Wind variations

[Guest]

Hi DamianLat weekend I flew my first RW ILS approach with screens up in a PA28 - quite an interesting experience!This made me think about how FS and RW winds seem to differ, and if I'm right, it might be something to consider for v2.0.During the approach the wind direction changed a little, so I had to keep working the heading to correct. In the Northern hemisphere, I think the wind backs by about 25 to 30 degrees during the last 1000 feet, and also the strength at circuit height is approximately twice that at ground level.I'm not close to how FS2002 or FS2004 model winds in this respect, but each approach I make in FS seems rather more "dead" and predictable than the RW. (Also, using FS2002, I'm not clear of the relationship between the value set in the Weather Wind section, and what is shown on screen with CTRL z!)Could Activesky 2 have a feature to allow the user to select a degree of random variation in the wind speed/direction around the ATIS/TAF figure? (ie much slower changes than FS generates through it's "turbulence" feature). Also, would it be practical to model the veering/backing effect and changes in wind strength from ground level up to circuit height?Apologies if I've misunderstood anything about how FS and Activesky model winds, but I was hoping these thoughts might help improve realism.RegardsAndy

[Damian Clark 1,144]

Hi Andy,Congrats on your first ILS approach! Wait until you do it in real IMC down to minimums :) A great experience I will always remember...Yes the surface friction effect can be modeled - where winds tend to parallel isobars aloft but gradually tend to shift 90deg to the right or left depending on hemisphere.. and slow when they meet the surface. This wasn't practical in the existing wxRE model because it was designed with FS2002 global weather in mind, and we needed the actual winds at the time of approach to correspond with the landing winds (for proper ATC runway selection and approach planning).Now with FS2004 we have more control... We no longer need to use "lock/force" modes. Instead, the destination station weather is set appropriately for both surface and aloft conditions, and this may be quite different a little further away from the airport on your approach... Now the good news! AS2's design already supports this (mostly). We are now using high-resolution WORLD FORECAST MODEL winds simulation, coupled with FS2004 station-based weather. This will automatically represent the surface friction effect. The only problem is that the transition between the aloft and surface levels would be abrubt, unless addition layers were used for the destination airport to provide smoothing transitions. We could probably provide 5 layers to go between the 3000ft and surface altitudes, with less noticeable abruptness between these layer changes (1/5th). I will try it out later in the development cycle and perhaps we can get something like this working...Thanks a bunch for the input... A great idea!-Damian

[Guest]

DamianThanks for the quick response - excellent news! :DRegardsAndy

[Guest]

DamianA further thought, as I noticed a similar theme about adding wind layers in http://forums.avsim.net/dcboard.php?az=sho...g_id=4382&page=IF too may wind layers turns out to give a FS2004 perfomance hit:(, then would it would be possible for AS2 to detect the aircraft altitude, and only create wind layers around that altitude :-hah?(eg if I'm at 2000 feet, then any FS2004 processing to calculate multiple wind layers around at FL350 seems pretty academic!)If performance was an issue, might this be a way to dynamically set wind layers only where they are needed, and minimise any compromises due to limitations in the number of wind layers?RegardsAndy

[Damian Clark 1,144]

Hi Andy,I don't think that will work - FS2004 does too many automatic manipulation / smoothing processes with winds and replacing the entire wind parameter set with different altitudes is not something you can do in a small-interval loop. It would result in a huge performance hit/stutter everytime you change altitude and strange in-sim wind values would prevail.So far using many winds layers has not resulted in any kind of performance hit that I can tell, but then again I haven't tried huge amounts of AI traffic and turbulence yet....-Damian[table border=0" cellspacing="30" cellpadding="0][tr][td align = "left"]Damian ClarkHiFi Simulation SoftwareDeveloper of ActiveSkyThe next-generation weather environment simulation for FS2002!http://hifi.avsim.net/activesky[/td][td]http://hifi.avsim.net/activesky/images/wxrebeta.jpg][/td][/tr][/table://http://hifi.avsim.net/activesky/ima...[/tr][/table