1 hour ago, simbol said:

I haven't seen any reports of such bug in the private beta forums.. Which is surprising.. So maybe there is something else behind it or REX / HiFi reported it privately..

Regarding REX and HiFi weather apps. These guys always push the boundaries of the simulator as far as they can.. I am sure they are at this point exploring the limits in order to implement new stuff. I would also like to remind you guys a Hotfix is coming so it makes no sense to release a product update now to then have to do it again after the Hotfix!. Many are holding up for this reason, including myself..

Please give these companies the space to deliver the quality of products your are hoping to receive.

Regards,

Simbol 

The bug with wind direction was noted by the developer of Opus FSI after he made his weather engine capable of connecting to V5. I have confirmed it on my own system using that product, as well as FSXWX and Rex Sky Force.

Basically, if wind is injected into the sim from an external weather program at a constant direction and velocity, the velocity is fine, but the direction will intermittently and instantly shift left and right of the target value - sometimes by up to 90 degrees. This happens on the ground and in flight. If the wind velocity is low, and one is flying a relatively slow GA aircraft like the default Rockwell Commander, it can be an annoyance, but the aircraft will handle the wind shifts relatively well. At higher altitudes, with high speed winds, the sudden directional shifts can make the aircraft completely unflyable.

This happens even when using default clouds with Enhanced Atmospherics not enabled. 

It’s not just a problem with winds injected from an external program. Even wind direction embedded in one of the default weather themes shows sudden shifts. The default “Fair Weather” theme should have a wind layer that gives a constant velocity of 25 knots and a constant direction of 270 degrees when flying at high altitude.
 

I did a 2-hour flight at FL350 in the FSW Lear 35 with the Fair Weather theme active, (no external weather program running) while monitoring the wind direction and velocity in the info gen display. The velocity remained constant, but the direction would suddenly shift every now and then, causing the aircraft to yaw side to side whenever this would occur. I have never seen winds embedded in a pre-packaged theme change either direction or velocity in any previous version of P3D.

This appears to be a bug within the internal weather model. Perhaps it was only introduced in a late build just before release. 

12 minutes ago, JRBarrett said:

The bug with wind direction was noted by the developer of Opus FSI after he made his weather engine capable of connecting to V5. I have confirmed it on my own system using that product, as well as FSXWX and Rex Sky Force.

Has this developer reported it to LM on their forums? I am just saying it is very odd I saw nothing regarding this on the "Private" beta forums, specially because well known weather injection developers are part of the group.. maybe you need to use the V5 SDK to inject weather, maybe you need to inject the weather in a different way, etc... lots of companies are still updating their software for V5..

Many people are just copying and pasting add-on's from V4 to V5 in the "hope" everything works..

S.

Ok I found OPUS Post.. https://www.prepar3d.com/forum/viewtopic.php?f=6311&t=136929

From that post I can tell he is using PDK services from the older version of the SDK (4.3 -- IWeatherStationV430) there is a new weather service for V5 (IWeatherSystemV500) so there is a chance the latest version is required for this to function correctly.

There is also the possibility that SimConnect functions are working correctly as long as you use the latest SDK as I know HiFi uses SimConnect instead of PDK.

So my point it.. let's not jump into conclusions, there are many variables here at play..

Regards,
Simbol

13 minutes ago, simbol said:

Has this developer reported it to LM on their forums? I am just saying it is very odd I saw nothing regarding this on the "Private" beta forums, specially because well known weather injection developers are part of the group.. maybe you need to use the V5 SDK to inject weather, maybe you need to inject the weather in a different way, etc... lots of companies are still updating their software for V5..

Many people are just copying and pasting add-on's from V4 to V5 in the "hope" everything works..

S.

The Opus developer reported it in the public SDK forum at LM, but there have been no responses to his post.

Very possible that something has changed in how external weather engines will interface going forward. I think Opus has always used standard SDK techniques to inject weather, while I’m pretty sure that AS uses a custom interface that goes beyond what the SDK offers.

However, all other weather parameters injected using the same SDK techniques that have been used in the past still work fine. Cloud cover, visibility, temperature, barometric pressure and wind velocity will match the current METAR perfectly. It is only wind direction that is unstable.

I know that there are some changes in the weather engine with True Sky enabled. Only one layer producing precipitation is possible with True Sky, and apparently cloud layers can no longer overlap, though that has always been possible in the past.

The thing that concerns me is that I see unstable wind direction even using the canned “fair weather” theme, where the wind should always be a constant 270 degrees at 25 knots at altitude. The only add-on I am currently using is FSUIPC. Since FSUIPC can affect weather, possibly it is having an affect - I can try removing it to see what happens.

The easiest way to verify if it is happening on your own system would be to go to “standard” weather (not enhanced atmospherics). Load the default “fair weather” theme. Use an aircraft capable of flying at higher altitude, like the F-35. Establish a level cruise at say FL 350, then monitor the environmental wind by bringing up the red info gen display in the upper left corner. What I am seeing is a constant velocity of 25 knots, but the direction is not constant. It will occasionally flip to something other than 270 degrees.

 

32 minutes ago, simbol said:

Ok I found OPUS Post.. https://www.prepar3d.com/forum/viewtopic.php?f=6311&t=136929

From that post I can tell he is using PDK services from the older version of the SDK (4.3 -- IWeatherStationV430) there is a new weather service for V5 (IWeatherSystemV500) so there is a chance the latest version is required for this to function correctly.

There is also the possibility that SimConnect functions are working correctly as long as you use the latest SDK as I know HiFi uses SimConnect instead of PDK.

So my point it.. let's not jump into conclusions, there are many variables here at play..

Regards,
Simbol

I’ll mention the requirement to use the V500 PDK functions to the Opus developer at his support forum. That may have a bearing on the problem. The engine also has the ability to not use PDK, by simply creating a custom theme - but the directional shift still occurs. 

As mentioned in the my previous post, I will remove FSUIPC (currently at version 6.03), to see if I still see directional instability using the default fair weather theme.

59 minutes ago, JRBarrett said:

As mentioned in the my previous post, I will remove FSUIPC (currently at version 6.03), to see if I still see directional instability using the default fair weather theme.

I had the problem before installing FSUIPC and still do.  The wind problem definitely seems to be affecting the canned weather themes too. 

Lately this forum looks more like FSX forum in 2008.