January 15, 201412 yr Just a small bit of early feedback. Standard bgl traffic used, testing KATL, keeping things simple with only 4 hand written SIDS, 4 hand written STARS, and 6 final procedures. I am not seeing any SIDs assigned for new departures. SIDs are being assigned on load to close spawn AI, and after some time SIDs are sometimes assigned to new departures but not until very late. The only one I have been able to catch so far wasn't assigned a SID until having reached ~9000ft, not sure if that means anything to you. For arrivals, STARs seem to work as they always have, and approaches are indeed working. I do think you might be a bit too aggressive with speed control however, to the point that the user wouldn't be able to fly in the pattern. At one point I noted 5 medium jet a/c on approach to a runway all at ~70kts. Not really sure that final approach is the stage to be that aggressive with speed control. That said, progress is progress, and this is a major step forward. Maybe a few tweaks to de-rotation (if possible) and a less aggressive speed intervention on final is about all I could say for now. If you have any ideas what I might have screwed up with the SIDs please let me know, and thanks again, Roland. Regards,Brian Doney
January 15, 201412 yr Thanks guys for the positive feedback. Now that I pushed Q is out the door, I can get back to answering questions! Sorry for the delays. For example in FSX is RWY 07 active. AISIDSTAR is chossing after starting FSX RNAV TRANSITIONS for RWY 25 and I got goarounds due to bad runway heading, cause 07 is active. Klaus, You can turn off "bad heading" checks by adding this parameter to your AISIDSTAR.ini file: AIAntiOffAngleApproach = 0 The first few STAR assignments might be incorrect based on stale or no data about runway in use, but after a few minutes the program should start selecting STARs corresponding to the runways in use. You might also want to try a ver. Q custom, final approach file, which will lock the AI approach down after it leaves the transition, unless FSX is late assigning a runway (then the AI will circle hold for a bit). Regardless, the AI will eventually land on the runway in use. When a custom final approach file is used, the program doesn't check for bad headings or anything like that because the user-generated waypoints in the file trump. Remember however that after exiting the STAR, the AI will intercept the FSX-generated IAF, then the final approach becomes active. I am not seeing any SIDs assigned for new departures. SIDs are being assigned on load to close spawn AI, and after some time SIDs are sometimes assigned to new departures but not until very late. The only one I have been able to catch so far wasn't assigned a SID until having reached ~9000ft, not sure if that means anything to you. Brian, please send your SID file to my support address and I'll take a look. Andiqui, I still need to look at your files as well to verify the problem didn't migrate to ver. Q. I do think you might be a bit too aggressive with speed control however, to the point that the user wouldn't be able to fly in the pattern. At one point I noted 5 medium jet a/c on approach to a runway all at ~70kts. Not really sure that final approach is the stage to be that aggressive with speed control. Probably, I'm looking for some good default values. I tend to test with insane amounts of AI to see what breaks, so perhaps the default speed reductions are overdone for normal AI levels. You can reduce AI speed reductions on final approach (caused by traffic conflicts) by adding this parameter to AISIDSTAR: AIfinalapproachslowdownrate = x (note: default is 1.5 - a lower number means a smaller speed reduction, but stay at 1 or above). -Roland
January 15, 201412 yr Only main issue I have is having aircraft approach at 63 knts. Also, AI seem to be stuck in their final approach phase for quite awhile. How better can I streamline their arrivals? This is at LAS.
January 16, 201412 yr AISIDSTAR keeps crashing on me. Using version Q. Seems to crash when I use my converted PMDG SIDSTAR files but worked fine with everything default from the Q download. Problem signature: Problem Event Name: APPCRASH Application Name: AISIDSTAR.exe Application Version: 0.0.0.0 Application Timestamp: 52d62b8c Fault Module Name: AISIDSTAR.exe Fault Module Version: 0.0.0.0 Fault Module Timestamp: 52d62b8c Exception Code: c0000005 Exception Offset: 0002a00f OS Version: 6.1.7600.2.0.0.768.3 Locale ID: 1033 Additional Information 1: 0a9e Additional Information 2: 0a9e372d3b4ad19135b953a78882e789 Additional Information 3: 0a9e Additional Information 4: 0a9e372d3b4ad19135b953a78882e789Read our privacy statement online: http://go.microsoft.com/fwlink/?linkid=104288&clcid=0x0409If the online privacy statement is not available, please read our privacy statement offline: C:\Windows\system32\en-US\erofflps.txt
January 16, 201412 yr Brian, thanks for bringing this bug to my attention. SID vectoring was completely broken! Here's the fix (I pulled the old version Q). http://www.mediafire.com/download/1jb3oqpomjdddkn/AISIDSTAR12Q.zip I became so focused on the final approach code that I didn't notice a final approach tweak broke the SID vectoring. It should be working now, but everyone should replace the old version Q with the one in the above link (if you want working SIDs). While I was looking at the SID code again, I revised it for better SID selection based on destination airport. Andqui please give this new version a try and see if it helps with the problems you mentioned in your email. Russian, try this version and see if it helps. I accidentally disabled a lot of code and that could have caused a crash. Regarding AI slowdowns on final (caused by traffic conflicts), add this parameter to AISIDSTAR: AIfinalapproachslowdownrate = x (note: default is 1.5 - a lower number means a smaller speed reduction, but stay at 1 or above). -Roland
January 16, 201412 yr Will give it a go and report back. Been using PHX at night as a tester. Very open approach path and easy to see the lights of the aircraft to assess how it's working.
January 16, 201412 yr SIDs are being assigned again :Applause: I am still seeing the same SID assigned to every departure though. Using ATL as my example, I have set up a SID for the north side and one for the south side for both west ops and east ops. The reason being, since we cannot time our departures, I want north dep vs south dep to turn away from each other at all times. For this example assume west ops. I have CADIT8 for 26L and JCKTS7 for 27R when west ops are in effect. For each of these SIDs, I have set an initial INTC waypoint 3 miles west of the field at the exact same longitude, but along the extended centerlines of 26L and 27R respectively. it is my understanding that since the longitude of these initial waypoints are identical, the waypoint directly in front of the departing AI, along the centerline of it's dep runway, should always be closest. At least the difference in longitude should be FAR smaller than the difference in latitude between 26L/27R. I am seeing all deps assigned CADIT8, no matter how I adjust the waypoints. I thought maybe it's exit was simply more attractive for the first departures that were headed to destinations north, but after an hour I've seen not a single AI assigned JCKTS7, even those heading south and west. Any ideas ? Regards,Brian Doney
January 16, 201412 yr AISIDSTAR still crashing. Here's the error. I've also included the link to the folder and files I am using. Question: if I only have files active for runways that aren't in use, will that cause a crash? https://www.dropbox.com/s/lipdd6t0c0e9idj/KDCA.zip Problem signature: Problem Event Name: APPCRASH Application Name: AISIDSTAR.exe Application Version: 0.0.0.0 Application Timestamp: 52d79c85 Fault Module Name: AISIDSTAR.exe Fault Module Version: 0.0.0.0 Fault Module Timestamp: 52d79c85 Exception Code: c0000005 Exception Offset: 0002a00f OS Version: 6.1.7600.2.0.0.768.3 Locale ID: 1033 Additional Information 1: 0a9e Additional Information 2: 0a9e372d3b4ad19135b953a78882e789 Additional Information 3: 0a9e Additional Information 4: 0a9e372d3b4ad19135b953a78882e789Read our privacy statement online: http://go.microsoft.com/fwlink/?linkid=104288&clcid=0x0409If the online privacy statement is not available, please read our privacy statement offline: C:\Windows\system32\en-US\erofflps.txt
January 18, 201412 yr I am having problems at FSDT LAX like AI aircraft landing at the same time and one behind the other closely. I also having problems that the AI planes like TFS, FAIB ect have their gears not retracting when taking off. Mike
January 18, 201412 yr Hi everyone, I must say, I sign up to AVISM forum almost specifically for this post. WOW! What a great utility we have now I found out about this program a few fays ago and have been reading through the entire topic before posting this. Only one remark I dare say here, more of my opinion actually, about this in the new features you exposed for version Q: Rev. 1.2Q (beta) can be found here: http://www.mediafire.com/download/esad5t1ydc4dwbe/AISIDSTAR12Q.zip [...]The AI rollout distance after landing has also been drastically reduced, aiding in more prompt evacuation of the runway thus increasing landing rates. [...] Thanks! -Roland When first reading this I first was really excited, because I thought you meant the AI landing roll would actually be lengthened. From what I have been reading on forums for years now, it seems that the general consensus would be that AI have the shortest landing roll so they can vacate the runway and avoid too many go-arounds for the following landing aircrafts. However, with the growth of programs like AI Smooth and AI Separation (and now yours!), the number of go arounds at busy airports was greatly reduced. it seems to me that AI users are now looking for realistic vacating of the runways. Rather than having the plane vacating via the first exists (more often than not one in a reverse directions, or not the one intended in real life) I prefer it to continue rolling to the appropriate rapid exit taxiway (which in the ned is not only more realistic but also better for the rest of the AI flow on the ground). You can see on many forums that people seek to tweak their AI FDEs to achieve this purpose (basically lengthening the landing roll-out, increase AI taxi speed on runway, sometimes ajusting the AI approach speed). It appears to me that this new feature of AISIDSTAR would work against this tweaks. Am I right to assume that? Therefore, if this feature could be made optional/activable like some other options in your great program, I guess that would serve many users too. I am even thinking that you could maybe implement the actual complete opposite option and allow the program to extend the AI landing roll after landing. This is my remark, although I have to admit that I haven't had the time to fully test your program, so all my apologies if this is not relevant. Just an idea however. On a different note, if I may offer a suggestion for your next round of new implementations: Could it be possible that, if go arounds occur, the AI would follow the Missed Approach procedure rather than a SID? Anyway congratulations on this great utility and thank you for bringing it to our FS, it's been much needed and appreciated. Best Regards Bib974
January 18, 201412 yr Guys, thanks for the great feedback, you helped me find several bugs. I appreciate the help. Here's the fix (I pulled the old Q version): http://www.mediafire.com/download/s6t9zhpdr2w5s1v/AISIDSTAR12Q.zip >> I am still seeing the same SID assigned to every departure though. Found the problem, this should definately be fixed now (or at least a lot better)......... >> AISIDSTAR still crashing. Found the problem. Thanks for sending me the files, which helped. >> I am having problems at FSDT LAX like AI aircraft landing at the same time and one behind the other closely. Yep, default values are set aggressively to ensure a high landing rate (low landing rates have been my number one complaint). To back things off a bit, try these values (from the documentation): AISeparationDistance = 5.0 AIHoldingAllowed = 4 AIDecreasingTimedRelease = 0 AIForcedReleasePriority = 0 AITargetLandingRate = 0.25 >> I must say, I sign up to AVISM forum almost specifically for this post. WOW! What a great utility we have now Thanks for the kind words, I appreciate it. >> When first reading this I first was really excited, because I thought you meant the AI landing roll would actually be lengthened. Yes, at default I'm trying to evacuate the AI from the runway as quickly as possible. However, to increase the AI landing roll, try these values (from the documentation). AIRollOutTime = 30 AIRollOutSlowDownRate1 = 1.025 AIRollOutSlowDownRate2 = 1.1 AIRecordRollOutSpeed = 25 If you start seeing the AI behave erractically during the landing roll, then back-off the numbers. If you're making custom final approach files, make sure to make the last roll out waypoint at the end of the runway to give the AI plenty of time to slow down. For auto-gen finals, the AIRollOutTime and AIRecordRollOutSpeed accomplish the same thing when recording the end waypoint. AIRollOutSlowDownRate1 and AIRollOutSlowDownRate2 apply to both custom and auto-gen and specify the de-accelleration in two steps. Lower numbers mean lower de-accelleration, but keep the number above 1. >> Could it be possible that, if go arounds occur, the AI would follow the Missed Approach procedure rather than a SID? Yes, now that I have control over the final approaches, I'll look into that!
January 18, 201412 yr Dear Roland, Wonderful!! Please give me some time to digest all that and I will give you more feedback on the program and the tweaks you offered. Again, many thanks!
January 18, 201412 yr Roland, Thank you very much for your long description. I will try version Q :-) ... And I will report ... Greets Klaus
January 19, 201412 yr I also having problems that the AI planes like TFS, FAIB ect have their gears not retracting when taking off. Mike Mike, I looked into this issue and I noticed the same. The problem is the height at which departing AI are put under control of AISIDSTAR is too low at default (300 ft AGL). The next version I'll bump this up to 1000 ft AGL, which solves the problem. In the meantime, you can add this to your .ini file: AIClimbOutHeight = 1000 -Roland
January 19, 201412 yr ^ on that I would actually suggest 900ft as that will let AISIDSTAR hook in just before the default ATC gives it's first vector...and should let you take control with wings level etc. --------------- SID assignment is working very well now, as they say a picture is worth 1000 words: Other idle thoughts: I'm not to sure about the rollout control. It's working well enough but I'd be curious to see a version that released control just after, or even just before, touchdown. Speed transitions are quite abrupt. Not an operational issue just a cosmetic one, and not really important at this stage, just an observation. When you command 410KTS, that isn't KIAS is it ? If so that's really a bit high. 170/250/300/250/170 would really be just fine. ------- I will work on east ops/sids a bit and then switch over to arrivals. SIDs are looking pretty good so far though, really. Regards,Brian Doney
Create an account or sign in to comment