Jump to content
Sign in to follow this  
v8_laudi

Helicopter physics impressions - simplified

Recommended Posts

Well not entirely, The Alpha FM was good in terms of responsiveness but it still had serious problems.

-Helicopters picked up speed to fast and even in turns and stuff

-You could do unrealistic maneuvers without consequences (like flying upside down for some minutes)

-Completely impossible to loose hight and speed at the same time (This is still a problem)

So it is both a control authority and a FM Problem (I use M+K, by the way.)

Compared to what we have now and to what we had in the Alpha, Arma 2 has the superior helicopter handling.

Two of these things would increase the difficulty of flying. I'm not saying that it's optimal as it is, I'm just testing the assertion that the fm can be more realistic and easier. It really sounds like it's more complicated than that, and realistic changes usually mean increased difficulty.

Share this post


Link to post
Share on other sites
Two of these things would increase the difficulty of flying. I'm not saying that it's optimal as it is, I'm just testing the assertion that the fm can be more realistic and easier. It really sounds like it's more complicated than that, and realistic changes usually mean increased difficulty.

Wich two would that be? Improving those things would make flying easier. Maybe I worded it wrong.

Share this post


Link to post
Share on other sites

I believe part of the problem is inconsistency between the authority of different control methods AND that the ranges and linearity of some analogue axis is screwed up mathematically, in that +100% to -100% of the physical axis ends up as something less and as low as +50% to -50% which then gets shuffled up to fit in the +100% to >0% zone while ignoring the <0% to -100% zone.

Collective axis for instance, you are either forced to use half the axis range, thus end up with an overly sensitive top half of the axis and a none functioning bottom half, or if you used a method to get it working across its full range you then end up deficient in the minimum collective value forcing you to map a digital button to the analogue collective down axis to yield the same decent rate as a keyboard.

You get a similar effect if you map an analogue steering wheel axis to vehicle steering, you end up ~1/2 the steering angle of the keyboard working a none linier fashion even though none is applied, it’s as if a +512 to -512 axis just ends up as 512 split as +256 to -256.

Try mapping an axis you know works like the lateral cyclic axis or the rudder/tail rotor of a joystick etc to vehicle steering and compare it to the steering angle of the keyboard.

Take throttle brake applied to a combined axis of +256 to -256, turbo mode ends up being mapped to the very beginning of the reverse and forwards axis rather than the very end of the forwards axis and it also feels that while the physical axis extends from +100% to -100% the game is using +99% to -99% which causes when at or near the the centre point to osculate between +1% and -1% perpetually causing maximum forward speed or maximum reverse speed when the physical axis is at/near the centre position.

Try mapping an axis to the throttle/brake function like your collective or forwards/backwards cyclic axis of your joystick then position the joystick at/near the centre point, in essence there should be NO input, now switch to keyboard to make a throttle/brake/reverse input, you will end up in a perpetual forwards/reverse change as the “0†points (physical device & game) keep sliding past one another switching between +100% to -98% and +98% to -100%

In essence as far as I can tell there are only 3 principle axis in the game that work correctly with parity between keyboard and a controller, they are the cyclic function and rudder/tail rotor function, every other principle axis is screwed up un some way mathematically be it steering of vehicles, collective or throttle>brake<reverse function.

Share this post


Link to post
Share on other sites
Wich two would that be? Improving those things would make flying easier. Maybe I worded it wrong.

Decreasing acceleration and creating consequences for impossible maneuvers would increase the difficulty of flight, I think. I should say that I would like to see realistic acceleration, and also consequences for upside down flight/ negatively loaded rotors, etc.

Share this post


Link to post
Share on other sites
Decreasing acceleration and creating consequences for impossible maneuvers would increase the difficulty of flight, I think. I should say that I would like to see realistic acceleration, and also consequences for upside down flight/ negatively loaded rotors, etc.

Decreades acceleration (compared to the Alpha) actually makes flying easier. In the Alpha it was very hard to fly tight precise maneuvers because the helicopter always picked up speed. Turning in on a attack run or a landing was therefore extremely difficult since you had to decrease speed after the turn. And decreasing speed fast was only possible if you gained hight at the same time. Doesn´t work well for landings^^

Share this post


Link to post
Share on other sites
Decreades acceleration (compared to the Alpha) actually makes flying easier. In the Alpha it was very hard to fly tight precise maneuvers because the helicopter always picked up speed. Turning in on a attack run or a landing was therefore extremely difficult since you had to decrease speed after the turn. And decreasing speed fast was only possible if you gained hight at the same time. Doesn´t work well for landings^^

I dunno, I always find flight easier to predict if I have more control over how the helicopter is behaving... so I know if I tip it back 10 degrees, it's having an effect, and that effect will cause the helicopter to stop its forward motion at such a distance, rather than having this huge vague area. I find that fosters a lot of over-controlling, because the helicopter is so sluggish that you have to give it a lot of control to get an immediate response, causing you to need to correct immediately, causing a pilot induced oscillation.

Share this post


Link to post
Share on other sites
I dunno, I always find flight easier to predict if I have more control over how the helicopter is behaving... so I know if I tip it back 10 degrees, it's having an effect, and that effect will cause the helicopter to stop its forward motion at such a distance, rather than having this huge vague area. I find that fosters a lot of over-controlling, because the helicopter is so sluggish that you have to give it a lot of control to get an immediate response, causing you to need to correct immediately, causing a pilot induced oscillation.

I get the impression that you're trying to apply the schema of fixed-wing aircraft handling to choppers. I'm not sure if I'm right but that's the impression that I get. It just doesn't work, choppers can't be flown like jets, they're completely different kind of machines. However, I agree that controls of the choppers are really sluggish at this moment (the controls were great in the latest Alpha builds, I want them back so badly :pray:). It feels so weird that if you keep your collective down to keep the lift at very minimum and tip the chopper back that 10 degrees, the thousands of kilograms weighing large block of metal starts to gain altitude! It feels like you're flying a sheet of paper rather than chopper.

I'm absolutely sure that it makes flying choppers much more difficult, because it's pretty much impossible to kill your speed and decrease your altitude at the same time with most of the choppers now – partially caused by the reasons mentioned by b101_uk. This causes a lot of issues to regular (or newbie) chopper pilots, because when they try to slow down to land, they tip their nose back and lower the collective, but – thanks to the flight model – they still end up skyrocketing to high altitude, then shot down by AA within 5 seconds. You have to make highly unrealistic maneuvers to make a fast insertion like doing circles, multiple U-turns and snaking, and sometimes it feels that you simply can't slow down (especially with PO-30 Orca), the speed can even increase when you turn.

How should it be improved then? In my opinion, there are four crucial fixes that should be made:

1. increasing the mass of choppers

2. increasing the amount of inertia of choppers (most of them feel like that they don't have it at all atm in stable build)

3. retrieving the responsiveness of controls that was present in the last (stable) Alpha builds, there was almost no input lag at all

4. fixing the inconsistency between the authority of different control methods and the ranges and linearity of some analogue axises (check b101_uk's post)

But what are the influences of these fixes to the newbies and regular pilots? How to avoid them crashing all the time then?

I believe that all those fixes would make it not only more realistic but also much easier to fly choppers, because if you want to slow down, you can do it without gaining altitude thus being able to approach your LZ faster and in cover, that would decrease the risk of getting shot down by AA remarkably, and on the other hand, the responsiveness of the controls allows you to make accurate corrections. Additionally, pilots should always be "one step forward" from the situation, so that you're kind of "realizing your plan" all the time instead of touching the controls first and then thinking. This could be taught to players with simple chopper training mission in the singleplayer included in the core game. It could contain simply exercises of basic handling of choppers, eg. hovering, taking off, proper landings and also some more theoretical stuff like that "always being one step forward" (like planning the flight route and LZ before taking off), fast and tactical insertions, how to avoid getting shot down by AA, eg. flying in cover and how to make autorotation in case of engine failure (well, it should be made possible by devs first :p). Very simple to create (I can barely script and even I have done a mission like that!), but big influence to the players.

Edited by Ezcoo

Share this post


Link to post
Share on other sites

I agree with Ezcoo, helicopters weigh like feathers. They are unplayable.

Share this post


Link to post
Share on other sites

While helicopters should be able to lose speed and altitude at the same time, there is no magic brake! At a 10 degrees pitch up attitude with collective all the way down, it takes about 7 seconds for a Ka-50 to slow down from 250 to 200 km/h. Now imagine you want to land: that'll take another 25-30s at least, unless you start flaring (which won't allow you to lose altitude as quickly). Descent rate is about 3 to 6 m/s (watch this video, from DCS Ka-50:

).

Now of course, this is from a simulator, and I entered a vortex ring state twice in this video (also, terrible piloting...).

Arma's collective is weird due to its auto altitude hold (kinda), but otherwise does its job well, maybe needing a tiny bit more responsiveness. However, descending while in hover would already be fast enough to enter a VRS in real life, so descending faster in Arma would be unrealistic, since VRS is not simulated.

Also, holding collective down from a long distance still allows you to lose some speed and altitude, but you have to take into account that starting a descent at 300+ km/h just isn't gonna work (again, my only experience with real life is simulators, such as DCS Ka-50 and some X-Plane demo). I still agree helicopters tend to gain speed/altitude too easily in that case, as pitching up 10 degrees is impossible without gaining altitude until you get below 150 km/h or so.

Responsiveness of helicopters, as stated previously, is too low in the Beta, and was much better in the Alpha. When flying with the mouse or a joystick, it should also be even slightly higher than in the Alpha for quick inputs.

Then of course, there is the speed issue: world record helicopter speed is around 400 km/h. Faster speeds mean the blades of the rotor start losing lift, because they spin in the vacuum they create momentarily by moving the air.

Most helicopters have a top speed of around 250 to 350 km/h.

Acceleration is also a big issue, as has now been repeated enough: helicopters don't accelerate that fast, even with full collective and an aggressive pitch down attitude. (which, by the way, slows the rotor down due to the high stress in the blades, eventually leading the rotor to stall in some extreme situations).

From DCS Ka-50 Black Shark again, my fastest acceleration in a level flight is around 7 seconds to accelerate from 100 to 200 km/h; in Arma 3, it takes about as much, or even less, to get from 0 to 200.

Again, I'm not asking for a realistic flight model - that would make inexperimented pilots run away - but I still want some realism to be added to helicopters, and a decent flight model helps a lot.

Share this post


Link to post
Share on other sites
I believe part of the problem is inconsistency between the authority of different control methods AND that the ranges and linearity of some analogue axis is screwed up mathematically...

Totally agree and hope devs take note if they ain't already aware of the issue! Maybe making a ticket would be a good idea?

/KC

Share this post


Link to post
Share on other sites

the lift effect when you pull up at high speeds is correct, the rotor is a wing and if you pul up 10 degrees you climb in proption to the speed you are going. most helicopters land along a glide path for this reason, they almost never do a vertical descent into the LZ.

easiest way to counter it is to come into the LZ off center, and bank before you start pulling back so you "lift" into the LZ itself.

once you understand how the heli will act in different parts of a landing sequence (so do them in the editor) you can go from 300km/h to landed in about 20 secs

Share this post


Link to post
Share on other sites
Totally agree and hope devs take note if they ain't already aware of the issue! Maybe making a ticket would be a good idea?

/KC

Several I think. Although it would be good if Ezcoo could add his nicely summarized technical explanation to the active one...

Share this post


Link to post
Share on other sites
Several I think. Although it would be good if Ezcoo could add his nicely summarized technical explanation to the active one...

Done :) Put it in this ticket, hopefully it's the right one...

I agree that you shouldn't be able to fly close to LZ with high/extreme speeds and then quickly kill your speed without gaining much altitude, but I think that most of the choppers feel way too light currently. I'm asking to rather correct it than overdoing it so that you can fall like a rock at any speed. I also think that there are severe issues with flight models of some choppers - PO-30 Orca is maybe the best example from that: it gains speed even in U-turns – but I'm hopeful that those will be fixed soon.

Share this post


Link to post
Share on other sites
I get the impression that you're trying to apply the schema of fixed-wing aircraft handling to choppers.

I'm talking about flying helicopters in the arma series. In Arma 1 they had relatively low acceleration but very high control authority compared to arma 2 and the flight model they have now. I find I over-control the helicopters on landing almost never now, but frequently did in Arma 1, even after years of practice.

Share this post


Link to post
Share on other sites
Helicopters are capable of hovering upside down to a certain extent. Helicopters increase and decrease altitude by tilting there blades not by changing the speed of them so in order to hover upside down they just tilt them the opposite way. Now you would not do that in real life because it's pointless and dangerous, but it can be possible.

I don't think that's right at all

IMG00039.GIF

The lift is generated by the upper part of the wing (rotor blade) being longer then the lower section. As Bernoulli's law states, paraphrased here, the air flying over the top of the wing is traveling faster because it is traveling further than the underside. The result is a difference in pressure which causes lift. If the blade was upside down, the lift would be generated to the ground, right?

Share this post


Link to post
Share on other sites
I'm talking about flying helicopters in the arma series. In Arma 1 they had relatively low acceleration but very high control authority compared to arma 2 and the flight model they have now. I find I over-control the helicopters on landing almost never now, but frequently did in Arma 1, even after years of practice.

Ok, thanks for correcting me. I'm sorry for misunderstanding.

I don't think that's right at all

http://www.cavalrypilot.com/fm1-514/IMG00039.GIF

The lift is generated by the upper part of the wing (rotor blade) being longer then the lower section. As Bernoulli's law states, paraphrased here, the air flying over the top of the wing is traveling faster because it is traveling further than the underside. The result is a difference in pressure which causes lift. If the blade was upside down, the lift would be generated to the ground, right?

I don't know if it works with real choppers as well, but you can fly remote-controlled choppers upside down.

Share this post


Link to post
Share on other sites

Im no pilot but I find it odd that so much emphasis is placed on making it easier for new people to fly. My experience is (arma 2) you get told to go practice then play. That is what I did and I am somewhat proficient at flying now. When making missions I often spawn in a littlebird and do a flyby of the island to find interesting places and practice flying.

Share this post


Link to post
Share on other sites

Hi guys

Its beyond ridiculous. I've read that the FM from ToH is being ported to ARMA III but, IMHO its just as bad (I have ToH as well so I know). Have any of the devs tried the helicopter FM in say FSX or DCS or even BF3 (heaven forbid)??? you should be able to control your vertical movement not only just with your collective but also with your cyclic by trading airspeed for altitude. I tried this yesterday in the blackhawk (whatever this is called now) and nothing...ie. I dropped off a team at Kosmino coast and flew along the coast back towards the airport at about 20m altitude and 250kmph. I decided to try going over the hills so I put in full collective (more speed) and pulled back on the cyclic and.......nothing. I didn't gain a metre. All I did was bleed off airspeed. It seems that the only way to gain altitude is to go directly vertical from a standing position then forward (as if its an arcade type vehicle).

Not only that...but try bleeding off airspeed in a heli. Oh thats right...you almost can't.

Maybe I'm wrong I dunno but atleast try to get it resembling RL a little bit.

Please please BI...do something with the FM. Helis are an important asset in ARMA. You can't make a 'realistic combat simulator' and yet have heli's using a dumbed down flight model. Helicopters are inherently difficult to fly...make it so ingame so that players have to actually upskill themselves to fly it.

Share this post


Link to post
Share on other sites
Ok, thanks for correcting me. I'm sorry for misunderstanding.

I don't know if it works with real choppers as well, but you can fly remote-controlled choppers upside down.

Please take note of the swift loss of altitude when this helicopter loops,

Another reason you would rarely see helicopters upside down is due to articulation, otherwise it would potential damage if not break the mast.
...

You need to go into Options-Controls- Helicopter and change "Collective" to Analogue Collective", then your cyclic will influence lift and descent.

Edited by NodUnit

Share this post


Link to post
Share on other sites

anyone used the most recent dev branch update? as alot of the BLUFOR helicopters seem to be more slugish, like what happened to the littlebird.

also i just found out small arms fire (6.5mm) causes damage to buildings

Share this post


Link to post
Share on other sites
I don't think that's right at all

The lift is generated by the upper part of the wing (rotor blade) being longer then the lower section. As Bernoulli's law states, paraphrased here, the air flying over the top of the wing is traveling faster because it is traveling further than the underside. The result is a difference in pressure which causes lift. If the blade was upside down, the lift would be generated to the ground, right?

As mentioned, RC helicopters, which work on the same principle, can fly upside down. Also, the rotor is really just a wing. Fixed wings have the same profile, and airplanes can fly upside down. It'a matter of angling the wings to the air, the wings can work upside down, just not very efficiently.

---------- Post added at 09:49 ---------- Previous post was at 09:47 ----------

Hi guys

Its beyond ridiculous. I've read that the FM from ToH is being ported to ARMA III but, IMHO its just as bad (I have ToH as well so I know). Have any of the devs tried the helicopter FM in say FSX or DCS or even BF3 (heaven forbid)??? you should be able to control your vertical movement not only just with your collective but also with your cyclic by trading airspeed for altitude. I tried this yesterday in the blackhawk (whatever this is called now) and nothing...ie. I dropped off a team at Kosmino coast and flew along the coast back towards the airport at about 20m altitude and 250kmph. I decided to try going over the hills so I put in full collective (more speed) and pulled back on the cyclic and.......nothing. I didn't gain a metre. All I did was bleed off airspeed. It seems that the only way to gain altitude is to go directly vertical from a standing position then forward (as if its an arcade type vehicle).

Not only that...but try bleeding off airspeed in a heli. Oh thats right...you almost can't.

Maybe I'm wrong I dunno but atleast try to get it resembling RL a little bit.

Please please BI...do something with the FM. Helis are an important asset in ARMA. You can't make a 'realistic combat simulator' and yet have heli's using a dumbed down flight model. Helicopters are inherently difficult to fly...make it so ingame so that players have to actually upskill themselves to fly it.

It's been mentioned before, but the other factor about FMs is how the AI handle it.

Share this post


Link to post
Share on other sites
...

It's been mentioned before, but the other factor about FMs is how the AI handle it.

Couldn't AI use current FM and let players choose which one they want for themselves?

Share this post


Link to post
Share on other sites
As mentioned, RC helicopters, which work on the same principle, can fly upside down. Also, the rotor is really just a wing. Fixed wings have the same profile, and airplanes can fly upside down. It'a matter of angling the wings to the air, the wings can work upside down, just not very efficiently.

Well, the power to weight ratio of an RC helicopter is more akin to a UFO than a helicopter. Also, they are normally less articulated than a real helicopter's rotors. When an RC helicopter is flying upside down, it is not supporting a 6,000 lbs metal aircraft above the rotorhead on hinges. All in all, I don't think aircraft scale very well in terms of performance.

Share this post


Link to post
Share on other sites
Well, the power to weight ratio of an RC helicopter is more akin to a UFO than a helicopter. Also, they are normally less articulated than a real helicopter's rotors. When an RC helicopter is flying upside down, it is not supporting a 6,000 lbs metal aircraft above the rotorhead on hinges. All in all, I don't think aircraft scale very well in terms of performance.

Agreed, with my previous post I rather meant that it's possible in theory, but probably not in practice due to those reasons. Rotors might even hit the tail of choppers and mast would be under extreme stress (I believe masts are not designed to handle that in the long run).

Has anyone tested the changed made to FM in recent dev builds? Unfortunately I didn't see any major difference, MH-9 and AH-9 still feel very sluggish, but I think that the problem is now actually in handling the controls input (both responsiveness and the weird "smoothing" that appeared in Beta) than in the actual flight models themselves. Here's a nice video that shows how agile they really are.

Btw, one thing that has always bothered me: I think that the overall feeling (of realism) is more important than single details. With that I mean that because everything can't be simulated (eg. you can't feel the position of chopper and changes in it that equilibrioception would do IRL, and your actual FOV is much smaller than it would be IRL – I mean the restriction caused by screen sizes, also those slight changes in the airspace around chopper that make it to twitch a little bit aren't simulated), some parts should have stronger effect than in real life to compensate the effect of restrictions. One example of this is the agility of MH-9 and AH-9 that was present in the last Alpha builds: if we took a look at it very closely we'd maybe find out that the responsiveness of controls and agility of those choppers was actually higher than in real life, but because of the aforementioned restrictions in simulation, the overall experience might still be more realistic and enjoyable with those too high agility and responsiveness of controls to the most of the people. I think that many people won't agree with me, but it really felt like I was flying a chopper when I flew MH-9 and AH-9 in Alpha. Their speed accelerated quite quickly though, but overall experience was still much better than in Beta in my opinion.

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
Sign in to follow this  

×