correct as is Yaw and Pitch inertia values are Reversed.

[NineLine]

On 5/16/2023 at 3:34 AM, Fox One said:

 Here is a knife edge flight without altitude loss:

 

As can be clearly seen from the video, to maintain horizontal flight in knife edge at this alt/speed the pilot must create a beta angle of approximately 10 deg.

I have tried in simulator the maneuver in similar alt/speed conditions. Pressing the rudder pedal fully, the beta angle that can be generated in no more than about 5 deg. The lift produced is insufficient and altitude is quickly lost.

When you tried in the simulator everything was equal to the flight seen in the video exactly?

[FusRoPotato]

21 hours ago, DummyCatz said:

Never heard of that. IIRC the Angle of Sideslip feedback function is unavailable for the FLCS when AOA < 10 deg.

Sorry I meant to say 10 degrees. This isn't part of AOS feedback function, this is just a made-up hard limit that ED said matches publicly available data. Was it actually limited to 5 degrees in the game?

[TheBigTatanka]

Any updates on this? If these values are indeed inverted, it would explain a lot of the wonky feeling in the flight model.

Sent from my Pixel 4a (5G) using Tapatalk

[Fuggzy]

I watched Thunderbird Falcons fly knife edge passes a good 1/3 of the runway at multiple airshows as a kid. Is this the Mandela Effect? 

Edit: Doubtful on mandala. Seems the Tbirds do knife edge to at least some extent.  There's a table in this pdf with the High Low Flat maneuver packages, definitely includes knife edge:   https://www.faa.gov/about/initiatives/airshow/media/military/Thunderbirds_Maneuvers_Package.pdf

 

Edited June 7, 2023 by Fuggzy

[FusRoPotato]

I've not seen any evidence that any hard limit ever existed for any block. I think it's just some nonsense someone made up on a forum somewhere and ED rolled with it. It's certainly not like it'd be the first time...

[TobiasA]

Back to the original post regarding the inertia values... Any updates / comments on this from ED?

It would indeed explain a lot, and people have been noticing the sloppy or delayed pitch for quite some time.

[Raven (Elysian Angel)]

6 hours ago, TobiasA said:

sloppy or delayed pitch

To be fair to ED, I think at least part of the problem people are experiencing might be due to their specific hardware setup: If I bind pitch to my Virpil stick for example, it indeed feels very sluggish as if there is a built-in dead zone. But with my normal setup (for flying the Falcon) which is an R3L, there is no problem at all.
And developers model the flight models after the real thing, not after consumer hardware so you can't blame them for that.

That's not to discount the research people have been doing in this thread regarding the NASA research data, obviously. I'm just saying that using a force-sensing stick such as the R3L, there is no sloppy nor delayed pitch.

[TheBigTatanka]

Other guys with the same force sensitive stick say they also get the "sloppy" pitch.

I'm on a virpil stick, and I get a noticable rubber-banding of pitch. The FPM doesn't stop where you place it, instead, it oscillates up and down for a few seconds.

Going back to the original post, I think it's reasonable to think that the yaw and pitch inertia values could be reversed in the code, giving this kind of excessive pitch inertia.

Sent from my Pixel 4a (5G) using Tapatalk

[TobiasA]

vor einer Stunde schrieb Raven (Elysian Angel):

To be fair to ED, I think at least part of the problem people are experiencing might be due to their specific hardware setup: If I bind pitch to my Virpil stick for example, it indeed feels very sluggish as if there is a built-in dead zone. But with my normal setup (for flying the Falcon) which is an R3L, there is no problem at all.
And developers model the flight models after the real thing, not after consumer hardware so you can't blame them for that.

That's not to discount the research people have been doing in this thread regarding the NASA research data, obviously. I'm just saying that using a force-sensing stick such as the R3L, there is no sloppy nor delayed pitch.

The thing is that you get your gun cross where it should be, and instead of keeping it there, the FLCS corrects back to 0.5G before it comes back to 1 instead of easing up until it reaches 1G.
It is is if you had to much gain on the pitch controller inside the FLCS. Or not enough. Or the wrong inertia.

[TobiasA]

The FLCS overshoots the 1G mark when releasing the stick, for example.

[Raven (Elysian Angel)]

That would explain why I am so crappy with guns-only fights in the Falcon 
Well, let's hope it gets fixed then.

[Nealius]

On 7/15/2023 at 4:10 AM, TobiasA said:

The FLCS overshoots the 1G mark when releasing the stick, for example.

Would this explain the current tendency for the FPM to slowly rise when hands-off stick, even though it is showing 1.0G in the HUD?

[TobiasA]

vor 10 Stunden schrieb Nealius:

Would this explain the current tendency for the FPM to slowly rise when hands-off stick, even though it is showing 1.0G in the HUD?

The FLCS trims for 1G, so it has a tendency to rise the nose, the more AOA you have the more noticeable it becomes. This should be correct. 

[TobiasA]

Just a question... Is this fixed? Feels like it isn't as unstable as it used to be.

[FusRoPotato]

No, hasn't been touched as far as I can tell.

[TobiasA]

Am 14.7.2023 um 21:10 schrieb TobiasA:

The FLCS overshoots the 1G mark when releasing the stick, for example.

But this seems to be gone...

[FusRoPotato]

13 hours ago, TobiasA said:

But this seems to be gone...

It was never a specific enough claim to begin with. There are plenty of conditions where it would have undershot and overshot returning to 1 depending on speed, altitude, and loadout. 

I once thought in a previous patch that they did something to improve pitch control, but what it ended up being was just a curve adjustment. The pitch response, delay, and overshoot behavior was completely identical. I haven't gotten around to measuring this update, but nothing feels different yet.

Edited July 25, 2023 by FusRoPotato

[TobiasA]

vor 12 Stunden schrieb FusRoPotato:

It was never a specific enough claim to begin with. There are plenty of conditions where it would have undershot and overshot returning to 1 depending on speed, altitude, and loadout. 

I once thought in a previous patch that they did something to improve pitch control, but what it ended up being was just a curve adjustment. The pitch response, delay, and overshoot behavior was completely identical. I haven't gotten around to measuring this update, but nothing feels different yet.

 

It might as well be store related. However, I have not witnessed such a big overshoot when correcting to 1G coming from 5G +.

I'd ask my squad buddies, but many of them have switched to other airframes for this and other reasons and do not intend to go back...

[SFJackBauer]

On 7/25/2023 at 2:16 AM, TobiasA said:

But this seems to be gone...

+1 on seeing a difference in pitch response. I used to go to the Free Flight mission in Caucasus to test it under different speeds, and the wobbling in pitch was always noticeable, but not anymore.

[FusRoPotato]

Well I took some measurements and compared to much earlier results:

• No change in delay, still 60 ms. Should be near zero because those systems were either using an analog or digital signal at 400hz.
• No apparent change in response gains or inertia. The value in the FM file is untouched.
• Change in amplitude, which may be from a faster release but is more likely a general reduction of max controller gain. Something changed. I'm not sure what, but whatever it was, it has nothing to do with inertial response or input latency. The curves leading up to release and settling behavior looks to have very similar behavior and frequencies.

Maybe one of you can clarify what you mean by overshoot from 5G to 1G? I do notice there is a hump at high speed where, when returning to 1G from higher G's, it briefly stops short. I don't know if this is correct behavior or not because I have not created or found high quality state and control matrices to apply the FLCS schematic to, nor do I think it will really matter much because it is a very brief motion. At very slow speeds, you can see it does get stuck a bit and continues to climb for a while before settling back to 1G.

 

Edited July 26, 2023 by FusRoPotato

[NineLine]

Dear all, sorry for taking so long to get back to you.

So for the OP, as it was explained to me, there seem to be some differences in Western and Eastern aerodynamics axis naming. In our aerodynamics X is pointed forward, Y is up and Z is to the right. To add to this, we have different values in alternative source documents as well. Based on this, the sim is correct as is for the issue of Yaw and Pitch inertia values. 

As for other discussions beyond the original report, you will be required to make your own report and provide proper evidence for it to be looked at. 

Thanks.