Wind shifts with 4.20

[Alias]

Hi Peter,

I have installed FSX+SP1+SP2, ASX and FSUIPC. But I still get enormous wind shifts at all stage of flight. In the pictures below I have attached my settings of both programs so that you can have a look at them. I have tried any possible combinations of wind smoothing but with low success. The most worrying thing is that I've found FSUIPC rather ineffective notwithstanding the installation has been carried on according to your manual.

Another thing I wanted to ask you is: is it still available the barometric pressure smooth utility? It was a feature of the version 3.xx but in the one I have 4.20 I can't find it.

I hope you can help.

Regards

Lorenzo

[Pete Dowson]

I have installed FSX+SP1+SP2, ASX and FSUIPC. But I still get enormous wind shifts at all stage of flight.

You've evidently not reviewed the two other recent threads here on this subject?

It has been a real pain hacking through FSX's code to try and get some sort of control over the winds, but by all reports it is a helluva lot better with version 4.219, available in the FSX Downloads.

Do please also read the other relevant threads.

Another thing I wanted to ask you is: is it still available the barometric pressure smooth utility? It was a feature of the version 3.xx but in the one I have 4.20 I can't find it.

None of the intimate weather control we had with FS2004 and before is easy in FSX. The facilities in FS2004 and before cost me many thousands of hours hacking into FS code. With the coming of SimConnect in FSX I thought I was over all that (I am getting too old), but it wasn't to be.

I've tackled the wind smoothing now. I cannot make the visibility smoothing work no matter what I try. And no one has, so far, mentioned any need at all for pressure smoothing in FSX. Are you saying that this jumps up and down badly as well?

The user guide for the FSX version of FSUIPC does explain a lot of this, if you'd care to look. It does show a complete lack of pressure smoothing, you'll notice. ;-)

Regards

Pete

[Alias]

Wow Pete,

this is what I call a "rush" reply.

:oops: sorry I didn't realize that there was an update, but simply because the first link showed on the FSX thread sends you to your homepage where 4.20 is still the version available.

But that's OK, I have just installed the new package, shall try tomorrow and report.

Concerning barometric smoothing..well it was just to say that anytime I'm descending through 18000 ft. I always have to push the "B" key many times to get the most updated value, as it change constantly until touchdown. In FS9 it was easier I hit "B" and got the very same value I had once on the ground, but this ain't a major issue.

Another personal remarks on you. All I can say is a big THANK YOU for what you are doing. I can't imagine how many hours of endless job you've been doing to help the sim community, and please...PLEASE don't say you're old..who is going to take care of us after you? :lol:

God bless you

Lorenzo

[Pete Dowson]

... simply because the first link showed on the FSX thread sends you to your homepage where 4.20 is still the version available.

It isn't actually my page at all, but Enrico Schiratti's, as it points out. There is a link there to this Support Forum, and the Support Forum is nicely equipped with an Announcement facility, where I make Announcements and keep folks up to date with what is going on. ;-)

I don't release major updates as often as interim ones here, even if Enrico would actually update his page so often, which he doesn't.

I was made extraordinary clear with the "official" 4.2 release that the wind smoothing, new in that release, was "experimental" -- it actually says as much on the screen you so kindly posted here (as if i'd not seen it before! ;-) ). There's also quite a lot written in the documentation as well explaining a lot of the problems with doing much with FSX weather.

... it was just to say that anytime I'm descending through 18000 ft. I always have to push the "B" key many times to get the most updated value, as it change constantly until touchdown.

There is no "B" key in a real aircraft, and 18000 is only the "standard" level in North America. Here in the UK the TA varies from about 3000 to 6000. It actually varies considerably all over the world. And unlike in North America the rest of us have two different altitudes to consider -- TA and TL. The TA is generally published in the charts, but the lowest Flight Level you can fly, known as the TL, varies with the QNH and is computed by ATC to ensure adequate vertical separation betwen those on QNH and those on STD (i.e. flight levels).

The "B" key "cheat" in FS doesn't obtain the QNH for you, it actually sets it on the altimeter. What you are doing is wrong (and of course not possible in the real world). When descending below flight levels you should set the altimeter to the QNH value you get from ATC or an ATIS weather report. That wouldn't change so often. You shouldn't have to be worried about what the ACTUAL sea level pressure is immediately below you -- other aircraft will be using the ATC/ATIS QNH values as well.

Also, the fact that the actual QNH is changing slowly means that FSX is actually smoothing the pressure -- otherwise it would keep jumping. If it kept jumping you would see your altimeter jumping too. Do you?

In FS9 it was easier I hit "B" and got the very same value I had once on the ground

That means, really, that there must have been a sudden jump somewhere -- from another area's QNH to the one at that airport. That isn't smoothing, it is the opposite of smoothing!

PLEASE don't say you're old..who is going to take care of us after you?

I'm 65 this year and really would like to fly more in my nice 737 cockpit on which I've spent quite a bit over the years. FSX's SimConnect was supposed to take over from me, and let me have more time .... now it looks like FSXI before I get time off! :-(

Regards

Pete

[Bill Casey]

Hi Pete,

Ref barometric pressure jumps - yes it does jump around in FSX. Since all my flying is on Vatsim I tend to notice this more because obviously there are 'real' ATC and aircraft around and keeping altitude is rather important. The extent of the jumps varies on what you have running: with ASX running the jumps can be frequent and very large (sometimes even 20+mb!) and I've seen altitude jumps of 1,000ft before now. With FSX online weather it does still jump but not quite so severely, having said that altitude movements of 2-400ft whilst cruising on 1013mb are really quite common (in smooth air). Using FSInn weather the pressure seem to be vastly more stable in all phases of flight (although in FSX you can't use FSInn weather above cFL240 thanks to a simconnect problem with TATs). In terminal areas we quite often have to use "B" or we will find ourselves at hopelessly the wrong altitude on the ATC screen if we use the controller's QNH, bizarre but true.

Somehow, somewhere MS made some complete boo-boos with the wind and pressure didn't they? Like you I am starting to wonder of FSXI will be the finished article and that FSX was little more than a beta for some of the weather "features"!

[Pete Dowson]

Since all my flying is on Vatsim I tend to notice this more because obviously there are 'real' ATC and aircraft around and keeping altitude is rather important. The extent of the jumps varies on what you have running: with ASX running the jumps can be frequent and very large (sometimes even 20+mb!) and I've seen altitude jumps of 1,000ft before now. With FSX online weather it does still jump but not quite so severely

Hmm. I'm not seen them with either FSX downloaded weather or with ASX. You have updated ASX to the very latest version, haven't you? It seems pretty good.

I wonder if your problems are related to the on-line operations. There's a lot going on with AI traffic creation, to emulate multiplayer, with those programs.

Anyway, I'll experiment here with pressure control, to see if I can change that without messing other things up. Not easy -- aircraft systems like those in the PMDG range are very sophisticated and use values like air density and temperature correlated with pressure in the flight and A/P computations, so I have to find a way of manipulating one aspect and hoping FSX will adjust the others accordingly. :-(

Now you are going to tell me you get temperature jumps as well, aren't you? ;-)

Regards

Pete

[Bill Casey]

You have updated ASX to the very latest version, haven't you? It seems pretty good.

Oooh yes, I never miss an ASX update! It is extremely good but seems to present problems for online flying with getting winds that match the ATC with the ASX wind smoothing enabled which can and does change the wind direction from that in the Metars. I have to fly online with that feature of ASX switched off. That obviously means I get the mega windshifts at altitude and hence my following your progress very closely Sir Pete.

I wonder if your problems are related to the on-line operations. There's a lot going on with AI traffic creation, to emulate multiplayer, with those programs.

Oh I absolutely agree with you. Online flying adds a whole different dimension to the issue!

Now you are going to tell me you get temperature jumps as well, aren't you? ;-)

Might be :wink:

[Alias]

Ref barometric pressure jumps - yes it does jump around in FSX. Since all my flying is on Vatsim I tend to notice this more because obviously there are 'real' ATC and aircraft around and keeping altitude is rather important. The extent of the jumps varies on what you have running: with ASX running the jumps can be frequent and very large (sometimes even 20+mb!) and I've seen altitude jumps of 1,000ft before now. With FSX online weather it does still jump but not quite so severely, having said that altitude movements of 2-400ft whilst cruising on 1013mb are really quite common (in smooth air). Using FSInn weather the pressure seem to be vastly more stable in all phases of flight (although in FSX you can't use FSInn weather above cFL240 thanks to a simconnect problem with TATs). In terminal areas we quite often have to use "B" or we will find ourselves at hopelessly the wrong altitude on the ATC screen if we use the controller's QNH, bizarre but true.

Well, I'm happy someone explained the issue with better words. Anyway, since I fly off line, this isn't a big issue to me..but I remembered the old days in FS9 when everything was perfect. Today I have spent many hours trying different configurations to smooth the wind and I have seen some good improvement with new version. I have used 747 PMDG and Level D for testing, I could see wind struggling to stay smooth, very nervously I must say, but sufficiently to avoid stupid overpeeds :mrgreen:

Shall be back with further tests.

Regards

Lorenzo

EDIT LATER: I wanted to take a piece of footage of my last landing Pete, but I failed. Anyway with the latest release 4.222 I have found that from TO to say 8000 ft., due to the shifts, the airpeed tape jumps violently up and down very nervously. The direction of the gust doesn't change at all. Then, above 8000, the speed and direction seem to get calm and they are not swinging too much during the trip. The same applies for the descent (which is a bit of a problem for a landing). It also true that the test was effected in the "windiest" place in the world: Keflavik. The touchdown happened with a 240° wind at 32 Knt.

The very but very strange thing in all of this is that I couldn't find any difference when "Wind smooth" option was "on" and when it was "off". Do I have to reload FSX anytime I check and un-check those options?

[Bill Casey]

As it happens Lorenzo, I too did a test flight out of Keflavik on Sunday! It's not a fair test at all since on departure I had winds of 230 @ 45G60! I can't imagining anything smoothing that kind of weather :shock: However the excellent news is that once climbing out and into the cruise I only had one major wind shift all the way to Stansted. Also no sudden pressure changes and altogether the smoothest cruise I've ever had on FSX :D This was using FSX download weather (15 minute update), flying online with FSInn (weather off), v4.219 of FSUIPC with smoothing on and flying the Level D 767X. Excellent result!

One point to note, I did see the rapidly fluctuating airspeed (+/- 2-3kts) on the initial climb but with no change of wind direction. OK it was gusting heavily at Keflavik but a later flight elsewhere showed the same effect in quieter conditions. It doesn't disrupt anything in a heavy except making me constantly watch it and wonder what's happening!

Altogether I'd say this smoothing enhancement is one huge leap forwards :D

[Alias]

As it happens Lorenzo, I too did a test flight out of Keflavik on Sunday! It's not a fair test at all since on departure I had winds of 230 @ 45G60! I can't imagining anything smoothing that kind of weather :shock:

Hi Bill,

well good to know I'm not the only one in Keflavik :wink:

I have one question since I'm using the LevelD too: can you actually see the plane shaking while you are on the ground before TO? It's incredible how the a/c reacts to the strong wind gust; I don't know how realistic this is, but it looks great.

And yes, as I said, after take off the direction is stable but the airpeed tape jumps up and down, pushing “SHIFT+Z” (but also looking at the EHSI) I can see the wind jumping up and down nervously from 40 to 48 knots (for example). This, of course, effects also the groundspeed; but all of this disappears while at cruise level where no changes seem to happen and everything comes back to normal.

Are you having the same experience? I dare to ask you this to be sure everything is normal and I’m not having some corruption in my system.

Thank you

Lorenzo

[Bill Casey]

Lorenzo, yes exactly the same effect with the "nervous" speed shifts on climb out. Aircraft being knocked around on the ground with heavy wind gusts is very realistic, look at that tail sticking up there, it's a like a big sail! Real world I was once sitting on the tarmac in a 737 at Port Elizabeth (S Africa) as a violent storm cell passed overhead. Our aircraft was shaking incredibly in the wind that was with it. The pilot said "we're not moving until the storm has" but another a/c did depart and our Captain said "I hope God is with those guys today". A wind gusting to 60kts will batter anything quite badly in the real world and many catering trucks (for example) wouldn't be safe to operate in that wind. Also the meximum crosswind component for many airlines/aircraft is only about 25kts so 60kts is downright dangerous!

[Pete Dowson]

Anyway with the latest release 4.222 I have found that from TO to say 8000 ft., due to the shifts, the airpeed tape jumps violently up and down very nervously. The direction of the gust doesn't change at all. Then, above 8000, the speed and direction seem to get calm and they are not swinging too much during the trip.

How big were the airspeed jumps? If there are turbulence, gusts or variance settings, then you'd expect to see the airspeed jumping about a little, in proportion (something like +/-2 knots per unit of turbulence). For turbulence I also try to make the ride feel "bumpy" by variations in the vertical speed, but I suppose without a motion base you wouldn't feel that! ;-) (but you should see it from outside the aircraft!)

The "suppress turbulence" and "suppress gusts" options on the Winds and Cloud pages should now work though. If you don't like the disturbance simulation, try turning it all off.

The very but very strange thing in all of this is that I couldn't find any difference when "Wind smooth" option was "on" and when it was "off". Do I have to reload FSX anytime I check and un-check those options?

No, not at all. If it was the same with wind smoothing off then it was FSX's very own simulation doing it all. They have implemented it, you know. The only reason for my efforts is that it seems it doesn't always work -- that's what folks have been complaining about!

The only reason I simulate disturbances when smoothing is enabled is that otherwise the smoothing action cuts out any disturbances which FSX is putting on which should really be there. I'm only trying to put back that which I'm removing, when it should be so! I should have done that really in FSUIPC3 -- the wind smoothing there did smooth out all those types of features, incorrectly IMHO.

Regards

Pete

[Alias]

Lorenzo, yes exactly the same effect with the "nervous" speed shifts on climb out. Aircraft being knocked around on the ground with heavy wind gusts is very realistic, look at that tail sticking up there, it's a like a big sail! Real world I was once sitting on the tarmac in a 737 at Port Elizabeth (S Africa) as a violent storm cell passed overhead. Our aircraft was shaking incredibly in the wind that was with it. The pilot said "we're not moving until the storm has" but another a/c did depart and our Captain said "I hope God is with those guys today". A wind gusting to 60kts will batter anything quite badly in the real world and many catering trucks (for example) wouldn't be safe to operate in that wind. Also the meximum crosswind component for many airlines/aircraft is only about 25kts so 60kts is downright dangerous!

..aaah..OK good to know Bill. Then it's not bad :mrgreen: BTW in South Africa storms are incredible (been there in honeymoon lately).

How big were the airspeed jumps? If there are turbulence, gusts or variance settings, then you'd expect to see the airspeed jumping about a little, in proportion (something like +/-2 knots per unit of turbulence). For turbulence I also try to make the ride feel "bumpy" by variations in the vertical speed, but I suppose without a motion base you wouldn't feel that! ;-) (but you should see it from outside the aircraft!)

Well, actually it depends on the place I have flown, but more or less I should say that yes it varies of a couple of knots, but very nervously. Anyway shall do some other tests in a quiter and more realistic situation when I'm back home.

The only reason I simulate disturbances when smoothing is enabled is that otherwise the smoothing action cuts out any disturbances which FSX is putting on which should really be there. I'm only trying to put back that which I'm removing, when it should be so! I should have done that really in FSUIPC3 -- the wind smoothing there did smooth out all those types of features, incorrectly IMHO.

You are right there, but personally I'd like to have a "smooooth" and maybe unrealistic climb rather than a bumpy and so nervous one. FS9 was perfect under this point of view altough I have never had the opportunity of a single high cross wind landing since everything was too smooth (which is also unreal).

All I can say is: keep up the good job.

Cheers

Lorenzo

EDIT: I have noted that using PMDG 747 it was also worst. The plane couldn't even turn and catch the set heading because of this nervous behaviour. But I have heard other comments somewhere on this issue.

[Pete Dowson]

Ref barometric pressure jumps - yes it does jump around in FSX.

I've added some code to FSUIPC4 to see if the variables I found for pressure and temperature can be used to smooth these things.

I am rather worried that I'm only seeing part of the story and that by smoothing some values others may be incorrect and upset finely tuned aircraft like those from PMDG and LevelD. So, for now I am not publicising the facilities, nor are they accessible through the options, only in the INI file.

If you want to test these, please download 4.223 using this link:

http://fsuipc.simflight.com/beta/FSUIPC4223.zip

Then you have to add these parameters (or change them if you find them there) in the [General] section of the FSUIPC4.INI file.

PressureSmoothness=0

TemperatureSmoothness=0

With these zero values the options are off. Otherwise set the number of 1/100ths of an hectoPascal (hPA or mb), or degree Celsius, which you wish to allow changing, at most, each second.

For example:

PressureSmoothness=20

TemperatureSmoothness=50

sets pressure smoothness at 5 seconds per hPA, and temperature smoothing at 2 seconds per degree.

I am rather dubious about temperature smoothing and might not keep it in. After all temperatures can drop and rise considerably in and out of clouds and, obviously, with changes of altitude. For the latter I don't smooth a change if the altitude changes by more than 10 metres in the same checking period. I'm not sure if this is good enough. I won't have time to do much testing this week I'm afraid.

The smoothing shouldn't happen when the aircraft is on the ground.

Please try these facilities, especially the pressure one, and let me know how you get on.

Regards

Pete

[Bill Casey]

Thanks Pete - I tend to agree that temperature isn't perhaps so crucial unless you're getting massive changes in TAT at higher flight levels since that'd muck up Mach numbers. At present any temp changes aren't nearly significant enough to get worried about. I'll try the latest version tonight in something exotic like a LevelD and see what happens. I'll try the pressure tweaks next time I'm flying in lower airspace since that's where it can be more of a menace. Things are certainly heading hugely in the right direction (no pun intended) with the winds though :D

[Alias]

Hi,

just made some tests with the new beta version, and I must say that is not bad at all. The wind was still a bit struggling but not so much until 8000 feet and barometric pressure was much more reliable and reminded me the old times in FS9. I think you made a good step foward with this Pete.

Cheers

Lorenzo

[Pete Dowson]

Hi,

just made some tests with the new beta version, and I must say that is not bad at all. The wind was still a bit struggling but not so much until 8000 feet and barometric pressure was much more reliable and reminded me the old times in FS9. I think you made a good step foward with this Pete.

This is with the LevelD 767, is it?

Thanks. I need someone to test with the PMDG 747 now.

Regards

Pete

[Alias]

Yes it was with LevelD. I couldn't test PMDG yet, but will be a pleasure.

Lorenzo

[Alias]

Well I have tried with PMDG and found that she behaves more or less like LevelD.

The only difference is that PMDG can't keep the heading assigned. Of course this is not related to FSUIPC but it's strange. Anyway the problem is ASX. It's a tremendous FPS eater and this, of course, effects the whole system. The fact is that basically FSX is nothing more than a beta test simulator, I wonder how other add-ons can interact with a program that, in itself, isn't working fine.

Cheers

Lorenzo

[JSkorna]

Hi,

So many people are using ASX with little or no FPS hits. You might want to e-mail us for help with this.

[Pete Dowson]

Well I have tried with PMDG and found that she behaves more or less like LevelD.

The only difference is that PMDG can't keep the heading assigned.

But that was the problem with the original implementation of my new wind smoothing which everyone so far, except you, has said is now working perfectly with the latest improvements.

What is deflecting the A/P off the heading, can you tell? Can you give any more information? I don't understand how you have this conflicting result, especially when you are only really checking the pressure smoothing at this stage ...

... unless it IS somehow due to the pressure smoothing? Are you saying it was okay before you enabled that? If so, then I will just have to remove the pressure smoothing, or at least publish strong warnings.

the problem is ASX. It's a tremendous FPS eater

Are you saying the PMDG 747 works fine without ASX?

I don't get any noticeable hit from ASX at all, so I'm surprised by what you say. There seems to be quite a bit more of a hit from the PMDG 747, but I have only tried it on my test PC in any case. It won't be any good in my cockpit where I don't use any FS panels at all.

Regards

Pete

[Alias]

Well, actually without ASX the PMDG works perfectly, there must be something in ASX that effects the way the heading is managed, I don't believe it is connected to FSUIPC, as I have tried also with an unregistered version and had the same problems.

But Mr. JSkorna is listening I would really like to help but I don't know in which way.

Lorenzo

[Pete Dowson]

Well, actually without ASX the PMDG works perfectly, there must be something in ASX that effects the way the heading is managed

Hmmm. That is very strange. As far as I know ASX is just setting the weather via SimConnect, using the METAR setting facilities, effectively just like FSX's own download weather updating system. I think its superior weather effects come from clever massaging of the data before it is submitted, careful selection of the weather stations being updated, and the generation of interim increments between real world METAR changes, all things not done by the FSX simplistic downloads.

I don't believe it is connected to FSUIPC, as I have tried also with an unregistered version and had the same problems.

When you say "unregistered version" I assume you mean the normal version but with the FSUIPC4.KEY file removed? There aren't separate versions. What about trying it with the current main user release, 4.20, before I added the new smoothing options? That might be more telling.

Regards

Pete

[Alias]

Thanks Peter, I will try this evening.

Lorenzo

[JSkorna]

Support is best handled by e-mail or our forum at AVSIM. I hate to tie up this thread with our support as well. Just write to me and give me some of your details.

Thanks!