Any idea on FM bug fixes?

[S. Low]

I see in patch notes (good stuff in there btw!) you mention some items you hope to get into next update, like TSE, which is very awesome, but I’m wondering if there’s any idea on corrections to FM bugs such as roll on negative g, excessive crabbing, apparent instability perhaps related to scas issues (I don’t understand scas ..), auto rotation etc?

[admiki]

3 hours ago, S. Low said:

I see in patch notes (good stuff in there btw!) you mention some items you hope to get into next update, like TSE, which is very awesome, but I’m wondering if there’s any idea on corrections to FM bugs such as roll on negative g, excessive crabbing, apparent instability perhaps related to scas issues (I don’t understand scas ..), auto rotation etc?

Hind still has some FM bugs that are here from introduction, so I wouldn't hold your breath.

OTOH, Apache has all the hype now, so, who knows?

[dedlike.]

im quite dissapointed how the SCAS behaves aswell , especaly when trying to do smaler adjustments at low speeds its fighting you alot, good example is trying to yaw while hovering

the only way you get it to yaw is by oversaturating the SCAS but once you let go of the pedals it trys to overcorrect and it just makes the heli wobble all the time

same when going faster than 120-130 knots , i feel like they gotta adjust the PIDs on the whole thing, but they should get the flightmodel itself fixed first ofc....

i just wish it was higher priority, funny enought, i tought the apache flew better at release, now its fighting me more than helping me

[Photon]

Agree with every point made. I really wish it would take top priority. So tired of rolling around and wobbling 

[Floyd1212]

This has been mentioned elsewhere, but worth mentioning again here, and maybe a SME can chime in to verify:

The correct way to fly the helo is to press and hold the FTR button for the entire time you are making adjustments with the cyclic or pedals.  While you have the FTR button held, the SCAS is disabled and will not fight you.  When you are done moving the controls, you release the FTR button and the new position of the controls are now set as the "trimmed" position, and the SCAS kicks back in to do its job trying to keep the roll/yaw/pitch as close as it can to where you left it trimmed.  If you make movements to the controls without holding the FCR button down, yes, it will fight you.  And of course the SCAS is going to get over-saturated while you are trying to make changes to the trimmed attitude of the aircraft without disabling the system that is designed to maintain that trim for you.  The only way it knows not to fight you is by you holding the FCR button down.

That being said, I think most of us (myself included) simply tap the FCR button to set the trim occasionally, and have grown accustomed to how the helo behaves while flying with SCAS enabled and fighting us.  It does make for some weird behavior when adjusting yaw with the pedals while in a hover, for example, but on the other hand, it does a pretty good job smoothing out oscillations in the roll axis during fwd flight. 

That doesn't explain some of the FM issues that still need to be addressed, like the nose diving down and to the left at speed, but I'm not sure we can complain about the "SCAS fighting us all the time" when we aren't flying it the way it was intended.

It's like setting the Cruise Control on your car and then complaining that the car is fighting you when you lift off the gas and the car works to maintain its speed for you.

[Raptor9]

Ok, I'll respond to this thread given the nature of the topic, but I want to stress two very important things before I do. The first is the most important, which is that I am not on the dev team, nor do I speak for them, I'm only speaking as someone with real-world experience with this SCAS behavior. The second is the usual reminder to those that may not be familiar with the Early Access process of complex DCS modules, that this module is still in a relative infancy compared to other modules that have been in Early Access for substantially longer periods of time and are still getting fine tuning and tweaks done to their flight controls. Modeling helicopter flight controls in a flight sim is hard enough, now add in something as complex as the F-16's FLCS system that also interacts with the flight controls on top of that. Anyway, with disclaimers out of the way...

53 minutes ago, Floyd1212 said:

The correct way to fly the helo is to press and hold the FTR button for the entire time you are making adjustments with the cyclic or pedals.

I really don't like to use absolutes when talking about aviation practices any more than I do when discussing military theoreticals and "what ifs". I am personally of the preference where I press the FTR until I finish moving the controls, and then I let go of it. This practice proved to be particularly helpful when learning to play the DCS Ka-50, since the Shark's autopilot without the flight director turned on will fight you, unless you press the trimmer button when maneuvering. However, sometimes you can't press the Force Trim Release while maneuvering, such as if you need to use your thumb to make a radio call, or action/de-action a weapon, change symbology modes, etc.

53 minutes ago, Floyd1212 said:

While you have the FTR button held, the SCAS is disabled and will not fight you.  When you are done moving the controls, you release the FTR button and the new position of the controls are now set as the "trimmed" position, and the SCAS kicks back in to do its job trying to keep the roll/yaw/pitch as close as it can to where you left it trimmed.

Not really, and this is where the discussion diverges between real-world behavior and current DCS AH-64D behavior, which is as I said, still WIP (this has been mentioned multiple times on the forums by actual members of ED by the way, this isn't coming from me). Further, I haven't flown the new updated flight model yet, so I can't speak to how it is behaving right now.

When the FTR button is held down, the real-life SCAS isn't disabled, but it's logic is altered somewhat; and how it is altered is dependent on airspeed, attitude, and whether or not there is an attitude hold sub-mode engaged at the time or not.

53 minutes ago, Floyd1212 said:

If you make movements to the controls without holding the FCR button down, yes, it will fight you.  And of course the SCAS is going to get over-saturated while you are trying to make changes to the trimmed attitude of the aircraft without disabling the system that is designed to maintain that trim for you.  The only way it knows not to fight you is by you holding the FCR button down.

Again, speaking to how it behaves in real-life: no it does not, with a few exceptions. With attitude hold engaged, once the controls are moved beyond a certain threshold without FTR being pressed, referred to as "breakout values", attitude hold will stop attempting to hold the attitude and will let the pilot maneuver the aircraft as desired, until the controls are back within the original control positions and the pitch/roll and/or yaw rates are back within certain limits, depending on what axis/axes went through the breakout values. Once this criteria has been met, the attitude hold will re-engage and set the new attitude values as the reference it will hold.

In the same scenario, if the FTR is held down, when the FTR is no longer pressed and it meets the rates criteria, then the new reference will be established. Holding FTR suspends any engaged hold modes within the pitch/roll/yaw axes (ie, Attitude/Velocity/Position sub-modes or Heading hold), which means the aircraft will not fight you, but because of the breakout values, it wouldn't anyway if the controls are moved deliberately. So if you bump the controls or make very very small movements to them while a hold mode is engaged, so small that it won't violate the breakout values, then yes it will attempt to null out the movement, because it assumes it was an inadvertent control input. The breakout values themselves also change based on airspeed and what hold sub-modes are engaged...again, very complex.

The best example of when it will actively fight you is in Position hold. You've told the aircraft to maintain a hover here, and then you move the cyclic without pressing the FTR, and the aircraft will actively fight you until it becomes saturated. So this is a prime example of when you need to press and hold the FTR until you get established and stable at your new hover position, and then let go of the FTR button.
__________________________________________________

So now that the sea of words has been typed out, here's the TL;DR, the meat and potatoes, the bottom line...The SCAS system is a lot more complex than most people realize, and I do not envy the dev team one bit in trying to get this all laid out and accurate. I imagine this will take time, just like the flight controls on the F-18 or F-16.

Here's a recent photo from the office of the dev members that are in charge of coding the SCAS system in DCS AH-64D:

(it's a joke, but a joke to make a point, and I wouldn't be surprised if it wasn't far off)

Edited June 8, 2022 by Raptor9

[Floyd1212]

Thank you for your time, and knowledge on the subject.

You always do a great job helping us understand what is going on under the hood.

[cow_art]

Thank you for your time and for explaining this so clearly.

So my takeway message is:  in reality there are at least two techniques (either press and hold FTR or move controls enough to exceed breakout values) both of which are currently not working correctly in DCS

... which is fine, it's early access after all. But for me that also means I can stop trying to "get good" flying the Apache until the FM is much closer to finished. Currently I feel it's an exercise in frustration and now I understand whatever I'll learn now, I'll have to unlearn when the SCAS is updated. Thanks again for the clear communication!

Edited June 9, 2022 by cow_art

[jubuttib]

Looking forward to getting the behavior a bit closer to what Raptor9 describes. It's not terribly bad for me right now, the main thing getting in my way is how doing even small adjustments to my anti-torque pedals in a hover currently leads to the yaw channel of the SCAS bouncing left and right, making the whole heli wobble. Didn't use to do that before the updates, or at least was much less of an issue.

[Swift.]

3 hours ago, cow_art said:

But for me that also means I can stop trying to "get good" flying the Apache until the FM is much closer to finished. Currently I feel it's an exercise in frustration and now I understand whatever I'll learn now, I'll have to unlearn when the SCAS is updated. Thanks again for the clear communication!

 

Don't be mistaken, it is possible to 'fly good' in DCS right now. 

I'll be honest, I'm not really seeing all the issues people describe with the rudder oscillating or having to 'fight' the SCAS. Admittedly I haven't flown in the patch from Wednesday, but this is from the last update:

The only oscillations I see are ones I induce myself, because I'm just a dumb fixed wing pilot that thought apache looked cool

I'd be interested to see some clips or tracks of the problems people are having!

[dedlike.]

when i made my coment earlyer it might have sound like im REALY complaining about it, im not, its just something i notice, trying to turn left/right in a hover just wont let you do it, you can hold down the trimm button but that usualy makes it worse for me when im hovering lol, its a good start, and the SCAS simply just is complicated, i do feel like some high speed ocelations could be fixed by adjusting PID values but there is most likely alot more going into that....

 

but just to put it down here, i still enjoy flying around and doing missions in the Apache, and i guess seeing that there is room to improve its gonna give me something to look forward to

[dedlike.]

allright, just got to test the recent update, it might not sayso in the patch notes but it does feel smoother to fly, maybe im going crazy who knows

still trying to figure out a way how to yaw in a hover without having to depress the trim, since that will esentialy put you out of the hovermode aswell

[jubuttib]

On 6/9/2022 at 1:57 PM, Swift. said:

Don't be mistaken, it is possible to 'fly good' in DCS right now. 

I'll be honest, I'm not really seeing all the issues people describe with the rudder oscillating or having to 'fight' the SCAS.

Was probably a bit unclear, my issues with the yaw channel started with the update before Wednesday, when the hover hold and attitude hold modes were introduced. You can still fly it well definitely, but I noticed multiple times when in hover/near hover that adding a bit of pedal caused the yaw channel to wobble side to side. If it was just the helicopter wobbling that'd be one thing, but the autopilot yaw channel readouts themselves were being really aggressive.

[Pixel of Life]

I don't have a track I can post right now (I have one but the file is too big), but I did a simple test where I came to a hover a few feet off the ground, engaged position hold mode, and then made a few very slow and gentle left and right turns. What I noticed is that the yaw autopilot seems to be making a sudden erroneous input in the direction I'm trying to turn which destabilizes the heli.

Luckily I also recorded a video of this:

Pay close attention to the green autopilot yaw channel indicator on the controls overlay when I start my first turn at 0:30. I'm very slowly and carefully applying left pedal and the autopilot keeps fighting me, and then at 0:44, you can just about see the autopilot make an unexplained yaw input to the left which causes the nose to snap to the left.

And then again, at 1:14 when I'm trying to make a right turn, the autopilot makes a sudden input to the right.

And again, at 1:42 when I'm trying to make another left turn, the autopilot makes a sudden input to the left again.

Can anyone else test this and verify? This is on latest Open Beta.

[dedlike.]

vor einer Stunde schrieb Pixel of Life:

I don't have a track I can post right now (I have one but the file is too big), but I did a simple test where I came to a hover a few feet off the ground, engaged position hold mode, and then made a few very slow and gentle left and right turns. What I noticed is that the yaw autopilot seems to be making a sudden erroneous input in the direction I'm trying to turn which destabilizes the heli.

Luckily I also recorded a video of this:

Pay close attention to the green autopilot yaw channel indicator on the controls overlay when I start my first turn at 0:30. I'm very slowly and carefully applying left pedal and the autopilot keeps fighting me, and then at 0:44, you can just about see the autopilot make an unexplained yaw input to the left which causes the nose to snap to the left.

And then again, at 1:14 when I'm trying to make a right turn, the autopilot makes a sudden input to the right.

And again, at 1:42 when I'm trying to make another left turn, the autopilot makes a sudden input to the left again.

Can anyone else test this and verify? This is on latest Open Beta.

thats exactly what i have been trying to talk about

[kgillers3]

5 hours ago, Pixel of Life said:

I don't have a track I can post right now (I have one but the file is too big), but I did a simple test where I came to a hover a few feet off the ground, engaged position hold mode, and then made a few very slow and gentle left and right turns. What I noticed is that the yaw autopilot seems to be making a sudden erroneous input in the direction I'm trying to turn which destabilizes the heli.

Luckily I also recorded a video of this:

Pay close attention to the green autopilot yaw channel indicator on the controls overlay when I start my first turn at 0:30. I'm very slowly and carefully applying left pedal and the autopilot keeps fighting me, and then at 0:44, you can just about see the autopilot make an unexplained yaw input to the left which causes the nose to snap to the left.

And then again, at 1:14 when I'm trying to make a right turn, the autopilot makes a sudden input to the right.

And again, at 1:42 when I'm trying to make another left turn, the autopilot makes a sudden input to the left again.

Can anyone else test this and verify? This is on latest Open Beta.

What I think you're experiencing is the heading hold was active at that point, which is going to attempt to keep the aircraft within a certain amount of degrees on heading of which it's established.  When you displace the pedals past a certain point the aircraft goes "The pilot wants to make an input" and disables / places on hold the heading hold. Which is why you see the jerk because it's no longer trying to maintain the heading.  It's important to understand that all of the hold modes are WIP and will undoubtably become smoother to use I think in the future.  But to answer your question on what you're experiencing.

What I would recommend is try a little counter pressure after initiating the turn and reaching your desired heading (push the opposite way a little bit) and release pressure off the pedals to see if it re-engages, or what's easier just retrim and allow your pedals to recenter, if you're using dampeners where your physical pedals remain in place, just release pressure and the heading hold should reactivate (after the counter pressure / retrimming)

TL/DR - By displacing the pedals you're disabling the heading hold which means the aircraft is expecting you to maintain the heading until it tries to reactivate, retrim once on your new desired heading should alleviate alot of this.  

Edited June 11, 2022 by kgillers3

[Pixel of Life]

1 hour ago, kgillers3 said:

What I think you're experiencing is the heading hold was active at that point, which is going to attempt to keep the aircraft within a certain amount of degrees on heading of which it's established.  When you displace the pedals past a certain point the aircraft goes "The pilot wants to make an input" and disables / places on hold the heading hold. Which is why you see the jerk because it's no longer trying to maintain the heading.  It's important to understand that all of the hold modes are WIP and will undoubtably become smoother to use I think in the future.  But to answer your question on what you're experiencing.

What I would recommend is try a little counter pressure after initiating the turn and reaching your desired heading (push the opposite way a little bit) and release pressure off the pedals to see if it re-engages, or what's easier just retrim and allow your pedals to recenter, if you're using dampeners where your physical pedals remain in place, just release pressure and the heading hold should reactivate (after the counter pressure / retrimming)

TL/DR - By displacing the pedals you're disabling the heading hold which means the aircraft is expecting you to maintain the heading until it tries to reactivate, retrim once on your new desired heading should alleviate alot of this.  

 

But if you look at the behavior of the heading hold system (like I said, pay very close attention to the green indicator just before the nose swings to the side), it doesn't simply disengage and stop trying to maintain the heading. It suddenly switches from fighting my inputs to amplifying them and actively tries to nudge the heli in the direction I'm trying to turn.

It's this behavior that throws me off and causes the terrible wobble, and it shouldn't be happening if, as you say, the heading hold disengages (i.e. turns off and stops doing anything, which is the only way you can interpret that word) when you give it enough pedal input. It's completely unexpected and there's no way to anticipate it.

[DummyCatz]

On 6/9/2022 at 7:26 AM, Raptor9 said:

...

When the FTR button is held down, the real-life SCAS isn't disabled, but it's logic is altered somewhat; and how it is altered is dependent on airspeed, attitude, and whether or not there is an attitude hold sub-mode engaged at the time or not.

...

Thanks for the insight. So it sounds like there's some kind of transition state in-between that would help alleviating/removing the jerkiness when the FTR button is held down, as in the current DCS implementation?

Another question is about the above mentioned Heading Hold. I remember you mentioned that the heading hold function is active all the time even if Hold Modes are not active. And when the pilot starts kicking the rudder above the dead zone, the heading hold momentarily disables to allow a heading change. So the moment when heading hold disables, does the SCAS also provide a smooth transition, or does it has the jerkiness like currently in DCS?

Edited June 11, 2022 by DummyCatz

[Kharrn]

afaik Heading Hold engages when you press "Force Trim Release" (UP).

 

vor 9 Stunden schrieb Pixel of Life:

I don't have a track I can post right now (I have one but the file is too big), but I did a simple test where I came to a hover a few feet off the ground, engaged position hold mode, and then made a few very slow and gentle left and right turns. What I noticed is that the yaw autopilot seems to be making a sudden erroneous input in the direction I'm trying to turn which destabilizes the heli.

[snipped]

Pay close attention to the green autopilot yaw channel indicator on the controls overlay when I start my first turn at 0:30. I'm very slowly and carefully applying left pedal and the autopilot keeps fighting me, and then at 0:44, you can just about see the autopilot make an unexplained yaw input to the left which causes the nose to snap to the left.

And then again, at 1:14 when I'm trying to make a right turn, the autopilot makes a sudden input to the right.

And again, at 1:42 when I'm trying to make another left turn, the autopilot makes a sudden input to the left again.

Can anyone else test this and verify? This is on latest Open Beta.

i will try to test later, but some observations from my humble opinion and my assumptions:

0:25 SCAS YAW is to the left because it wants to tell you "add anti torque pedal left to keep this attitude at current collective setting";

0:41 SCAS YAW and your pedal input match up (nearly no difference in heading, only 2-3° left);

0:46 i assume SCAS heading hold kicks in wanting to "steer" back right;

0:47/0:48 heading hold gets on "hold"/disabled (as kgillers3 mentioned), just afterwards you let slightly go of your left pedal input and i assume SCAS YAW is applying left pedal again to aply anti torque for the current colelctive setting;

0:58 SCAS YAW is again left (see 0:25).

 

I hope this might be of help!?

If i turn when having "ATT HOLD" enabled at 0 knots, i do minor pedal input and then "trim" (FTR UP) - and again - and again. It kinda works. Will  re-try later!

K

[Pixel of Life]

20 minutes ago, Kharrn said:

afaik Heading Hold engages when you press "Force Trim Release" (UP).

 

i will try to test later, but some observations from my humble opinion and my assumptions:

0:25 SCAS YAW is to the left because it wants to tell you "add anti torque pedal left to keep this attitude at current collective setting";

0:41 SCAS YAW and your pedal input match up (nearly no difference in heading, only 2-3° left);

0:46 i assume SCAS heading hold kicks in wanting to "steer" back right;

0:47/0:48 heading hold gets on "hold"/disabled (as kgillers3 mentioned), just afterwards you let slightly go of your left pedal input and i assume SCAS YAW is applying left pedal again to aply anti torque for the current colelctive setting;

0:58 SCAS YAW is again left (see 0:25).

 

I hope this might be of help!?

If i turn when having "ATT HOLD" enabled at 0 knots, i do minor pedal input and then "trim" (FTR UP) - and again - and again. It kinda works. Will  re-try later!

K

I don't think any of those timestamps are relevant. Again, look at the exact moment the heli starts yawing. Watch it in full screen and slow motion if you need to. Look at the control overlay when that happens and watch what the autopilot yaw channel is doing.

You can clearly see it more or less matching my manual inputs until 0:44 where it suddenly jumps to the left of my current pedal position.

1:14 is much easier to see. The yaw channel indicator is all the way at the left edge of its range of movement, and when I apply enough right pedal, the autopilot suddenly just gives up and removes some of the left yaw trim it was previously using, which causes the nose of the heli to snap to the right.

And again at 1:42, you can see it matching my inputs and then suddenly jump to the left.

If it had only happened once, it might've been an accidental input on my part. But three times in a row, with alternating left-right-left turns, and with the autopilot yaw channel making the same weird movement every time? No way. I'm 100% convinced that the heli itself is doing something it shouldn't be doing (or something that's not properly documented).

[Floyd1212]

23 hours ago, Pixel of Life said:

Pay close attention to the green autopilot yaw channel indicator on the controls overlay when I start my first turn at 0:30. I'm very slowly and carefully applying left pedal and the autopilot keeps fighting me, and then at 0:44, you can just about see the autopilot make an unexplained yaw input to the left which causes the nose to snap to the left.

 

12 hours ago, Pixel of Life said:

and when I apply enough right pedal, the autopilot suddenly just gives up and removes some of the left yaw trim it was previously using, which causes the nose of the heli to snap to the right.

 

You are asking the helo to hold your position by using the Position Hold mode, but then when you deliberately apply pedal inputs to the controls, you seem to be surprised that it finally "gives up" and does what you are asking. 

As pointed out above, there is a "breakout" point where the SCAS stops trying to hold the position you asked it to, and it instead tries to assist you with your new inputs.  If you want to avoid this, you can do one of two things: 1) disable Position Hold before making your pedal inputs, then re-enable Position Hold when you are done, or 2) re-trim the helo in the new desired position.

I have recorded a video with some narration to describe my experiences with this, and how I change heading while holding a hover.

 

[Pixel of Life]

7 hours ago, Floyd1212 said:

As pointed out above, there is a "breakout" point where the SCAS stops trying to hold the position you asked it to, and it instead tries to assist you with your new inputs.

So the bold part is the issue then. Why does it do this? What's the purpose of it suddenly switching to assist mode with no warning, instead of simply maintaining full left or right trim once it runs out of control authority, and expecting the pilot to override it which would be much more predictable and controllable?

If realistic, in what situation would this behavior be useful? If I'm just trying to make a slow turn, I don't need or want it to "assist" me because it's actually doing the opposite by making unexpected inputs. And if I need to make a quick turn to evade an incoming missile or whatever, that little blip of left or right yaw it gives me makes little to no difference because I'll be giving it a lot of pedal anyway.

[Floyd1212]

Hopefully this will improve over time, but in the meantime there are the two options mentioned above to avoid this behavior.  1) Disable ATT Hold before making your adjustments, or 2) trim during/after making your adjustments.

I have a feeling a pilot in the real aircraft would press and hold the FTR button while making his/her adjustments, then release when they are done.  As the SCAS is implemented in DCS right now, it won't fight you as long as you are holding the FTR button.

[Akiazusa]

This may not related to the topic,but a small question here.

Which direction will the Apache tend to roll in a low G situation?  Just curious.

[Grennymaster]

If you are using the "hover" hold. why do you have to move the nose so much? The taktic is: you sneak into your fire pos and use the Hellfire to engage your targets, wich will steer to the laser you or a buddy is firing. If you fired on one or two targets you will change your position.

You need to work with the Systems not against it. Thats what flying a attack chopper is. Thats what all formes Apache Pilots always talk about in there Vids. Im pretty sure the now better than any of us.

I dont whant to be bad or personal, im just confused why so many guys complain about the bad FM. ED works with real Pilots of that Bird so the should get it as near to RL as possible.

so if anyone feels personaly attaked by my sry mate, whas never ment to be