Old 744 Forum

Archived posts

Weather engine

Page: 1

Author Post
Member
Registered: Apr 2012
Posts: 68
Dear Hardy,

please accept my apologies, if this has been discussed earlier. I could not find anything via 'Search function'.

I would like to ask, how does the weather system works. When I am flying 'online' (download the latest weather every 30mins), the winds, temperature seem to be in-line with PFPX weather model.

I do understand that PFPX is only a forecasting tool, and expected winds may not lived up to what is actually out-there at FL340.

However, I would presume that when I am flying offline (i.e. no internet updates), the wind would somewhat resemble the foretasted ones (for example currently I am enroute from ZSPD to EDDK and winds are 271/41, but forecast over PAROM fix call for 110/34) I can send a SITU file with scanned flight plan. I am noticing that the weather remains relatively static for a very long time.

I have checked the 'Set Zones by flight track...'.

Now the question arises. How does the weather behaves, once PSX is offline. Does the weather model 'lives on its own' (it continues to develop; i.e. after 7 hours it has changed), or remains at the download state (i.e. after flying more than 7 hours, the weather has not evolved and is 7 hours old).

If the latter is true, would it be possible to encompass TAF in the weather model? For example the weather around the World we evolve in 6/12 hours to reflect TAF prediction (to a certain limit). I understand that the weather might not be what have been predicted, but it is the next best thing we have got.

As I have no coding skills, I do apologize if the scope is way to time consuming.

It has been said numerous times and it needs to said again, thanks for everything! 8)

Cheers,
_______________
Andrej
Moderator
Registered: May 2009
Posts: 5140
Hi Andrej,

PFPX doesn't control PSX -- unless an add-on injects data into PSX.

In PSX there are 3 weather modes:

1. Set zones by flight track and downloaded METARs -- with Internet ON
(METARs will change twice an hour by downloads.)

2. Set zones by flight track and downloaded METARs -- with Internet OFF
(METARs will not change; downloading is disabled.)

3. Don't set zones by flight track and downloaded METARs
(The zones are manually set and won't automatically change during the flight.)

That's the basic principle.

A TAF system is not possible in the near future; way too expensive.


Regards,

|-|ardy
Member
Registered: Apr 2012
Posts: 68
Dear Hardy,

thank you to your reply. Perhaps, I was not clear in my initial post. I do not expect PFPX data to be fully reflected in the PSX and vice versa (unless as you state, an add-on injects such data into PSX). I use PFPX as planning tool only.

From the three modes, I am mostly interested in mode 2 (as sadly, I cannot devote more than 3-4 hours to PSX :) ).

Does that mean, that METARs should be updated based on the nearest airport/METAR will not change at all? More specifically, what happens to winds in higher elevation (speed and direction). Do they change/adjust accordingly? Or do they stay the same?

I am under the impression, that if the 'Set zones...' is checked, the weather should be updated based on the closest airport/METAR. I am not sure about the upper winds.

When using Mode 1, the winds and speeds are pretty much spot on (despite the fact that PSX and PFPX use different data sets).

Speaking of TAF, maybe one day? :lol: :mrgreen:

Cheers,
_______________
Andrej
Moderator
Registered: May 2009
Posts: 5140
P.S.: The basic idea behind the current weather model is this:

In the terminal area of an airport the aircraft may be in a hold at relatively low altitudes for a very long time, or it may be at the gate for a very long time. That is, the aircraft may be within a certain zone for a very long time. In this case, the weather will not change by flying from zone to zone across the planet; instead, the weather will change only over time. In this case, the control parameter for the weather around the aircraft is the time, not the space (if Internet download is enabled; the download will change the METAR of the zones from time to time).

In the cruise area, however, on a normal flight, the aircraft will never be stationary. The aircraft will mainly fly in the tropopause system of the weather model, outside of downloaded METAR winds. In that case, the control parameter for the weather around the aircraft is the space, not the time (if the jetstream system is enabled). I.e. the weather around the aircraft changes by the fact that the aircraft flies through the jetstream system which provides different winds on every spot of the planet.

Low altitude: Travel through space and time.

High altitude: Travel through space.

With this principle, there is always a dynamic weather change through the whole flight. It is entirely static only if the user disables the Internet downloads and deactivates the internal jetstream system.


Regards,

|-|ardy
Moderator
Registered: May 2009
Posts: 5140
Hi Andrej.

andrej wrote
Does that mean, that METARs should be updated based on the nearest airport/METAR will not change at all? More specifically, what happens to winds in higher elevation (speed and direction). Do they change/adjust accordingly? Or do they stay the same?


METAR wind data refer to surface winds only. In PSX there is a parameter for the surface wind and one for the tropopause wind. The wind in between is smoothly interpolated based on a geographic algorithm as described on page 42.

Before I repeat here on the forum what I wrote in the manual, could you tell me which part in the manual you didn't understand? It's not always easy for me to understand what the user doesn't understand :-) and so I sometimes end up writing overly long answers to a question I misunderstood :-)


Regards,

|-|ardy
Member
Registered: Jun 2015
Posts: 2
Hi andrej,

There is a tool developed by myself which injects weather from P3D into PSX; if you'de like to give it a try
PSXWx V1.5 Currently it only works with P3D as this is the simulator Simfest-UK uses.

It works by interpolating the surface wind/temp with the tropopause wind/temp at the current aircraft altitude. There are a few known issues where it doesn't allow the PSX to smooth the weather (it updates every 5 seconds from P3D.) So bearing in mind the P3D winds are smoothed, the injected winds should be also.

There is anther tool currently in development which has more advanced features, but it hasn't been released yet.

Kind Regards

Craig
Moderator
Registered: May 2009
Posts: 5140
Hi Craig,

there is a new Qi available that allows PSX to smooth your weather injections:

http://aerowinx.com/forum/topic.php?post=30006#post30006


Cheers,

|-|ardy
Member
Registered: May 2009
Posts: 131
Location: Reutlingen, Germany (near EDDS)
Hi Craig,

> There is anther tool currently in development which has more advanced features,
> but it hasn't been released yet.

by chance, will that other tool talk to FSX as well?

I haven't been following this topic, but is there any available tool at all which will read FSX upper winds and inject them in PSX?

Markus
Member
Registered: Jun 2015
Posts: 2
Thanks Hardy, it has already been incorporated in the latest (unreleased) version.. and it seems to be working well.

Markus,

Ideally I do intend in adding FSX support, it's something I tried to do with version 1.5 but didnt have much luck... so hopefully I will get it working in the latest version.

Kind Regards

Craig
Member
Registered: Dec 2011
Posts: 497
Location: Portugal
Does this mean that, having ASN feeding P3V2 with it's weather, this will get injceted into PSX too, overriding PSX's native weather engine ?
_______________
Jose Monteiro
Member
Registered: May 2009
Posts: 149
Just a related comment. I've been interested for some time, as an intellectual exercise, in matching up PS1/PSX world weather with flight planning software. I usually fly with real time wxr on in PSX and have found that, based on Hardy's suggestions of how the jetstream model works for world weather that I can get pretty good agreement for winds aloft by using PFPX winds at my projected cruise altitude (taken from real world predicts for when flight plan is produced) and tweaking the PSX jetstream troposphere "Wind at Tropopause numbers to more or less match that (assuming flight levels are in that range). Actually gives results close to real world uncertainties in many cases - and tests your contingency fuel planning for unexpected winds when they don't :).

Further comment - PSX's seasonal jetstream model is damn good for most purposes.

Cheers,
Torrence
Member
Registered: Aug 2014
Posts: 60
jcomm wrote
Does this mean that, having ASN feeding P3V2 with it's weather, this will get injceted into PSX too, overriding PSX's native weather engine ?


Yes. The only issue being that thunderstorms will display on the ND in a different place to where they appear out of the window.

torrence wrote
Just a related comment. I've been interested for some time, as an intellectual exercise, in matching up PS1/PSX world weather with flight planning software. I usually fly with real time wxr on in PSX and have found that, based on Hardy's suggestions of how the jetstream model works for world weather that I can get pretty good agreement for winds aloft by using PFPX winds at my projected cruise altitude (taken from real world predicts for when flight plan is produced) and tweaking the PSX jetstream troposphere "Wind at Tropopause numbers to more or less match that (assuming flight levels are in that range). Actually gives results close to real world uncertainties in many cases - and tests your contingency fuel planning for unexpected winds when they don't :).


This is pretty much what Craig's program does, except it updates it every 5 seconds rather than when you get around to it. This can be particularly important in long haul flights through changeable weather.
Member
Registered: Apr 2012
Posts: 68
Hardy Heinlin wrote
METAR wind data refer to surface winds only. In PSX there is a parameter for the surface wind and one for the tropopause wind. The wind in between is smoothly interpolated based on a geographic algorithm as described on page 42.

Before I repeat here on the forum what I wrote in the manual, could you tell me which part in the manual you didn't understand? It's not always easy for me to understand what the user doesn't understand :-) and so I sometimes end up writing overly long answers to a question I misunderstood :-)


Dear Hardy,

apologies for a delayed reply, but I was away from my main computer. Nonetheless, I have re-read the manual (page 42) as well as what you have written.

My main purpose of asking the question, was to make sure that I am not doing something wrong (i.e. have a wrong setup). I was puzzled by the fact that when the Internet is ON (option #1), then winds in the tropopause correspond to PFPX's estimates. At such instance everything works perfectly (as witnessed by Torrence).

However, when I select the the Internet OFF, then winds in the tropopause do not change with updated locations/new zones on the flight track (at least that's what I have encountered so far). Should not they update/adjust with the new zones? On my example flight from ZSPD to EDDK, wind was always ~260/70 (it did change slightly, but general direction was unchanged), despite the fact that winds were expected to change.

For the time being, I will conduct more flights and will report my findings (maybe there needs to be more samples on my side). :)

I am very sorry, if I seem obtuse, but sometimes the obvious information passes by. :oops: All the of the above is my observation only, and I am certain that the problem is on my side. I am just trying to enable PSX on my machine to be as perfect as it was designed. :)

craig1231 wrote
There is a tool developed by myself which injects weather from P3D into PSX; if you'de like to give it a try


Dear Craig,

thank you very much for the heads up. I am definitely planning to add P3D as my scenery generator (and hopefully one day) build me a fixed based sim. :) And I will surely look into the PSXWx. But for the time being, I am MacBook flier. 8) Nonetheless, the amount of add-on software for/by Simfest FBS team is astonishing.
_______________
Andrej
Moderator
Registered: May 2009
Posts: 5140
Hi Andrej,

the difference you noticed regarding "Internet weather on/off" is more or less a random effect.

The METARs will update the local zone weather variables only, not the tropopause winds.

The wind variables in the local zones contain surface wind data only.

The wind data between the local surface and the local tropopause is automatically interpolated by PSX using certain algorithms. As you fly along, wind data changes dynamically because the ground elevation varies and because the tropopause varies and because your position varies.

When you are flying below the tropopause, you are flying in an interpolated data world. The lower you are, the more METAR wind is involved. The higher you are, the more tropopause wind is involved. At or above the tropopause, no METAR wind is used.

When you use the jetstream system and click on "Randomize for [the current month]" you'll get a random jetstream constellation that represents a typical, season related average constellation. As this includes random variations to a certain degree, it may more or less agree with current real-world weather; and the "Internet weather on/off" effect will just add more variation -- but if this variation increases the precision in the current situation, then it's just pure coincidence.


Cheers,

|-|ardy
Member
Registered: Dec 2011
Posts: 497
Location: Portugal
Reading again through this thread I found something "new" :-)

In Post #2 HH explains the 3 variants of setting weather, and I thought there was yet a third one, where we do not enable METAR download at all, and the METARs are generated based on PSX's own weather model.

I thought this was possible - using for instance the jetstream constellation model, and disabling the METAR download.

Actually, for long flights I would rather prefer to use a completely independent weather generation. I don't mind if it agrees with the RW data for the date of the flight, provided it is sound, and I have actually used this approach ( by disabling Internet METAR download ) and was convinced that the area observations resulted from PSX's own weather model ?

Well, I really should take some time, even if a few minutes every day, to carefully read the whole Manual pages :-)
_______________
Jose Monteiro
Member
Registered: May 2010
Posts: 843
What Hardy is really trying to say....

https://www.youtube.com/watch?v=pllRW9wETzw
Moderator
Registered: May 2009
Posts: 5140
Exactly :-)

And then rocket up into the stratosphere: https://youtu.be/PRQImTprMgM
Member
Registered: Dec 2011
Posts: 497
Location: Portugal
Well, now you really got me :-/

I though HH was just a nickname for "Clyde", Mark Mardon's brother ..... Well... I really have to read the Manual pages ..
_______________
Jose Monteiro
« Last edit by jcomm on Sat, 15 Aug 2015 16:47:38 +0000. »
Member
Registered: Dec 2009
Posts: 334
Location: KTPA
Craig-

I was hoping for a pointer. I run vPilot Traffic Proxy, P3d, PSX, VisualPSX and TrafficPSX on the main computer. Then I run a PSX Client, ASN, vPilot, PFPX eat on a second machine.

Do I have to run PSXWx on the main machine, or is it possible to run it on the second? I tried, however, it didn't connect. I used Port 6120 for the P3d port (not sure about this).

I appreciate any thoughts! Ta!

C
Member
Registered: Apr 2012
Posts: 68
Dear Hardy,

thank you very much for taking your time and explaining the weather engine some more.

Hardy Heinlin wrote
The wind data between the local surface and the local tropopause is automatically interpolated by PSX using certain algorithms. As you fly along, wind data changes dynamically because the ground elevation varies and because the tropopause varies and because your position varies.

When you are flying below the tropopause, you are flying in an interpolated data world. The lower you are, the more METAR wind is involved. The higher you are, the more tropopause wind is involved. At or above the tropopause, no METAR wind is used.


Now I have a better understanding and deeper appreciation how complex the PSX is. So the best way is to leave the jet-streams to randomize based on current season, or adjust them accordingly latest maps/data/forecast?

Hardy Heinlin wrote
the difference you noticed regarding "Internet weather on/off" is more or less a random effect.


Then you have done an amazing job, as the interpolation is quite good. On all of my flights this has been very close to forecast (not 100% precise, but I do not expected that either).

I recall using SimWX in the PS1.3 days, it was very good add-on. But I not sure how that worked (did it used tropopause data?; is such data publicly available and regularly refreshed?). Time to read/study and hopefully give back to the community. :)

Thanks again for everything!

Cheers,
_______________
Andrej
Moderator
Registered: May 2009
Posts: 5140
Hi Andrej.

andrej wrote
So the best way is to leave the jet-streams to randomize based on current season, or adjust them accordingly latest maps/data/forecast?

I would use the randomizer. Additionally, if you have a wind map that shows the current real-world jetstream trajectory and speeds, then you could fine-tune PSX's jetstreams accordingly.

Why does it match the real world data so well? Because ...

- The earth always rotates from west to east, and always cycles through the same seasons, always generating relatively regular corridors of planetary winds, always having warmer air and a higher tropopause near the equator and vice versa near the poles -- which is all the case in PSX too --, so the average global weather is pretty constant from year to year.

- Short flights are not much affected by wind changes: a big wind forecast error causes only a small ETA error because you will land soon after take-off.

- Long flights are more affected by wind changes, but a long flight also encounters a more averaged amount of both negative and positive errors, which statistically compensate each other on a long flight. Long flights encounter a more average global climate rather than a local special weather.


Cheers,

|-|ardy

Page: 1

Old 744 Forum is powered by UseBB 1 Forum Software