FSUIPC 4.251 - Turbulence feedback

[MELKOR]

Hi Pete!

Tried out some scenarios with PMDG 747X tonight, with some manually configured heavy and severe turbulence layers.

While I really have no opinion on whether or not the simulation of turbulence effects is "real world" accurate, I can report that the bird had absolutely no trouble maintaining a heading with HOLD, SEL, and LNAV in said turbulence.

One thing I did notice while monitoring conditions with WeatherSet - I could see that my manually configured wind layers would occasionally vary (altitude, direction, speed, and turbulence all seemed to change at one point or the other). Do you know: Is this caused by FSUIPC, or FSX?

Thanks,

- Bill

[MELKOR]

Pete,

After I posted this, I saw that 4.252 was available, so I tried that too.

Same behavior from the bird, which is good.

I did notice that the wind speed and direction seemed to vary at a quicker rate than with 4.251.

Thanks,

- Bill

[Pete Dowson]

While I really have no opinion on whether or not the simulation of turbulence effects is "real world" accurate, I can report that the bird had absolutely no trouble maintaining a heading with HOLD, SEL, and LNAV in said turbulence.

Good.

With regards to "realism" I do have a little feedback already from a real-world 747 pilot, who says that the wind directional fluctuations in FSUIPC are good, but he thinks the wind speed changes are not actually large enough (!). Quote: "LIGHT+-5 KTS, MOD+-10 KTS, HEAVY+-15 KTS, SEVERE+-20 those values are much closer to reality and I don’t see any issues controlling the airplane in either regime A/P or MANUAL".

Those fluctuation values are 5 times my current ones! You could try that by setting

TurbulenceRate=5.0

in the INI file (though this would also increase the directional changes to 5, 10, 15, 20 degrees, too much I suspect). I think I shall have to have it set two values, thus

TurbulenceRate=1.0,5.0

for the directional changes and speed changes independently. Look out for this in 4.253, probably next week now.

One thing I did notice while monitoring conditions with WeatherSet - I could see that my manually configured wind layers would occasionally vary (altitude, direction, speed, and turbulence all seemed to change at one point or the other). Do you know: Is this caused by FSUIPC, or FSX?

By default WeatherSet is reading the Interpolated Weather at the aircraft. This is exactly as reported by SimConnect, and it seems to vary a lot. The weather is actually changing all the time in any case -- slower or faster according to the slider in FSX's options, but never static. But one of the reasons it appears to change so much, layer-wise is that, as the weather is 'developed', presumably because of bugs in the code, FSX generates many spurious wind and temperature layers. These used to build up and up until, in fact, the METAR string exceeded SimConnect's maximum allowed, after which they got truncated (or caused a crash in the RTM release). To fix this, when FS is allowed to change FS's own weather it removes these spurious layers and writes the METARs back. These are directly changing the nearby WX stations, and thus the interpolated weather.

On your second message:

I did notice that the wind speed and direction seemed to vary at a quicker rate than with 4.251.

Yes. In fact it wasn't right in 4.251 -- the target to which the speed and direction was moving was computed incorrectly.

If you'd like to test things further, could you see if you could find the highest value for "TurbulenceRate" (between 1.0 and 10.0) with which, with moderate turbulence set (i.e. "2") the 747X still flies correctly? I should then be able to set the default rate for speed changes to half of that (but I'd leave the directional value at 1.0 I expect, considering the Captain's comments).

Thanks for all your feedback. It is very useful.

Best Regards

Pete

[MELKOR]

Thanks for the insight into what is going on with layers.

And, later, I will post results here regarding the upper limit to the turbulence rate (at least, with respect to PMDG 747X).

Thanks, Pete.

- Bill

[MELKOR]

Pete,

I'll be darned... I thought I'd start with the worst case scenario and work myself down. But, even with TurbulenceRate=10.0, and a turbulence of 4, the 747X managed to keep her track quite well! :shock:

And, if a RW pilot is happy with your simulation, so should the rest of us be!

I am going to set my TurbRate=5.0 for the VATSIM FNO tomorrow night - I'll let you know if I note anything odd.

Well done, again, Pete - your diligence sorting out all these wind issues is much appreciated by us all! :D

Thanks,

- Bill

[Pete Dowson]

I'll be darned... I thought I'd start with the worst case scenario and work myself down. But, even with TurbulenceRate=10.0, and a turbulence of 4, the 747X managed to keep her track quite well!

I think having the individual wind changes kept small, and random targets approached slowly, does the trick. It allows the A/P time to compensate.

There are some more changes needed. The actual range of wind variations needs to be made proportional to the actual average wind speed. It isn't at present. A factor 4 turbulence in a wind of 5 knots obviously cannot vary by 40 knots! But that part is relatively easy.

I am going to set my TurbRate=5.0 for the VATSIM FNO tomorrow night - I'll let you know if I note anything odd.

Okay. Thanks!

It would be useful to know not only the turbulence settings, but also the mean wind speed -- cruise winds in particular have such a wide range.

Thanks & Best Regards

Pete

[Pete Dowson]

I'll be darned... I thought I'd start with the worst case scenario and work myself down. But, even with TurbulenceRate=10.0, and a turbulence of 4, the 747X managed to keep her track quite well! :shock:

Arrggghhhhhumble apologies. I've just found a "little" bug. :-( The TurbulenceRate was only affecting the direction of the winds, not their speed. So it isn't so surprising that it didn't make much difference.

Terribly sorry for wasting some testing time.

I'm working on it now, but I also need to make the wind speed changes proportional to the wind speed -- you can't have 40 knot changes with a 5 knot wind! ;-)

I'll see if I can get a new version uploaded tonight. Otherwise it might not be till late Monday now. I have guests this weekend.

Best Regards

Pete

[ahaka]

Tried with version 4.253 and the turbulence seems much more stable and better! Great job!

However, on occasions my aircraft keeps suddenly banking left/right, and even if the ailerons and rudder are centered it stays in that bank. This does not sound like normal behavior does it? It happens in both Level-D 767 and PMDG 747X. I havent tested it on any other aircraft than these two.

[Pete Dowson]

Tried with version 4.253 and the turbulence seems much more stable and better! Great job!

However, on occasions my aircraft keeps suddenly banking left/right, and even if the ailerons and rudder are centered it stays in that bank.

Those two statements sound totally contradictory. How is it "more stable" if you get such sudden banking all the time? I don't understand how you could report both at the same time.

Maybe the wind speed factor is too high at 5.0 -- that is 5 times what it was in 4.251, so I would expect it to be worse, not better. Can you clarify -- is it turbulance tipping the aircraft, or something else?

If it was with turbulence, try reducing the 5.0 value, the second in the TurbulenceRate= parameter.

Pete

[jannier]

Hi Pete

What should be the default value of TurbulenceRate= ?

I notice you show a value 1.0,5.0 in the download thread, but I noticed that the value is TurbulenceRate=5.0,5.0 after install and 1st run of 4.253.

I changed mine manualy back to 1.0,5.0 and wind turbulance and clound turbulance seems within limits so far...

Regards

Jannie

[Pete Dowson]

What should be the default value of TurbulenceRate= ?

1.0,5.0.

I notice you show a value 1.0,5.0 in the download thread, but I noticed that the value is TurbulenceRate=5.0,5.0 after install and 1st run of 4.253.

Yes, because you set it to 5.0, remember? If there's only one number it replaces the default in that position.

Unless I change the names of the parameters each time I change defaults, I'm afraid any previous user values stick.

Regards

Pete

[jannier]

Ahhhh , Ja I remember now I changed the value to 5.0 with 4.252

Thanks for clarifying.

Jannie

[ahaka]

Tried with version 4.253 and the turbulence seems much more stable and better! Great job!

However, on occasions my aircraft keeps suddenly banking left/right, and even if the ailerons and rudder are centered it stays in that bank.

Those two statements sound totally contradictory. How is it "more stable" if you get such sudden banking all the time? I don't understand how you could report both at the same time.

Maybe the wind speed factor is too high at 5.0 -- that is 5 times what it was in 4.251, so I would expect it to be worse, not better. Can you clarify -- is it turbulance tipping the aircraft, or something else?

If it was with turbulence, try reducing the 5.0 value, the second in the TurbulenceRate= parameter.

Pete

Sorry if what I wrote was unclear.

However, the overall stability is much better. In 4.25 it was hard to maintain direction in turbulence/gusts and sometimes the gusts would go up to 50kts even if the maximum set was something like 20. In 4.253 this does not happen anymore and it is possible to maintain direction even in turbulence without the aircraft getting too erratic. So it is overall much improved.

The banking, well, it must relate to turbulence/gusts, since it didnt happen to me in 4.25 if I supress all turbulence/gusts. So it already happened in 4.25, and still happens in 4.253 with turbulence on. However I cant find any particular reason for this since the wind speed/direction seems to be under control now and doesnt fluctuate to enormous speeds/changes.

[Pete Dowson]

The banking, well, it must relate to turbulence/gusts, since it didnt happen to me in 4.25 if I supress all turbulence/gusts. So it already happened in 4.25, and still happens in 4.253 with turbulence on. However I cant find any particular reason for this since the wind speed/direction seems to be under control now and doesnt fluctuate to enormous speeds/changes.

Odd. Sorry, there have been no other such reports, yet. I can only suggest reducing the 5.0 turbulent wind speed rate in the INI. Or suppressing turbulence altogether. Even those flying the originally sensitive PMDG 747X are now finding it okay.

Regards

Pete

[ahaka]

The banking, well, it must relate to turbulence/gusts, since it didnt happen to me in 4.25 if I supress all turbulence/gusts. So it already happened in 4.25, and still happens in 4.253 with turbulence on. However I cant find any particular reason for this since the wind speed/direction seems to be under control now and doesnt fluctuate to enormous speeds/changes.

Odd. Sorry, there have been no other such reports, yet. I can only suggest reducing the 5.0 turbulent wind speed rate in the INI. Or suppressing turbulence altogether. Even those flying the originally sensitive PMDG 747X are now finding it okay.

Regards

Pete

Pete,

It seems to work ok now! I reinstalled FSUIPC 4.25 and then replace it with the 4.253 again. Strange. Anyway, the aircraft still banks in turbulence, but it seems to be only small bank instead of a big tip over. I believe this is normal FS behavior.