I don't/didn't have any rolling cache and still have the AP issue

30 minutes ago, Noooch said:

I don't/didn't have any rolling cache and still have the AP issue

Do you have a manual cache? If so, delete that one and see if that brings you any improvement. (Check in Options/General/Data by the top of my head).

42 minutes ago, hvw said:

Do you have a manual cache? If so, delete that one and see if that brings you any improvement. (Check in Options/General/Data by the top of my head).

No I don't have any cache of any sort

this is a random issue like Iceman said.

Edited November 28, 20205 yr by Noooch

The a/p cannot magically change just because you deleted a cached folder. But just out of curiosity I did try that. No difference whatsoever. This is an a/p global setting issue. Please let me try to explain exactly what is happening:

Asobo has severely curtailed the heading control so any turn that begins with a healthy bank in the direction of the desired heading (so far so good) then starts rolling level WAY, WAY too early. So a healthy roll towards the target heading now becomes a battle to capture the heading with a TINY amount of bank. Consequently the last 10-20 degrees of heading change becomes so slow that let alone a standard rate turn, even a turn that takes twice as long is difficult to achieve. The faster your aircraft, the more time it takes to capture the heading. But you could say that is not so bad if you have the patience to wait.

BUT....now you need to capture an ILS, which relies not just on its own parameters but is also reliant upon the roll and heading modes in order to swiftly roll left or right if it gets away from the localiser. The closer it is to the localiser the slower it adjusts. You can test this very easily. If your capture starts drifting to the right or left of the runway, you'll see that once it reaches 30 degrees off, even if it almost at the threshold, it will suddenly start banking again because it is outside the heavily curtailed roll within 10-20 degrees. But it will not roll the aircraft enough earlier on, so you will miss the localiser. So now it is too late.

The other issue is that if you are within 20 or so degrees of the localiser it will bank far too late to get in line, even if you are several miles away. It will over-run to the other side, then limply try to get back through yaw. If you are lucky this might just about work but rarely, When Asobo did this "fix" to calm down the heading, they FORGOT that the very same heading and roll controls ALSO control the bank required to keep on a localiser.

The solution is not to emasculate the heading control, but to get rid of all the unwanted oscillations and bad a/p settings in each individual aircraft. They are not caused by the heading parameters. They are caused by something else.

There was not much wrong with the a/p that couldn't be fixed by tuning each individual aircraft. The "crazy" behaviour was NEVER caused by the heading control. Curtailing the heading control by reducing the bank angle way too early has implications not just for heading control itself but nearly every single a/p command apart from climb and descent.

It means you cannot properly capture any flight plan that has approaches or procedures that require turns over a short distance. It cannot capture ILS approaches now, as explained, and it cannot perform any more than very wide vectors under ATC. If ATC tells you to left 30 degrees you will wait a long time, because that is exactly the heading change that is the slowest with this updated a/p tweak by Asobo.

They just did not think this through.

Edited November 28, 20205 yr by robert young

4 minutes ago, robert young said:

The a/p cannot magically change just because you deleted a cached folder. But just out of curiosity I did try that. No difference whatsoever. This is an a/p global setting issue. Please let me try to explain exactly what is happening:

Asobo has severely curtailed the heading control so any turn that begins with a healthy bank in the direction of the desired heading (so far so good) then starts rolling level WAY, WAY too early. So a healthy roll towards the target heading now becomes a battle to capture the heading with a TINY amount of bank. Consequently the last 10-20 degrees of heading change becomes so slow that let alone a standard rate turn, even a turn that takes twice as long is difficult to achieve. The faster your aircraft, the more time it takes to capture the heading. But you could say that is not so bad if you have the patience to wait.

BUT....now you need to capture an ILS, which relies not just on its own parameters but is also reliant upon the roll and heading modes in order to swiftly roll left or right if it gets away from the localiser. The closer it is to the localiser the slower it adjusts. You can test this very easily. If your capture starts drifting to the right or left of the runway, you'll see that once it reaches 30 degrees off, even if it almost at the threshold, it will suddenly start banking again because it is outside the heavily curtailed roll within 10-20 degrees. So now it is too late.

The other issue is that if you are within 20 or so degrees of the localiser it will bank far too late to get in line, It will over-run to the other side, then limply try to get back through yaw. When Asobo did this "fix" to calm down the heading, they FORGOT that the very same heading and roll controls ALSO control the bank required to keep on a localiser.

The solution is not to emasculate the heading control, but to get rid of all the unwanted oscillations and bad a/p settings in each individual aircraft. They are not caused by the heading parameters. They are caused by something else.

Robert, I am sorry but ever after deleting my rolling cache and have FS2020 built a new one, I don't have any of the issues you mention anymore. Perhaps I am just lucky. My ILS approaches are right on the dot. I can't explain why it is working here, I won't even try to do so :). But I do agree, that Asobo should have done a far better job on this part of their "update". Perhaps they should consider to bring some betatesters into their team that have knowledge of how a plane should fly in general and how it respond on A/P in particular :).. It is a shame they released this "update" without having it thoroughly tested. Or perhaps they just thought that they did a phenomenal job in their opinion, I don't know.

With regard to heading changes, yes, what they call "smoothing" a turn turned out to be starting a turn way too soon, but it does the job correctly, at least for me. On a SID out of LEMD the plane  had to change from a heading 140 to hdg 65 (ordered by ATC) and the A/P did it just perfectly, albeit that the turn start too soon.

I didn't experience the incorrect banking you are mentioning when intercepting the localiser. Again, I might just have been lucky.

Let's just hope that Asobo will come up with a fix that works for everybody and every plane soon, very soon, as it is now it is a royal PITA for many of us. 

Anyway, I wish you and everybody else blue skies and enjoyable flights.

Hans

7 minutes ago, hvw said:

Robert, I am sorry but ever after deleting my rolling cache and have FS2020 built a new one, I don't have any of the issues you mention anymore. Perhaps I am just lucky. My ILS approaches are right on the dot. I can't explain why it is working here, I won't even try to do so :). But I do agree, that Asobo should have done a far better job on this part of their "update".

Hi Hans, 

I do believe you. Maybe you have something going on that others do not. I'm happy to learn about it. Can you please say which aircraft you have no problems with? Do you observe that a heading change starts with a normal bank angle then starts leveling off miles before you reach the heading and then turns very slowly, or are you seeing a healthy bank/roll right up to immediately before the heading is reached?

When you say your ILS captures are on the dot, I was refering not to the glideslope which is untouched and is fine. It is the localiser. Are you approaching from dead straight and far away? Have you tried capturing the localiser from an angle of say 30 degrees and 3-4 miles away? How does that go?

If you have other settings that might be relevant it could be useful for others to know. Is it ONLY your rolling cache or did you also delete or change something else? Thank you for any help you can offer.

Edited November 28, 20205 yr by robert young

36 minutes ago, Noooch said:

No I don't have any cache of any sort

this is a random issue like Iceman said.

Sorry to hear you're still suffering of this issue. But I didn't say it was THE fix, but a possible one. It worked for me so far and perhaps for a few others. Asobo should get its us a working update for everybody soonest.

Robert, I am only using the FBW A320NX (latest dev mode). My heading changes (A/P) are as expected except for the fact, that they start a bit too early, but other than that, they are fine. No levelling off before reaching the new heading etc as you mentioned. This morning I did a LEMD to EDDK flight. I did a DEPOK STAR and a rwy 14L ILS approach into EDDK without any glitch. It worked fine. This may sound strange to you,  but it's true. Go to YouTube and look for  a video by Simpilot320 or 320SImpilot. He made a nice video while flying the A320 after the latest update flawlessly. Actually his video made me search for a fix to the A/P problem and hence after I found this rolling cache deletion. 

Hans

1 minute ago, hvw said:

Robert, I am only using the FBW A320NX (latest dev mode). My heading changes (A/P) are as expected except for the fact, that they start a bit too early, but other than that, they are fine. No levelling off before reaching the new heading etc as you mentioned. This morning I did a LEMD to EDDK flight. I did a DEPOK STAR and a rwy 14L ILS approach into EDDK without any glitch. It worked fine. This may sound strange to you,  but it's true. Go to YouTube and look for  a video by Simpilot320 or 320SImpilot. He made a nice video while flying the A320 after the latest update flawlessly. Actually his video made me search for a fix to the A/P problem and hence after I found this rolling cache deletion. 

Hans

Ah, that explains it. Thanks. I think the FBW A320 probably has a custom autopilot. That might also explain the rolling cache deletion having an effect. When you have a moment, do try one of the default aircraft and you'll immediately see the problem.

My A / P problem is when I am flying with the A320 during the approach and close to capturing the ILS the A / P leaves the route and goes to the last waypoint that is the selected runway.
Is anyone having this problem too?

1 minute ago, Ricardo Andrada said:

My A / P problem is when I am flying with the A320 during the approach and close to capturing the ILS the A / P leaves the route and goes to the last waypoint that is the selected runway.
Is anyone having this problem too?

Hi Robert,

After deleting the rolling cache etc. I made three flights so far. Later today I will try to make another short one as I doubt that my ;luck will persist. If time permits I will try to make a few more than one, following your logic I should run into the malfunctioning A/P as well. 

I hope that Asobo got their ears on and are, as we speak, working on a solution. Will keep you (and others that are reading this thread) updated on my experiences.

Hans

This worked for me, basically I use the older version of the A32NX, so maybe the issue is now easier to find!

 

Edited November 28, 20205 yr by Noooch

1 hour ago, robert young said:

Thanks. I think the FBW A320 probably has a custom autopilot.

The devs have said they can’t alter the auto pilot functions, so I doubt it.

So maybe it isn’t the auto pilot that’s broken per-say, but something else on the aircraft is interacting with it.

Edited November 28, 20205 yr by Tuskin38

1 hour ago, robert young said:

Ah, that explains it. Thanks. I think the FBW A320 probably has a custom autopilot. That might also explain the rolling cache deletion having an effect. When you have a moment, do try one of the default aircraft and you'll immediately see the problem.

Robert,

Now I ran into this problem again, too. Upon entering the ILS approach the aircraft made a hard right turn and I had to disengage the A/P to get back to the first entry of the ILS approach again. (Again I used the FBW A320NEO). For the time being I will leave FS2020 alone and return to P3D to make some successful flights with the 737. I haven't got a clue why it worked great a few times and now once of a sudden leaves me out in the cold again. 

Did you try the fix explained in the video above? I just made a OKBK - OTHH with it and it was perfect