that I've read these voilent changes comes from shifting between weather stations, and that I can't really see why the changes would be so violent if that was the case.. :)Geir J.

I think you're underestimating the severity of it. How do you explain what happens when I install and uninstall it? The problems begin after an install and disappears when I uninstall. Without it I'm back to the normal "crappy/annoying" FSX weather where you get sudden gusts of wind every now and then, but nothing remotely close to what I experience with FSUIPC installed. What is that?

I am sorry but I can not explain what you have not shown. You see wind shifts, I say FSX has and always has had wind shifts. What is there to explain? If you could produce a repeatable weather scenario that showed a wind shift with FSUIPC installed, and no wind shift with it removed, then I (and others) would have to take you more seriously. So far all you have done is say "the wind shifts and I blame xxx".

Thanks, I know that ;) (if that was for me) ASE DWC works very well, haven't tried registerred FSUIPC, but I'm sure it does the same job. ASE DWC smoothes 2knots/2deg I think...?My point was that I've read these voilent changes comes from shifting between weather stations, and that I can't really see why the changes would be so violent if that was the case.. :)Geir J.

FSX does not maintain a model of the atmosphere to which it can apply changes (though hopefully, MS Flight will), instead, it works out every no and again what you should experiance, that is why the changes can be so violent. Try setting up a manual weather scenario and you will see what is available to play with and how limited it is.

FSX does not maintain a model of the atmosphere to which it can apply changes (though hopefully, MS Flight will), instead, it works out every no and again what you should experiance, that is why the changes can be so violent. Try setting up a manual weather scenario and you will see what is available to play with and how limited it is.

Thanks Paul, I' ve set up a manual scenario many times. Been using MSFS activly since 1994.My point was (which you know if you read my answer to CaptEm) that if the abrupt wind changes happen from leaping from onestation to another (or many), or interference between them (as has been claimed by some), one should expect only minor changes in the upper winds still, since upper winds are very steady over large distances (in reality) and would not change much fromstation to station. That's why I can't really fully understand a 180 deg. change in the wind between two closely positionedweather stations... Therefore I'm thinking it has to be some corruption of (or wrong) data happening somewhere..Geir J.

Real World if you cross the jet stream you can have abrupt wind shifts aloft. It is one of the leading causes of severe clear air turbulence and difficult to predict. If whatever the weather bureau uses to measure this gets caught up in it you'll get that much of a shift between winds aloft reporting stations.I was a member of the coffee head-to-toe club one day crossing Lake Erie I would imagine in the FL30s in clear air just west of Buffalo. The flight attendants even had a worried look on their faces. It was on Northwest Airlines KLGA direct to KMSP as I recall back around 1968. I don't recall the aircraft.I have crossed the lake before but never experienced more than the usual mild chop from lake thermal convective activity with mild bumps of crossing shorelines (allowing for wind drift offset).http://en.wikipedia.org/wiki/Jet_stream#Aviation might be of interest.

Guys,I said this in a different thread too, but honestly I don't know where this is coming from - there most definitely are crazy wind shifts with a completely clean vanilla FSX install with no addons and no FSUIPC. I've seen it a million times. The idea that FSUIPC is the source of wind shifts in FSX is not true in my experience. Neither is the use of SimConnect - SimConnect is just a way for our gauges to talk to the sim, we don't mess with weather at all.

That's why I can't really fully understand a 180 deg. change in the wind between two closely positionedweather stations... Therefore I'm thinking it has to be some corruption of (or wrong) data happening somewhere..

My point was that I've read these voilent changes comes from shifting between weather stations, and that I can't really see why the changes would be so violent if that was the case.. :)

It is a known bug in FS's weather interpolation code which has been in place since and including FS9. FS uses a triangular pattern of weather stations for interpolation -- every part of the globe is in one triangle or another, and the apexes of those triangles are the three weather stations whose weather is read in order to derive the weather at the aircraft.Some of those triangles are wildly stupid, by error. I think they are different for FSX to FS9, but I remember there was one in FS9 affecting a narrow area to the south of Rotterdam where two of the stations were quite close but the third was somewhere in North Africa!When the weather is not too wildly different in each of the three stations, the interpolation is fine. When one of them is wildly different, or had corrupted download data, you can get the windshifts depending on the relative wind directions. These are a result of a trigonometric sign reversal -- just a careless mistake in the code regarding cosine and sign quadrants.In FS9 I managed to hack into a place where the wind data was being passed from WEATHER.DLL into SIM1.DLL for implementation in the simulation engine. It is there where I apply the wind smoothing, which therefore operates well in FS9.In FSX I never managed to locate a suitable place to hook. Things in FSX were bundled completely differently (they moved even more towards the horrible complete "black box" approach of object-oriented programming), and modules called each other by normal links which all changed, with handles to objects instead of actual data, and not by address tables and real data as in the past. Therefore the wind smoothing in FSUIPC is not so good -- it is time based, trying to force the wind values at regular (short) intervals. It is better than nothing but cannot completely stop the shifts.This is the real reason for ASE's "DWC" mode. Global weather mode in SimConnect sets the same weather at EVERY weather station in the world, so the interpolation can't possibly go wrong. It is actually a reversion to what we used to have with weather programs in FS98 and FS2000, with the weather closely controlled for different areas as you travelled.Before ACES closed they were well aware of the problems in the weather engine, but were at that time unable to fix them. The person who wrote the weather engine (Ms Niniane Wang) left when FSX was released, and it seems no one else understood her code well enough to attempt to fiddle with it. So they were planning to re-write the whole thing for FSXI. It wasn't their main priority though. Of course performance was at the top of everyone's list.So, we are stuck with it. The only bright notes on the horizon are that the forthcoming Microsoft "Flight" will be all new code (with new bugs no doubt), and the possibility that Lockheed Martin's "Prepar3D" (based on ESP which in turn was FSX+Acceleration) is still undergoing development and they may well eventually fix the weather engine. Of course Prepar3D is intended only for professional use. (BTW I doubt whether I'll get involved as a developer for MS Flight. I'm getting too old).I use FSX + ASE in DWC mode and I've not seen any nasty wind shifts since. Mind you, in the areas I fly (Europe) we are usually relatively free of them in any case. During my researches into the problem I found the worst area for them was around Chicago.RegardsPete

Used to have many problems over Turley while flying from Greece to the Far East. Must 've been a triangle between Greece, Turkey and New Zealand that caused all this! However, reading all these very interesting things I am quite amazed by how a complicated thing such as the simulation of the globe (in some sence) is totally screwed by simple things like sine and cosine errors or triangles between Rotterdam and north Africa that only God knows how they were initially created!Thank you for the thorough explanation Mr Dowson. It answers many fundamental FS related questions!

Real World if you cross the jet stream you can have abrupt wind shifts aloft. It is one of the leading causes of severe clear air turbulence and difficult to predict. If whatever the weather bureau uses to measure this gets caught up in it you'll get that much of a shift between winds aloft reporting stations.I was a member of the coffee head-to-toe club one day crossing Lake Erie I would imagine in the FL30s in clear air just west of Buffalo. The flight attendants even had a worried look on their faces. It was on Northwest Airlines KLGA direct to KMSP as I recall back around 1968. I don't recall the aircraft.I have crossed the lake before but never experienced more than the usual mild chop from lake thermal convective activity with mild bumps of crossing shorelines (allowing for wind drift offset).http://en.wikipedia....stream#Aviation might be of interest.

Well, that's not the cause of the winds shifts in FSX... There are not jet streams all around, they are primarily associated with the large climate cells of the globe, and behind and in front of cold and warm fronts. The latter being the lowest ones, and mostly encountered in aviation.CAT are normally avoided to the highest extent possible, by staying clear of the jet stream core.

Thank you Pete, for your thourough explanation :)Well appreciated. Looking forward to MS Flight :)

During my researches into the problem I found the worst area for them was around Chicago.RegardsPete

Pete,Thanks for your work in FS and stating facts here so clearly.Chicago is called 'the windy city' but simulating that the FS way, no, don't like that. :( Regards,Harry

Thanks for posting that Pete, I learned a few things I didn't know there.