DCS: AH-64D Flight Model discussion

[CHPL]

vor 10 Stunden schrieb Floyd1212:

I fly the Apache exclusively, and I haven't personally noticed any changes in the way she flies after the big MT patch, or the two smaller follow-up patches since then.

Not a totally reliable way to test, but you could play back a track from before the March patch and see if the helo does something very different from when the track was recorded.  I remember going back to play an old track after the big adjustments were made to the tail rotor system and the aircraft was spinning out of control and crashing right after take-off.

Created a video with corresponding maneuvers right before the update and one right after.
In the second, maneuvers have succeeded that was not possible for me before. This concerns in the APACHE specifically the speed in the execution and the reactions of Tourqe and rotor speed.
This also becomes clear on the TFT displays for the performance in the APACHE. 
In video 1, yellow and red displays dominate, even during trivial maneuvers. In video 2, these colors are instead the exception.
It was also possible to fly more aggressively in 2 without constantly hearing the whining of the on-board computer "Low RPM - low RPM."  

[CHPL]

vor 10 Minuten schrieb Trigger:

Agree with Floyd1212. I also fly the Apache since its release mostly, and in the meantime didn't mess any more with the axes (all just neutral). To me, it's just a matter of patience and training. Couldn't find any difference since the last major update months ago. Only my personal learning curve is slowly but constantly rising.

No question many APACHE pilots in DCS started flying helicopters in the first place with this module.
I don't want to offend anyone when I say that the changes I'm talking about are in a range that many can't register because they are simply not realized due to lack of experience with other modules.
From experience (former CFI helicopter), I can also say that the APACHE is easy to fly but unsuitable as a training device for the helicopter basics. No RL APACHE pilot has learned to fly on an APACHE, and the question of cost is third-rate.
 
But there are also very, very few in DCS interested in the basics. Even the APACHE pilots in our squadron are fixated on electronic weaponry and avoid any realism in the sense of sneaking up, hiding, hitting, and disappearing.
The only recent exception is the ability to attack JATEC-lasered targets from cover.

INVITATION:
All those interested in the basics are welcome to join us for training. Individual instruction is also available by arrangement.
The UH-1 is an advantage in training, but not mandatory. 
               
https://forum.dcs.world/topic/244549-black-angels-low-level-heaven/#comment-5107028

[pii]

7 hours ago, CHPL said:

Thanks for your info and don't worry.
That was just a rhetorical question.  
However, with experience and a lot of practice, one becomes very sensitive to the control and recation of various modules.
However, the change in the APACHE was so substantial for me that an improvement of the flight model was out of the question.
Of course, it is even more interesting to find out what the reason for this change is if it does not concern the flight model.
I see this as a fundamental question not only concerning the APACHE. In our online squadron, many have made the experience that hardware is extremely noticeable, especially below 30 knots, but not only.
This is essential information for all those who have difficulties with takeoff, landing, hover, or other maneuvers and use, let's say, low-cost hardware. 

Well, I had a warthog and it is anything other than low-cost but I replaced most of it and the difference was night and day for the good. The warthog was a nightmare for the apache. So I replaced the pedals and my warthog base, with its heavy spring, for a Virpil base, and now I can fly like an ACE (sometimes) The difference is enormous!

So while hardware matters it's not always low-end hardware that is the problem.

[NeedzWD40]

I've only perceived incremental changes since release, nothing major or groundbreaking. As of the latest patch, I've noticed no real changes to handling.

I suspect for most people that the larger problem is sticks with springs that are far too stiff/strong for the smaller center adjustments needed for helicopters. It can be done, but it takes a lot of practice. Like bradmick says, it's more likely that players have stuck with it long enough to acclimatize to the characteristics.

[shagrat]

Am 25.3.2023 um 10:25 schrieb CHPL:

Created a video with corresponding maneuvers right before the update and one right after.
In the second, maneuvers have succeeded that was not possible for me before. This concerns in the APACHE specifically the speed in the execution and the reactions of Tourqe and rotor speed.
This also becomes clear on the TFT displays for the performance in the APACHE. 
In video 1, yellow and red displays dominate, even during trivial maneuvers. In video 2, these colors are instead the exception.
It was also possible to fly more aggressively in 2 without constantly hearing the whining of the on-board computer "Low RPM - low RPM."  

The major improvement is likely related to the immense increase in fps through the introduction of MultiThreading.

Now, what has fps to do with the flight model? Nothing, but it impacts the visual feedback loop between corrective movement of the cyclic stick (joystick) in your hand and noticable reaction on the screen.

Say we had 40 fps before and now stable 80 fps. For subconscious corrections you typically do with the cyclic to stabilize the helicopter it takes now half the time (actually time between frame changes, but let's simply say "time") to get a visual recognizable feedback on the screen. The movement seen on the screen is smoother and way more granular. 1/40th of a second vs. 1/80th of a second.

As for us sim-pilots there is no feedback from our inner ears movement sensor, but only the very limited 70-90° visual feedback of the screen, without peripheral vision, this "feels" now like the inputs are more controllable/smooth and responsive.

...and of course the fact we now fly the Apache for quite some time and training and muscle memory getting better. From my personal experience now my confidence will increase, I get a bit too cocky until I mess up more regular, again and increase my experience in addition to a good muscle memory and instinct, which let's me omit precarious situations before I am required to test my confidence and muscle memory in the first place.

[ShuRugal]

On 3/21/2023 at 9:07 PM, Raptor9 said:

No changes have been made to the AH-64D flight model.

Maybe nothing in the FM changed, but something in the behavior is different.  Picking up to a hover required roughly half the displacement of cyclic after the MT split was released as compared to a matter of hours previously.

My guess is performance optimization making the SAS operate more efficiently?  This isn't a case of "I got better" between the night before the patch was released and the next afternoon when i was able to play it.  My hover pickup technique in the apache calls for slowly picking up the collective while feeding in aft and left cyclic to keep level, and doing so since MT patch has required considerably less deflection of the cyclic to achieve a vertical pickup.

[Floyd1212]

1 hour ago, ShuRugal said:

Maybe nothing in the FM changed, but something in the behavior is different.  Picking up to a hover required roughly half the displacement of cyclic after the MT split was released as compared to a matter of hours previously.

Out of curiosity, do you have a track from before the MT patch that you can replay with the current build, to see if the helo misbehaves?

[cow_art]

I have no tracks and only subjective observations. But the first time I used the Apache in MT I also got the impression that it somehow feels better now. And I also find it highly unlikely that I just magically got better at the Apache over night. I wouldn't be surprised if the performance of the flight model / SAS is somehow degraded when a computer is limited by CPU performance. Or perhaps it is somehow related to frame rate consistency?

Anyway, according to the roadmap, SAS and flight model changes are coming with the next patch, so this discussion is hopefully going to be moot soon

Edited April 1, 2023 by cow_art

[solus]

As a person who have been flying helos in DCS since the the first installment of Black Shark that was sold on 2 CDs and who tried every helicopter in the game, I would say that flight model of AH-64 still has quite big room for improvement. 

Flying and turning became easy after 1 year of development, but hovering is still a lottery. Maybe 1 year later we will see easy-to-hover Apache. I can't imagine people in real life flying such unstable, unpredictable and super sensitive machine.

It seems that Apache module is also very demanding in terms of hardware - pedals, suitable joustick and high FPS. I must admit that I'm far from matching all those boxes, but it doesn't bother me when I'm flying other modules. 

[Sandman1330]

On 4/1/2023 at 1:39 AM, solus said:

As a person who have been flying helos in DCS since the the first installment of Black Shark that was sold on 2 CDs and who tried every helicopter in the game, I would say that flight model of AH-64 still has quite big room for improvement. 

Flying and turning became easy after 1 year of development, but hovering is still a lottery. Maybe 1 year later we will see easy-to-hover Apache. I can't imagine people in real life flying such unstable, unpredictable and super sensitive machine.

It seems that Apache module is also very demanding in terms of hardware - pedals, suitable joustick and high FPS. I must admit that I'm far from matching all those boxes, but it doesn't bother me when I'm flying other modules. 

Helicopters genuinely are hard to fly, especially hover. You can’t compare the KA50 as it’s coaxial rotor and stability systems negate much of the negative stability side effects of tail rotor equipped aircraft. The DCS Apache feels about right in this regard, it’s no harder to hover than the real helicopters I fly for a living. Your control setup is almost certainly holding you back.
 

Your next question would probably be, then, why do most helicopters not go with a coaxial configuration then, if tail rotors cause so much trouble? The answer is complexity. Coax rotor systems are extremely complex and maintenance heavy.

 

[corbu1]

vor 17 Stunden schrieb Sandman1330:

Helicopters genuinely are hard to fly, especially hover. You can’t compare the KA50 as it’s coaxial rotor and stability systems negate much of the negative stability side effects of tail rotor equipped aircraft. The DCS Apache feels about right in this regard, it’s no harder to hover than the real helicopters I fly for a living. Your control setup is almost certainly holding you back.
 

Your next question would probably be, then, why do most helicopters not go with a coaxial configuration then, if tail rotors cause so much trouble? The answer is complexity. Coax rotor systems are extremely complex and maintenance heavy.

 

I love this beautiful drawing of the Helo pilot doing his job……Seen it several times before, but every time I look at it at makes me so smile….Tank you!

[ShuRugal]

On 3/31/2023 at 5:38 PM, Floyd1212 said:

Out of curiosity, do you have a track from before the MT patch that you can replay with the current build, to see if the helo misbehaves?

Unfortunately, I do not.  I can only tell you that, prior to MT release, i had to hold about 25% left stick and ~33% aft stick to hover, and now those numbers are roughly halved.

What it feels like to me is that the SCAS was over-reacting prior to MT release (possibly due to longer compute cycles?) and that I have to fight it a lot less now.  

A good way to demonstrate this without the variability of human skill is to use the hover-hold mode.  Since MT release, the helicopter takes significantly less time to stabilize itself into a precision hover after engaging HH.  Prior to MT patch, it could take as long two minutes for the helicopter to stop "stirring" itself in a circle after engaging HH.  Now it takes as little as 30 seconds for HH to achieve a point hover.

[Floyd1212]

Interesting.  I guess under that theory you could switch back and forth between ST and MT and not just feel the difference, but see the difference on the controls indicator?

I haven't seen that kind of difference in where my controls are during a hover, but I'll play closer attention next time I fly.

[GrEaSeLiTeNiN]

I think the map (altitude, temp, wind) makes a diff in the case of the sensitive Apache. Maybe the 'bad experience' was in the mountains of Syria and the seemingly 'better' FM in the South Atlantic coast.  

[Rogue Trooper]

The Flight model improved with MT.

I have not tried ST as a comparison but in MT it improved without doubt. It is a small improvement but without doubt she is more stable and cleaner to pull into a hover. She offers the pilot more ability to anticipate what she will naturally do next and this allows the pilot to ride the chaos to his/her advantage. It is possibly due to more modelling time in the CPU as Multithreading is used to off load some workload to other cores..... perhaps.

I click trim by the way, I do not hold-adjust-release the trim..... without doubt the Apache has the best trim system of all DCS helicopters! ...... so forgiving .... so natural and liquid. It feels like it catches up with the pilot smoothly.... lovely!

Edited April 6, 2023 by Rogue Trooper

[Nereid]

On 3/18/2023 at 10:03 AM, CHPL said:

All are focused on multi-threading, while the APACHE flight model has significantly improved.

Well. I have just checked DCS again after a few weeks and mainly because of multi-threading support. And I tried the AH-64D just afer the F-18C and I have to agree. The flight model has improved A LOT. Before I always was struggeling to land it or to keep it under control in some situations. Sometimes it seemed to behave weird. And now all this is gone. It just flies like a charm.

[CHPL]

@Nereid

Thank you very much. You save my day.
But sometimes, it's funny with DCS. Especially with the different experiences people have.

[TZeer]

Could this be part of why it feels different?

 

Quote

Fixed: Aircraft weight remains same when FCR removed

This was fixed in the same patch as MT was introduced.

[Nereid]

18 hours ago, TZeer said:

Could this be part of why it feels different?

I don't think so. I was struggling with some weird behaviors that were hard to get used to for me. They are hard to explain but most often I lost a bit to much control when I was hard turning or slowing down. Sometimes the tail rotor suddenly seemed to do not any affect at all in one direction. But there were other minor effects, that were somewhat counterintuitive. I do not see them anymore.

Edited April 12, 2023 by Nereid

[GrEaSeLiTeNiN]

Some days it's good, some days not so good. Erratic at times. That's has been the overall development experience for me (regular warthog table top hotas) from day one. It's fun to shoot from, just not so fun to fly in.  

[Raven434th]

Seems to me to be a twitchy hot mess now

they need to publish what hotas setup they are calibrating the FM too....that would shed some light on what the FM is supposed to do vs what its doing on our systems.

Edited April 13, 2023 by Raven434th

[Apache 64]

vor 10 Minuten schrieb Raven434th:

Seems to me to be a twitchy hot mess now...

Really big changes this time. You're right about that. I fly almost every day for several hours and it has felt really good lately.

Did a few extra rides before today's patch, everything was fine.

Then patched and flew the same maneuvers in the same missions, but almost no decent flying possible anymore.

Looks like there will be many hours of new muscle memory training needed again. I also had to make some major adjustments to the axis settings.
Especially the collective reacts extremely tamer. You need much more input for the same result as before.

Everything will be fine. Hopefully.

[NineLine]

22 minutes ago, Raven434th said:

Seems to me to be a twitchy hot mess now

they need to publish what hotas setup they are calibrating the FM too....that would shed some light on what the FM is supposed to do vs what its doing on our systems.

 

1) no changes to the core flight model, just some updates to how the collective channel behaves

2) we do not model the FM to a HOTAS, we model the FM to the real thing. You would need to adjust your HOTAS to best represent that and what feels best for your personal taste.

Thanks

[Raven434th]

12 minutes ago, NineLine said:

1) no changes to the core flight model, just some updates to how the collective channel behaves

2) we do not model the FM to a HOTAS, we model the FM to the real thing. You would need to adjust your HOTAS to best represent that and what feels best for your personal taste.

Thanks

Which I did months ago...but it seems every patch of late changes the way that calibration behaves...so what's the point.And as you can see by the post above, I'm not alone.

[Amarok_73]

37 minutes ago, Raven434th said:

I'm not alone.

No, there's at least two of us with common feelings.

There was a time, like somewhere around last Christmas (before rather than after), when it was totally possible to get into stable, hands free hover on stabilizers and fly stable forward with stabilization enabled on both, altitude and attitude. Then next patch ruined it.
Now, whatever settings I'd try, I can't get into hover so stable, so I could try to leave the joystick to click anything with mouse.
I've tried different approaches, the curves, different special options, to purge totally the module and install it from scratch, all in vain.