|
Post by mjl1966 on Dec 26, 2021 15:14:28 GMT -5
The one scoring parm I have a beef with is the landing score. If I grease it, pax give me five points. Even if I'm 1000 ft. beyond TDZ and have to stomp on the brakes. It is known among the ranks of airline pilots that a firm ("positive") landing is safest, for several reasons - among them the better chance of hitting the TDZ without leaving runway behind you.
I'm wondering if there is a way to grade the landing based on the touchdown location relative to the TDZ vice the descent rate exclusively. 5 pts for anything between the threshold and TDZ. Bonus points for lower descent rate. Take away points for every 100 ft. beyond the TDZ.
That would be more in line with real world standards. TDZ is king.
|
|
|
Post by Travis on Dec 27, 2021 21:19:52 GMT -5
The only reason why what you describe is not currently an option is simply that I am not confident enough in our "runway detection" method to entrust it to something which could have a detriment to any Captain's career. This is an area in which I need to be exactly confident. And I have my other ideas to make "airport runways" an important element on their own, so this is something that I want. The reason "runway detection" hasn't improved in the past few years is nothing more than lots of bad timing. There always seems to be another "Very Important Item" that jumps the Project Queue - recently being, "a new simulator". After FSCaptain 1.8.3, I had to make FSCaptain more friendly with MSFS. But even with that as a mandate, enough "bugs" in 1.8.3 showed up after release that I had to stop work and shift over to "1.8.3 Service Pack 1". That's where we're at right now. After SP1 releases, I will block off a week or so to work on making the "runway detection" method work better. That schedule might work out nicely because I've chosen to not do anything to tax my current PC until I get a new one in late January (if I'm lucky). I've backups aplenty, but I don't want to rebuild a PC just before building a new PC. Best,
|
|
|
Post by mjl1966 on Dec 28, 2021 1:01:38 GMT -5
I would imagine that it is difficult to define with a variable that is easily read from the environment, where that TDZ actually is. You know where the airplane is, and you know when the wheels touch down, but how does one know where the TDZ is?
The fact that you've already thought of this same parameter is intriguing. Wizardry is afoot!
|
|
|
Post by Travis on Jan 2, 2022 22:16:59 GMT -5
Over the past week or so, I've had the lightest of head colds, so I've been not pushing myself in doing anything. But this issue (improper runway detection) keeps nagging me for a fix. I'm currently working on a different algorithm to support this issue, and I've had good results in a few airports where I know detection has previously failed for me! I'm not finished here - I'm currently having a problem telling my Right Side of the Runway from my Left Side - but I think I'll have something ready by the end of this new week. When I'm happy with that, I'll loop back to look at other current Bugs, and prepare to release 1.8.3 SP1 when all fixes are ready. ---------- However to make things work for "TDZ Scoring" we need to establish a definition of What a TDZ is for FSCaptain. My current model comes from an FAA AIM (Aeronautical Information Manual) which I found online. It defines a TDZ as the first 3000 feet of runway beginning with the runway's threshold. If the runway is less than 6000 feet in length, the TDZ would then end at the midfield location. (So for a 5000 foot runway, the TDZ would be measured starting at the threshold and extending only 2500 feet.)
Are there any other models that we should support? Bear in mind that the MakeRunways data for a runway consists of only two points - the Center-line Threshold Point, and the Center-line Midfield Point. I can calculate the Left Side Edge and the Right Side Edge (but see above) so I have what I need to support the FAA TDZ definition. If there are any other TDZ definitions, please let me know!!
Best,
|
|
|
Post by Travis on Jan 3, 2022 19:47:09 GMT -5
I renamed this topic to make the focus be more on the concept of "TDZ"s.
I have already received one emailed description of what a TDZ means, and it fits the model I described earlier.
All data / descriptions / questions are welcome - and there's no rush. I've now shifted to outstanding bugs and will circle back here in several weeks to pick up future work.
Best,
|
|
|
Post by DirkDP on Jan 8, 2022 4:32:58 GMT -5
|
|
|
Post by Travis on Jan 26, 2022 16:47:38 GMT -5
Thanks for the pointer, Dirk. I knew there was previous discussions, but you saved me from hunting them down. ----- I have done some test coding and I'm fairly happy with TDZ detection, but I do have a few corner cases to test as well before I'm happy enough to make a build to open up testing for more. While we're in SP1 - Alpha 1 testing, the TDZ results will be only written to the FCDU log, and possibly displayed on the FCDU during your "rollout phase" which doesn't last more than 30 seconds or so. Once we're all in agreement that TDZs are being called correctly, I can then progress to allocating penalty points for missing a TDZ. I'm of the opinion that you shouldn't receive any Bonus points for a TDZ landing, since that's really part of your job. LOL But when the time is right, I will be open to arguments and suggestions from any quarter. In a day or so, I may post some screenshots of my upcoming testing at CYTH. (Peter found it for my specification: "I need at least two runways with bisecting TDZs, in a location that's away from any complex scenery.") Btw, because I don't want to crash my PC, my testing is limited to taxiing around barren airports using my old testing warhorse - the FSX Trike! Best,
|
|
|
Post by DirkDP on Jan 29, 2022 13:24:03 GMT -5
Hi,
Make sure to test RWYs with displaced threshold... I think my VA's ACARS had problems with not correctly detecting them in FS9/FSX.
Regards, DDP.
|
|
|
Post by Travis on Jan 30, 2022 16:19:57 GMT -5
Hi Dirk,
Using default simulator scenery (FSX or P3D preferred), can you give me an airport & runway that you would like me to test?
Preferably somewhere in the boondocks for me right now, but I can give a Metro Hub Airport a quick try.
Best,
|
|
|
Post by Travis on Feb 1, 2022 14:15:48 GMT -5
For my first tests, I see that at my P3D4 stock KBFL, 12L/30R has a length of 10859ft, but 30R has an offset of 3428ft. By my reckoning, that leaves 7431ft free for landing. I'll test elsewhere if someone can find a tricky situation! Best,
|
|
|
Post by Travis on Feb 4, 2022 14:16:00 GMT -5
Just to update that testing is still going well. I can set up on KBFL 30R and our Runway Detector sees that we're on 30R... but if I then press the secret "Pretend I just landed here" button (PIJLH), the FCDU will not place me in a TDZ. I can then taxi a few thousand feet to the runway ID markers and press PIJLH again and get an acknowledgement that I'm in the TDZ! A little birdie told me that the P3D default KVCV has a large displaced threshold. In reality there isn't one, but this will work for my testing. Advice on any other default P3D airports with potential "corner cases" are still welcome!! I'll also accept advice for any P3D freeware airport, as well as "ORBX region" airports. I think I have all of them.... ----- I can only manage about 45 minutes of very easy in-sim testing before my system temps begin to rise and I'm forced to shut down the simulator. Still though I should be ready to make a test build for Peter and Dutch by this weekend. Best,
|
|
|
Post by Travis on Feb 19, 2022 18:23:13 GMT -5
My simple testing has gone well, but I'm holding off on an Alpha SP1 release until I can actually make an honest-to-goodness flight. This is how the FCDU would look after landing within a TDZ - nearly 1900ft short of the TDZ's end. ----- My new PC should be ready next week. It will take a while to get it ready for use, so I'm hoping to have something ready by early to mid March. (Famous last words, I know....) Best,
|
|
|
Post by Travis on Apr 27, 2022 18:07:18 GMT -5
Captains, Release of 1.8.3 Service Pack 1 - Alpha 1 is very close. One of the biggest changes in this release is in how our "Runway Detector" functions, and from that change there seems to finally be a way to judge landings based on "hitting the TDZ". The Alpha and Beta releases of Service Pack 1 will include a way to test the runway (and TDZ) detector. Here is the current relevant notes from the README file so all TDZ-focused Captains can be ready to test our calculations.
We hope this will bring increased realism, and as noted in the notes, future FSCaptain features....
Best,
|
|
|
Post by alaskanflyboy on May 2, 2022 19:41:33 GMT -5
Just to update that testing is still going well. I can set up on KBFL 30R and our Runway Detector sees that we're on 30R... but if I then press the secret "Pretend I just landed here" button (PIJLH), the FCDU will not place me in a TDZ. I can then taxi a few thousand feet to the runway ID markers and press PIJLH again and get an acknowledgement that I'm in the TDZ! A little birdie told me that the P3D default KVCV has a large displaced threshold. In reality there isn't one, but this will work for my testing. Advice on any other default P3D airports with potential "corner cases" are still welcome!! I'll also accept advice for any P3D freeware airport, as well as "ORBX region" airports. I think I have all of them.... ----- I can only manage about 45 minutes of very easy in-sim testing before my system temps begin to rise and I'm forced to shut down the simulator. Still though I should be ready to make a test build for Peter and Dutch by this weekend. Best, What's the culprit on the overheat? Anything you can run a liquid cooler on? For the TDZ, as I recall the standard definition by the FAA is the first 3000 feet of a runway that is allowed for touchdown. When calling out the elevation for the TDZ on an approach chart, it's the highest point in that first 3000 feet. I'm sure there are exceptions, but I'd start there.
|
|
|
Post by Travis on May 3, 2022 11:28:24 GMT -5
John, I suspect it was the power supply, but I needed to update my mobo so I could run Win 11, so I chose the route of least resistance - a new rig! ("Least resistance" = "Mo$t co$t"!) ======
In my implementation of TDZ Detection, I made certain to allow for different rulesets. The current rule is "The first 3000 feet of a runway that is at least 6000 feet in length... otherwise, half of the runway." I'm eager to get feedback from you and from Dirk and from the OP! ( Download the 1.8.3 Service Pack now - instructions are here.) Currently, TDZs are seen everywhere and all the time - on concrete or grass strips, and in water and snow. Both your A350 and your Trike will have them calculated. We can later apply any restrictions that the group feels are logical. Best,
|
|