Feedback on wind smoothing (FSUIPC 4.222 & FSX+SP2)

[kerke]

Pete, I run FSX and FS9 and on both I use the latest FSUIPC versions for FSX and FS9. Here are my observations with regards to Winds:

WINDS UNDER 1000 FEET

I typically try to suppress winds under 1000 feet (I set ‘Max wind speed up to 1000ft’ to 1) because I don't really enjoy what I personally consider FS9's and FSX's "excessive" wind effect on the runway, specially with cross winds. On FS9, FSUIPC does indeed set the winds to 1 knot and when I land or taxi, the winds are actually 1 knot. However, on FSX, FSUIPC seems to have no effect on FSX's winds under 1000 feet or on the runway.

WIND SMOOTHING

I also don't particularly enjoy FS9's and FSX's unrealistically strong wind changes which cause airspeed to jump up and down and cause abrupt heading changes. On FS9, my FSUIPC "wind smoothing" setting (I set it to limit changes to 1 know per second) seems to work very well; when FS9 weather does its crazy abrupt wind changes, FSUIPC comes to the rescue and takes control of the situation. The net effect is that wind changes are smooth the way they are supposed to be. However, on FSX (either with SP1 or with SP2), it seems that FSUIPC is unable to get a handle on FSX's abrupt wind changing behavior. With the latest FSUIPC 4.222 it seems to be a little better but not as good as with FS9. On FSUIPC 4.222, I check the “Allow changes to FS own weather”, I check the “Smooth wind changes near aircraft with changes limited…” to 1 knots or degrees per second, the “or this many seconds for each 1 knot or degree” to 0, and uncheck the “Smooth only when airborne” box. I also set the “Max wind speed up to 1000ft” to 1. Nevertheless, wind changes at cruise altitude are still quite strong with the undesirable KIAS speed changes and heading changes. And on the runway, FSX winds are higher than my FSUIPC set limit of 1 knot, as if FSUIPC was not even installed.

These are my .02 cents with respect to Winds. I will be happy to test newer FSUIPC versions and report back my findings to you. Thank you and keep up the good work!

Kerke

[Pete Dowson]

However, on FSX, FSUIPC seems to have no effect on FSX's winds under 1000 feet or on the runway.

Currently the facility to limit the surface wind speed is simply trying to do this via the METAR settings at the nearby weather stations, the system which has proved fairly ineffective for wind smoothing.

If the wind smoothing changes made recently look to be reasonably effective, and with no unwanted side effects, then I will continue and try to implement facilities such as that one. I would like to implement FS9 "taxi wind" facilities which operated by only limiting side wind, not head and tail wind, but I'm afraid the variables for that elude me in FSX.

In other words, thank you for this feedback but it doesn't actually relate to the current changes being evaluated. Not yet.

I check the “Smooth wind changes near aircraft with changes limited…” to 1 knots or degrees per second, the “or this many seconds for each 1 knot or degree” to 0, and uncheck the “Smooth only when airborne” box. I also set the “Max wind speed up to 1000ft” to 1. Nevertheless, wind changes at cruise altitude are still quite strong with the undesirable KIAS speed changes and heading changes.

Can you tell me what sort of sudden wind changes you are getting, because this is contrary to other reports. Are you sure they are not caused by actual Wind Turbulence, Gusts and Variability, possibly set by the incoming METAR specifications? In FS9 my wind smoothing suppressed all of that, wrongly in my opinion. In FSX I simulate it. Possibly my simulation is excessive, but by all accounts so far not unreasonable.

If you are getting big wind changes with the "Suppress" options set for turbulence (on Wind and Cloud pages) and gusts (on the Wind page), then I need some saves Flights (FLT + WX files) to find out why. They shouldn't happen ...

... though there is one exception. If the simulation stops for any reason, for instance you visit the menus to display the map or do something else, then I cannot smooth winds in this period, When you return to normal flight mode there may be sudden changes to whatever FSX thinks the winds should be by then.

And on the runway, FSX winds are higher than my FSUIPC set limit of 1 knot, as if FSUIPC was not even installed.

See the answer above. This is the same point. I'll get to that soon if it is easy, or else if and when I am reasonably happy that the current wind smoothing changes work well enough. They are a pretty awful "hack" into FSX code and I don't want to retian this if it isn't working, as you seem to think.

If it doesn't work then it gets withdrawn completely. However, most feedback is opposite to yours, so I need to know why yours is different.

Meanwhile, I will look at how to make the surface wind limit work using what I've learned so far.

Regards

Pete

[kerke]

Thank you for your quick reply Pete.

If you are getting big wind changes with the "Suppress" options set for turbulence (on Wind and Cloud pages) and gusts (on the Wind page), then I need some saves Flights (FLT + WX files) to find out why. They shouldn't happen ...

Pete, on all my tests so far, both the "Supress turbulences" and "Suppress gusts" checkboxes have been unchecked on the Winds page. I wonder which one of the 2, turbulences or gusts, should I check to get a handle on sudden speed and heading changes while not making the whole experience unrealistically smooth.

On the "Max wind speed up to 1000ft" subject, by setting to 1 knot, my main goal is to limit cross winds to 1 knot while the aircraft is on land (e.i. taxiing, on the runway landing and/or taking off) the way it works on FS9 is very good IMO.

Thanks again Pete, I appreciate your taking the time to work with FSUIPC users and your continued dedication to improve your product for so many years.

Kerke

[Pete Dowson]

on all my tests so far, both the "Supress turbulences" and "Suppress gusts" checkboxes have been unchecked on the Winds page. I wonder which one of the 2, turbulences or gusts, should I check to get a handle on sudden speed and heading changes while not making the whole experience unrealistically smooth.

If you are comparing it with FS9's smoothing by FSUIPC3 you would need to suppress both, and the turbulence on the clouds page too. The smoothing I did then got rid of it all. :-(

When you get these changes look at the wind reports for your altitude in WeatherSet2, see if there is a non-zero value against Gusts, Turbulence or Variance.

Gusts changes only wind speed, not direction.

Variance changes only direction, not speed

Gusts changes both, but much faster and usually with a lot less amplitude -- i.e. a sort of twitchiness. I also make it change the vertical speed a bit, to make it "bumpy", though you'd need a motion base to feel it -- you might be able to see it from outside.

The turbulence is currently arranged to have maximum amplitude of +/- 2 knots or 2 degrees per unit in the setting (which ranges from 0 for off to 4 for severe).

The other two are set in their appropriate units, knots maximum gust speed and degrees variable direction.

When you suppress turbulence I also suppress variance I think (or is it when you suppress gusts? Sorry, it's in the docs).

On the "Max wind speed up to 1000ft" subject, by setting to 1 knot, my main goal is to limit cross winds to 1 knot while the aircraft is on land (e.i. taxiing, on the runway landing and/or taking off) the way it works on FS9 is very good IMO.

Currently I haven't a way to control only the cross-wind. I think I'd have to limit the wind speed completely. Very unrealistic, but I'll put it in, probably tomorrow.

Actually, by eliminating crosswind so much on landings you are missing out on one of the most fun things in flying -- crabbing to land, or (in a GA aircraft) using crossed controls. Mind you, if you have no rudder pedals both are a bit difficult (impossible in the latter case) I suppose.

Regards

Pete

[Pete Dowson]

On the "Max wind speed up to 1000ft" subject, by setting to 1 knot, my main goal is to limit cross winds to 1 knot while the aircraft is on land (e.i. taxiing, on the runway landing and/or taking off) the way it works on FS9 is very good IMO.

Well, the best I can do at present is limit the wind speed itself, rather than limit only the crosswind as I managed to in FS9 and before.

Try version 4.231, now available in the FSX Downloads announcement above. I've not finished all the testing it really needs, but so far it works well -- it even graduates to/from the correct windspeed at the 1000' AGL point.

Regards

Pete

[kerke]

Thank you Pete, I'll download it tonight.