Hi All,

 

Sorry I have not been able to spend too much time on this lately. However, I did test the latest version this afternoon and was pretty impressed. I am not sure if it is because it has been a long time or not bu the latest version seems to work much better that some before. The Ai Separation looked better.

 

I did notice some weird behaviours but not as much as in the previsous versions. The one thing that surprised me at the very begining is that it took a while for the arriving aircraft to show up (using FlightSim Commander), much longer than when tester before some time ago.

I also noticed one aircraft reaching the last waypoint of a STAR and go in a holding pattern (no mention of the command in the programm window) while there was no other aircraft in front . I might have missed some thing here

I havent started using final approaches again yet, as even with very precise coordinates for waypoints, some behaviors were still erratic.

 

Still, more and more wonderful program!

 

Thank you

 

BR

Well, after a couple of hours in EDDF i´m still unsure :mellow:

 

No doubt, the assign of the STARs regarding the Wind direction now works great.

Not perfect is, that it now happens that the AI follow e.g. a STAR dedicated to the RWY 07L and finally the ATC assign the RWY 07R.  That´s not perfect because obviously this swing overs crack the separation because the final is short.

 

What me really leave desperated is the fact, that it seems to be impossible to "trust" the ATC.

I wrote long spreadsheets of where P3D asssign the Runway (IAF).

Now you can think, it is simple to "optimize" the STARS and let each AI pass this Area.

At the certain day i noticed, that all Traffic of a certain STAR receive their Runway between Fix DF439 and DF444. Great.

 

Now you go to work 5 days and start the sim again one week later. What happens? 

From 10 Aircrafts only 3-4 get the Runway in the usual area, 6-7 NOT :angry: .

And you know what happens when you have a "sixpack" of Airliners entering the Final without clearance <_< .

 

Looking at the ATC window show tons of messages like "25 expedite your climb to 12,000 ft", "1048 expedite your climb to 15,000 ft"

It´s all the traffic that was failed to assign the runway while the AI in front at the same track did.

 

I think, as long as the fu...g built in ATC will not be replaced by a more smarter your tool will have a hard stand.

 

But also with assigned Runway it happend often, that the AI turned back at short final.

First it had a STAR and a Runway, after leaving the last STAR WP the Runway was left. After passing the first one or two Final WP the AI turned and the Runway wasn´t shown at the monitor as well. So the AI was wandering arround. This happend randomly for about 15-20% of the landing Aircrafts.

The next step is to find out how to force fsx atc to say "clear to land" so the ai planes actually land without issue. there has got to be a way...

Each time it has happened with a new .ini file:

 

Mike, what .ini file were you using?  It looks like a problem with one the files, a missing xml tag pair.

 

 

No doubt, the assign of the STARs regarding the Wind direction now works great.

Not perfect is, that it now happens that the AI follow e.g. a STAR dedicated to the RWY 07L and finally the ATC assign the RWY 07R.  That´s not perfect because obviously this swing overs crack the separation because the final is short.

 

Yes, I noticed that as well.  I've fixed that problem.  The "fast runway search" version you're using also locked onto one runway at airports with multiple active landing runways (I was afraid of that).  This has been fixed as well.  New version should be uploaded shortly.

 

 

 

 

 

Now you can think, it is simple to "optimize" the STARS and let each AI pass this Area.

 

Exactly, that's why I refer to it as a FSX (or P3D) generated IAF.  The sim is using an algorithm to determine  the waypoint.  You will also notice that sometimes the AI isn't even required to intercept this point.

 

 

 

 

I think, as long as the fu...g built in ATC will not be replaced by a more smarter your tool will have a hard stand.

 

I use RC4 most of the time, haven't used default ATC in quite awhile.

 

 

 

After passing the first one or two Final WP the AI turned and the Runway wasn´t shown at the monitor as well. So the AI was wandering arround.

 

 

I'll keep a lookout for that (I'll check EDDF), but I've only noticed this behavior when pausing/unpausing the sim. 

 

-Roland

 

    

New Rev. 1.2Q (beta) can be found here: 

 

http://www.mediafire.com/download/7oh5x7c3omedvlo/AISIDSTAR12Q.zip

 

Revised:  This version incorporates a previous alternate version that quickly propagates the most recently detected landing runway to the arriving AI for STAR assignment.  This alternate version however did have a tendency to lock onto a single runway at airports with multiple active runways (e.g., parallel runways).  This has been fixed.  AI will use all active runways.

 

Revised:  Revised AI separation code on short final.  Also introduced a limited go-around mode if separation can't be obtained by 500 ft AGL.

 

Fixed:  Eliminated following error:  EXEPTION = 20  = "six digit number" index = 1chData = 24

 

Important Revision:  A registered version (4.9x or later) of Peter Dowson's FSUIPC utility is recommended to use the full functionality of AISIDSTAR.  For example, the new AICull function requires FSUIPC.  For networked configuration users, Peter's WideFS (ver. 6.9) is also recommended.

 

Important New NOTAMS!

 

ENTERING FINAL APPROACH COORDINATES.  Do not use FSX when entering final approach coordinates for runway positions (e.g., touchdown, rollout) as the FSX coordinate display lacks the necessary precision (significant digits).  Use the freeware Airport Design Editor X (or similar) instead.

 

ENTERING RUNWAY DATA.   Ensure the runway number corresponds to the FSX runway description (for whatever airport scenery you are using).  Often, the latest FMS data will reference changed runway numbers that do not correspond to stock FSX runway descriptions.  

 

Please consult the following video documentation:

 

Overview of What's New in Version 1.2Q:

 

http://www.mediafire..._in_Ver_12Q.mp4

 

 

Video 1 - Working with the Included Sample SID/STAR Files and Creating New Files Using AIConv.exe:  

 

http://www.mediafire...sing_AIConv.mp4

 

 

Video 2 - Custom Editing SID/STAR Files - 

 

http://www.mediafire..._STAR_Files.mp4

 

 

Video 3 - Operating Parameters.

 

Part 1: http://www.mediafire...ters_Part_1.mp4

Part 2: http://www.mediafire...ters_Part_2.mp4

Part 3: http://www.mediafire...ters_Part_3.mp4

Part 4: http://www.mediafire...ters_Part_4.mp4

Part 5: http://www.mediafire...ters_Part_5.mp4

 

 

Thanks!

-Roland

 

New Rev. 1.2Q (beta) can be found here: 

 

Mike, what .ini file were you using?  It looks like a problem with one the files, a missing xml tag pair.

 

    

 

 

Just the latest one that was provided (not the one you just put out). It also happened in the version before. But it is funny because I didn't get the error when I open the .exe for the first time, but only happens later on.

I have experienced an issue.  Using UT2 at KSEA i cant seem to get the AI to roll out on landing more than 25% of the runway.  Even 777's and 767's are landing and stopping before the halfway point in the runway.  This leads to airplanes turning around and using the opposite direction high speed turnoffs.  I have tried playing with the "roll out" value, even pushing it up to 20.  Nothing seems to force these planes to roll out a realistic distance on landing (ideally between 50 and 75% of runway length.)  FYI I have been using the autogen landings not creating my own approaches.

 

Anyone else seeing this or have solutions?

FYI I have been using the autogen landings not creating my own approaches.

 

 

I checked out KSEA and saw what you were reporting (AI stopping short and using the opposite turnoffs).   However, this was happening to AI while under the control of FSX. Your autogen file was recorded during one of these short FSX rollouts, so the autogen file exhibited the same issue.

 

The default AFCAD for this airport is pretty bad.   Most of the AI were landing/departing runway 16L causing long backups.  If you're using stock, I would look for some better AFCAD/ADE files in the library.

 

Anyway, you can't change the rollout distance of the autogens because they represent recordings of FSX final approaches and rollouts.  Instead, tweak the autogen to make it a custom file.  Remove the word "autogen" from the pattern name, e.g.:

 

     <patternname>

          Final Rwy 16L

     </patternname>

 

Then make the following changes to the waypoints:

 

<waypoint>

     <waypointnumber>

          7

     </waypointnumber>

     <fixname>

          SECOND ROLLOUT WAYPOINT

     </fixname>

     <latitude>

          47.449967

     </latitude>

     <longitude>

          -122.307846

     </longitude>

     <altitudefeet>

          439.9668387209997

     </altitudefeet>

     <altitudemeters>

     </altitudemeters>

     <speedknots>

          130

     </speedknots>

     <comments>

         coordinates manually modified to be at midpoint of runway

     </comments>

</waypoint>

<waypoint>

     <waypointnumber>

          8

     </waypointnumber>

     <fixname>

          END ROLLOUT WAYPOINT

     </fixname>

     <latitude>

          47.433897

     </latitude>

     <longitude>

          -122.307986

     </longitude>

     <altitudefeet>

          439.9668387209997

     </altitudefeet>

     <altitudemeters>

     </altitudemeters>

     <speedknots>

          130

     </speedknots>

     <comments>

          manually added waypoint 8 to end of runway to increase rollout distance

     </comments>

</waypoint>

 

 

Then, save the custom file to your KSEA folder.  The rollout should be pretty long as is, but you can change various parameters mentioned in the readme to make it longer/shorter.

 

-Roland

Hello guys,

 

Little question: does anyone us AISIDSTAR with VoxATC?

Is the AI controlled by AISIDSTAR or VoxATC?

 

Thanks,

Hi Roland,

 

I am reverting to you again with my question regarding the AI behaviour after it exits the last point of the star. Here in the new airport I am designing, the VOR approach makes the AI overfly the VOR on terrain 5so they overfly the airport at 4000ft) then descend on a heading before turning right to align with the runway.

 

I am not sure it is doable with the final approach system of the program, so I wrote the approach as a prolongation of the STAR:

http://www.philskies.net/library/Charts/New/RPVM.pdf

See on this link the VOR rwy 04.

 

I created two dummy waypoints for the descent to 1500 and the turn to the right to align. The thing is they do align (mos of the times with correct altitude and speed) but then circle a wpt called FS04 in FSX (which seems to be a kind of FAF here), regain some altitude, realign again and then land.

 

It also happens sometimes to aircraft on another STAR coming in as a straight in approach.

 

Do you have any recommendations on how to write that type of approach? (With and without using the final to see the difference. Here I dont see how they could first overfly the FSX algorythm generated IAF before commencing the approach over MCT).

 

Thank you.

 

Best Regards,

Renaud

Is the AI controlled by AISIDSTAR or VoxATC?

 

I don't have the program, but from what I've read, VoxATC lines up AI onto final approach for atmospheric effect.  The AISIDSTAR AI are exiting real STAR(s) and joining onto real final approaches (custom), so they probably wouldn't mix.  It would be pretty easy for me to add a option to delete AISIDSTAR AI when they exit the STAR or gets close to (e.g., within 30 nm of) the monitored airport.  Regarding SIDs, AISIDSTAR won't control departing AI if no SID files are specified, however what you're left with is just departing VoxATC AI, which I don't think follow SID procedures (although there may VoxATC ATC "communications" that they do, again for effect). I would like to hear from anyone who has tried to run the two programs as well.

 

-Roland

 

 

Do you have any recommendations on how to write that type of approach? (With and without using the final to see the difference. Here I dont see how they could first overfly the FSX algorythm generated IAF before commencing the approach over MCT).

 

I wrote some files for you:  http://www.mediafire.com/download/vobeaun26bnr7cs/RPVM.zip

 

Just STAR 15 and custom final approach files for runways 4 and 22.  However, you can probably extend the concept to the other STAR(s) to handle AI approaching from all bearings.

 

I extended the STAR to the point where the AI exits the downwind (right before the AI turns onto final).  The AI overflies the airport, makes a procedural turn, and joins the downwind at a 45 degree angle.  The AI then exits the "extended" STAR at the end of the downwind and turns onto base to intercept the FSX IAF, which fortunately is in more or less the correct location relative to base.  After the AI intercepts the IAF, the custom final approach file is selected and the AI lands.  The pattern isn't as "tight" (e.g., within 2 nm of the runway) as specified in the noise abatement plate, but it seems ok.

 

The ILS approaches (including the circle to land portion) should work too if you decide to write them.  The AI would finish the "circle" and exit the STAR, intercept the IAF, then go onto final.

 

Note: if you add AIImmediateForceLandAll  = 1 to your .ini, you won't have to worry about designing these patterns to intercept the FSX IAF, however the trade-off with this option is the AI won't get taxi service after they land and thus are automatically deleted after roll-out (otherwise they would block the runway).

 

-Roland.

New Rev. 1.2Q (beta release 12) can be found here: 

 

http://www.mediafire.com/download/enizokcezydgdgs/AISIDSTAR12Q.zip

 

Fixed:  Still problems with arriving AI locking onto one landing runway at airports with multiple, active runways (e.g., parallel landing runways).  This should be fixed.  Thanks Ronald for bringing this to my attention.

 

Fixed:  Program incorrectly in some cases selects landing runways based on weather data.  Note:  runway selection (for STAR assignment) based on weather data only occurs if there is no actual landing data at the airport yet available (e.g., your user aircraft is approaching the airport, but it is still outside your FSX reality bubble).  Once landing data become available however, the AI will have their STAR(s) reassigned if necessary (this should be the exception) to match the actual landing data.

 

Important Revision:  A registered version (4.9x or later) of Peter Dowson's FSUIPC utility is recommended to use the full functionality of AISIDSTAR.  For example, the new AICull function requires FSUIPC.  For networked configuration users, Peter's WideFS (ver. 6.9) is also recommended.

 

Important New NOTAMS!

 

ENTERING FINAL APPROACH COORDINATES.  Do not use FSX when entering final approach coordinates for runway positions (e.g., touchdown, rollout) as the FSX coordinate display lacks the necessary precision (significant digits).  Use the freeware Airport Design Editor X (or similar) instead.

 

ENTERING RUNWAY DATA.   Ensure the runway number corresponds to the FSX runway description (for whatever airport scenery you are using).  Often, the latest FMS data will reference changed runway numbers that do not correspond to stock FSX runway descriptions.  

 

Please consult the following video documentation:

 

Overview of What's New in Version 1.2Q:

 

http://www.mediafire..._in_Ver_12Q.mp4

 

 

Video 1 - Working with the Included Sample SID/STAR Files and Creating New Files Using AIConv.exe:  

 

http://www.mediafire...sing_AIConv.mp4

 

 

Video 2 - Custom Editing SID/STAR Files - 

 

http://www.mediafire..._STAR_Files.mp4

 

 

Video 3 - Operating Parameters.

 

Part 1: http://www.mediafire...ters_Part_1.mp4

Part 2: http://www.mediafire...ters_Part_2.mp4

Part 3: http://www.mediafire...ters_Part_3.mp4

Part 4: http://www.mediafire...ters_Part_4.mp4

Part 5: http://www.mediafire...ters_Part_5.mp4

 

 

Thanks!

-Roland

One single question:

 

Do you have control about the landing gear during take off?

 

I noticed as i tried to set up "realistic" prucedures for TNCM (St. Maarten) that i need to handover the control to AISIDSTAR very early (at the latest 300ft over ground). But, doing this that means that the Aircrafts don´t have the time to fully retract the landing gear. Smaller Ai´s slightly better, the KLM 747 continues the climb with landing gear 3/4 outside. 

Not a showstopper, but it is looking strange though :rolleyes: .

I would like to hear from anyone who has tried to run the two programs as well.

 

 

Hello Roland

 

I've been testing AISIDSTAR with VoxATC, to see what happens.

The only problem I see is that VoxATC is injecting some traffic with VFR rules, without any departure or destination.

As I was watching the monitor program these AI never fell into AISIDSTAR control.

As far as I can see, you will never be able to control these.

 

Here is an screenshot. Here you can see that the problem is that VoxATC is injecting a jet aircraft with VFR rules. So this AI will land without the possibility for AISIDSTAR to take control over it.

Hope it helps, tell me if I can do anything else.

 

Any reason why the link is dead?