DCS headshake suggestion

[TX-EcoDragon]

There are some components of the DCS headshake that I think are pretty well done, but there are also a few overdone effects with respect to roll as well as under-damped lateral translation.

 

DCS forces the virtual pilot's head to roll opposite the bank angle as though the virtual pilot is trying to level the horizon with his head and this is something that is generally considered poor form in both combat and precision aerobatic flight - not to mention that it's pointless to model in the first place because a PC pilot can roll their head to level the horizon if they want to simply by rotating their actual head!

 

Most well trained pilots don't do it much in real aircraft, and I certainly don't want my head to do it without my consent, especially given how simple, and realistic it is that I can do it if I so desire, in the exact same fashion I would if I were actually in a banked aircraft. Because of this, I find enabling the headshake feature to be too distracting and frustrating and hope that I've made the point sufficiently clear that the horizon leveling head-roll should be removed or made optional in DCS.

 

Additionally, in DCS there is a little more lateral displacement during coordinated banks/rolls than I'd expect to encounter in real aircraft, even when rolling at rather low roll rates. Perfectly coordinated turns and rolls will not result in as much lateral displacement of the pilot's torso/head - especially in slow rollers like the Mustang, or the helis (a passenger may experience a more exaggerated displacement because they do not anticipate the accelerations as the pilot will).

Edited May 20, 2013 by TX-EcoDragon

[gavagai]

Here we go again. :)

 

When you fly you do use your abdominal muscles to hold yourself steady in a roll, don't you? With track-ir I can keep my view steady just by moving my body a little bit. It makes the sim more of a whole body experience.

 

The head-roll thing I can see though, but, again, it is easy to counter with track-ir. I'm going to do some real flying in a few weeks and I'll let you know if I revise my opinion.

[JG14_Smil]

It's a simulator, so the real-life argument is just subjective. To simulate shaking, something has to shake, regardless if your eyes really do that in real life.

 

You used to be able to edit this, I did so it would match my tactile feedback. I'll have to look where as 1.2.4 wiped it out.

 

I too wish the cockpit would stay level and the world did all the banking. It would be best for cockpit builders that way.

[JG14_Smil]

I explain what I did in this thread. Not sure if it works in 1.2.4. You do need to be careful with it as it can cause fatigue of overdone. It does seem to depend on FPS too.

 

http://www.ivibe.com/forum/showthread.php?t=2977

Edited May 19, 2013 by JG14_Smil

[TX-EcoDragon]

The head-roll thing I can see though, but, again, it is easy to counter with track-ir.

 

The suggestion that a DCS pilot should roll the aircraft left, and then as the sim forces him to lean right, the PC pilot should then lean left in equal and opposite magnitude in order to attempt to hold his head vertical is simply foolish.

 

DCS is pretty well establishing the current benchmark for flight sim fidelity, and the forced head roll has no place in the series, perhaps beyond making track playbacks look more dynamic...and more like the pilot is a bit sloppy.

 

There is ample evidence available that high level combat and aerobatic pilots are instructed to, and practiced at, resisting the temptation to roll their heads to level the horizon. This is not only for absolute precision in aircraft control, gunnery, coordination detection, prep for opposite roll/yaw acceleration, but also to prevent damage to the cervical spine when at elevated G's. [Don't bother asking the Cessna drivers]

 

 

It's a simulator, so the real-life argument is just subjective. To simulate shaking, something has to shake, regardless if your eyes really do that in real life.

 

Yes, there's no debate about headshake itself here - my post is with respect to the forced, pointless, and largely unrealistic pilot head roll opposite the banked direction.

Edited May 19, 2013 by TX-EcoDragon

[gavagai]

The suggestion that a DCS pilot should roll the aircraft left, and then as the sim forces him to lean right, the PC pilot should then lean left in equal and opposite magnitude in order to attempt to hold his head vertical is simply foolish.

 

Why is it foolish?

 

If you're in a 70 degree bank and undergoing radial acceleration, there is a vector component pushing your head/body toward the other side of the cockpit.

 

If it seems like too much that you shod nod your head a tad to counter it, then you can increase your track-ir roll sensitivity and you will hardly notice it.

[Echo38]

If you're in a 70 degree bank and undergoing radial acceleration, there is a vector component pushing your head/body toward the other side of the cockpit.

 

If it seems like too much that you shod nod your head a tad to counter it, then you can increase your track-ir roll sensitivity and you will hardly notice it.

 

Two issues with your suggestion--the sim pilot must move his real head-with-TrackIR when banking, in order to prevent his sim-head from tilting to match the horizon; this is unrealistic. As Ecodragon pointed out, a real pilot does not ever need to tilt his head to match the horizon, so a sim pilot should never need to tilt his head to match the horizon--whether it's his real head or his virtual head, he should never have to tilt either. And it should never automatically tilt to match the horizon, because this doesn't happen in reality (not unless you've taught yourself a bad habit, and even then you can simply consciously refrain).

 

Secondly, what about those of us--such as myself--who don't have (or cannot medically use) TrackIR? We're stuck with having our virtual heads forcibly aligned with the horizon, in an unrealistic manner, unless we turn off the head shake altogether (which is also unrealsitic).

 

In short, no realistic option regarding this exists for anyone, at present. While not an urgent matter, Ecodragon has a point--this could use some work in the future.

Edited May 19, 2013 by Echo38

[TX-EcoDragon]

Why is it foolish?

 

If you're in a 70 degree bank and undergoing radial acceleration, there is a vector component pushing your head/body toward the other side of the cockpit.

 

That is incorrect. In coordinated turns of any magnitude there is no perceptible force pushing you to either side of the cockpit. In your example of a 70 degree bank, assuming constant altitude, and coordinated flight, the pilot will experience a net load factor of +3Gz with no perceptible lateral G (Gx). In the case of uncoordinated flight, the slip/skid angle will result in a lateral acceleration that will move the pilot to the side of the cockpit if they do not correct this by restoring coordinated flight via proper rudder usage - this is the so called "seat of your pants flying" and in a well implemented headshake, the view will most certainly move laterally in this case...but NOT during coordinated flight.

 

If it seems like too much that you shod nod your head a tad to counter it, then you can increase your track-ir roll sensitivity and you will hardly notice it.

 

My TrackIR profiles have the roll axis disabled and the reasoning is much the same as my initial point. Think about it!

 

It makes sense that we must (until we get better VR setups) apply a multiplier to the pitch and yaw axis of TrackIR because we must still look forward at the monitor even when we are checking 6 etc. However amplifying roll serves no realistic function. If I tilt my head 10 degrees, what *should* happen is the view on my monitor rotates 10 degrees, and with your TrackIR roll axis disabled that's exactly what happens (WOOT! 100% REALISM!). However, with my TrackIR roll function enabled then my pilot appears to be sitting upside down in the airplane (BOO! 0% REALISM). Why introduce all of these artificial obstacles to the one part of our view system which works 100% realistically? Yeah...it's foolish, yeah, it's "too much."

Edited May 19, 2013 by TX-EcoDragon

[gavagai]

That is incorrect. In coordinated turns of any magnitude there is no perceptible force pushing you to either side of the cockpit. In your example of a 70 degree bank, assuming constant altitude, and coordinated flight, the pilot will experience a net load factor of +3Gz with no perceptible lateral G (Gx). In the case of uncoordinated flight, the slip/skid angle will result in a lateral acceleration that will move the pilot to the side of the cockpit if they do not correct this by restoring coordinated flight via proper rudder usage - this is the so called "seat of your pants flying" and in a well implemented headshake, the view will most certainly move laterally in this case...but NOT during coordinated flight.

 

What is perceptible is going to be very different for an experienced pilot versus a new pilot. Also, as for Josh's point, whether nodding your head or not is realistic depends on whether you want to simulate the muscle exertion in the pilot's neck and abdomen. Is it realistic if the virtual pilot does not have to use his neck or abdominal muscles? There is no fact of the matter here. It is a subjective preference of simulation (such factors do exist).

 

As I see it, the roll could be turned down a bit and other things left as they are, and it would be fine.

 

-------------------

 

Ok, the math is not wrong, but I left out gravity.:P That would seem to be in good opposition to the lateral force I show below. So, you would need about a 6g turn at 70 degrees bank to feel 1g of lateral force. Probably not possible in a coordinated turn.

Edited May 20, 2013 by gavagai

[TX-EcoDragon]

It's not a matter of perception, or opinion. I don't know how to be any clearer than I've already been. In a coordinated, level turn, there is no net lateral acceleration on the pilot. Case closed.

 

Yes, your drawing is wrong in a few ways. Most notably is that you aren't accounting for gravity, and the fact that the forces sum such that....you guessed it, there's no net lateral force on the pilot.

 

 

See page 176 of Aerodynamics for Naval Aviators

http://www.faa.gov/regulations_policies/handbooks_manuals/aviation/media/00-80T-80.pdf

 

You may also find these useful:

 

http://www.free-online-private-pilot-ground-school.com/Load_factors.html

 

http://selair.selkirk.bc.ca/training/aerodynamics/turns.htm

 

http://www.docstoc.com/docs/53946162/Turning-Flight

 

http://www.av8n.com/how/htm/maneuver.html#sec-turns

 

http://www.recreationalflying.com/tutorials/groundschool/umodule1b.html

 

and of course: http://www.youtube.com/watch?v=pMWxuKcD6vE

Edited May 20, 2013 by TX-EcoDragon

[theGozr]

I do have lots of experiences in many different aircraft here going from MiG15/17, All Yak9U, Turbine Legend, etc.. etc.. and I will have to side with EcoDragon on this one 100%.. I also do know and saw that EcoDragon fly in Aerobatics and very good in simulation and consultant in physics for various Simulators. I would recommend to listen. ;)

Edited May 20, 2013 by theGozr

[9.JG27 DavidRed]

i dont know whether i see the same what EcoDragon reports...i have headshake enabled, cause just like he pointed out, in some aspects ist very well done in my view.

it really helps to kind of feel your plane.

 

but i tried again after reading his post, and i just dont see this automatic leveling with the horizon.its only at the beginning of a roll, where your head lags a bit behind, but as soon as i have reached the desired bank, and continue with a coordinated turn, the view is steady again, without me having to counter any ingame headroll movement...

 

in my opinion, while in the air, this natural headmovent option is pretty good implemented...

but i think when on ground this option is a bit overdone...the plane may shake in real life pretty hard while taxiing or starting up on ground, but i think ist not as distracting with the headshake option disabled.there the plane still shakes, but not as pronounced, and mainly the dashboard is vibrating,while having a steady view.

with the headshake option enabled, the whole airframe shakes, and thus gives the impression of waggle head.

 

PS: dont get me wrong, i completely agree with the op and that it souldnt be this way,...but i just dont seem to have this problem....

Edited May 20, 2013 by 9./JG27 DavidRed

[MACADEMIC]

Will have to agree with the OP.

 

The head movement in the opposite direction of the turn effect could be a heritage of car simulations, where this would be appropriate due to the centrifugal forces. Sure it's a nice graphic effect livening up the simulation. Most people drive cars, so they probably can more easily be immersed when shown something they're very familiar with and that's deeply seated in their muscle memory.

 

However I agree it shouldn't be there in a high fidelity flight simulation.

 

In an airplane, all that matters is the total of the load vector, shown in cockpit with that ingeniusly simple slip/skid indicator (the ball in the curved glass).

 

As long as the ball is in the center, flight is 'coordinated' so no side forces occur. Only in a skidded turn (ball showing towards outside of turn) would the body and head have a tendency to move to the outside of the turn, while in a slipped turn (ball on the inside of the turn) to the inside.

 

MAC

[bkthunder]

Well, I'm a RL pilot too and I totally agree with EcoDragon, in a coordinated turn there is absolutely no force pushing you to the side. This by the way is one of the reasons for spatial disorientation, if you closed your eyes in a coordinated turn you wouldn't even know you were turning.

 

On the matter of leveling your head with the horizon, in my experience that's something that most pilots involuntarily do during shallow turns (this can be seen in many cockpit videos as well). In aerobatics though, it's something to avoid and as you pull more Gs you will naturally align your head with the G vector.

[Robert1983NL]

In a helicopter and when flying low to the ground it's better to turn your head to keep the horizon level in my opinion. It's more natural.

[IvanK]

Nothing Natural about Helicopters :)

 

I agree pretty much with Eco in this debate.

[Kuky]

I'm not a real pilot but I don't use head shake either and my TrackIR roll axis is mapped 1:1 (I tilt my head 10° and in game I see the same... I wish I could do the same with X and Y axis :)

[gavagai]

Yes, your drawing is wrong in a few ways. Most notably is that you aren't accounting for gravity, and the fact that the forces sum such that....you guessed it, there's no net lateral force on the pilot.

 

So I am wrong but I also guessed it right?:D I'll have a look at your links to see what is there. I see your point now that the forces have to sum to be zero, but what about gravity? That has to still be there, doesn't it?

 

Like I said before, I am open to revising my opinion. I'll pay attention to this point when I go flying next month.

Edited May 20, 2013 by gavagai

[gavagai]

Ok, I made a sum of the forces equation in order to prevent the sloppiness of my first attempt.

 

As the attached table shows, for a 60 degree bank, a load of 2g is required for a coordinated turn.

 

Am I reading it correctly that the load factor is opposite the lift vector, and not the radial acceleration vector?

Edited May 20, 2013 by gavagai
for a question

[TX-EcoDragon]

On the matter of leveling your head with the horizon, in my experience that's something that most pilots involuntarily do during shallow turns (this can be seen in many cockpit videos as well).

 

Yes, most pilots of lower performance aircraft where maneuvering flight is incidental to their pilot duties. As maneuvering flight becomes the focus, you'll see fewer puppy-dogs in the pilot seat.

 

In a helicopter and when flying low to the ground it's better to turn your head to keep the horizon level in my opinion. It's more natural.

 

I'm sure many people will be more comfortable head tilting. If you want to head-tilt in the sim aircraft you fly and the operations you fly it in, then that's simple enough to do...just tilt your actual head, even if you are a hat switch user your head actually tilts relative to the instrument panel and horizon just as in the real aircraft so there's no reason for the sim pilot's virtual head to tilt as well.

 

 

Nothing Natural about Helicopters :)

 

I agree pretty much with IvanK in this debate.

Edited May 20, 2013 by TX-EcoDragon

[Robert1983NL]

Yes, most pilots of lower performance aircraft where maneuvering flight is incidental to their pilot duties. As maneuvering flight becomes the focus, you'll see fewer puppy-dogs in the pilot seat.

Again, this is fixed wing only. In helicopters it's more natural and better imho to tilt your head so the horizon stays level.

 

I do agree that the basis should always be that the instrument panel is the horizontal 'baseline', and trackIR should let the user decide how to tilt the head.

[TX-EcoDragon]

Am I reading it correctly that the load factor is opposite the lift vector, and not the radial acceleration vector?

 

Yes

 

 

Again, this is fixed wing only. In helicopters it's more natural and better imho to tilt your head so the horizon stays level.

 

It's most natural and better imho that we can have consent over what our necks do, just as we would if we were in the aircraft. As my reply to you before suggested, I'm not here to tell heli pilots they gotta keep their heads straight up if they don't want to, I'm here to point out that we can do that even with headshake off and without TrackIR, so we don't need our virtual pilot doing it non-consensually, and those of us that do not want our heads to tilt are forced to turn off "Pilot head motion from G's forces" or whatever it's called in DCS.

 

In an airplane, all that matters is the total of the load vector, shown in cockpit with that ingeniusly simple slip/skid indicator (the ball in the curved glass).

 

As long as the ball is in the center, flight is 'coordinated' so no side forces occur. Only in a skidded turn (ball showing towards outside of turn) would the body and head have a tendency to move to the outside of the turn, while in a slipped turn (ball on the inside of the turn) to the inside.

 

MAC

 

Exactly right!

 

I do agree that the basis should always be that the instrument panel is the horizontal 'baseline', and trackIR should let the user decide how to tilt the head.

 

OK, we mostly agree then! But again - TrackIR should have no place in head tilt, unless we're talking about a guillotine simulator! Imagine the red horizontal line in this forum is the top of the glareshield of your instrument panel, say you want to tilt your head 10 degrees relative to that line, how do you do it? Hint: You don't need to turn on your TrackIR!

 

See what I'm getting at?? When you tilt your real, actual human head, your eyes perceive that tilt...there's no reason to further augment the tilt. If you do, you end up with this cosmic portal of death:

Edited May 20, 2013 by TX-EcoDragon

[gavagai]

Edit: ok, thanks for the help.

 

Going through a full free body diagram and a sum of the forces equation, I do get that the net force in the x-axis is zero. I definitely stand corrected! Anyway, ecodragon, you indicate that you hope you explained enough in your OP, but some of us need this stuff spelled out before we can really believe it. In the end, I think it's better to be convinced by the math and physics than to just take someone's word for it.

 

:thumbup:

[TX-EcoDragon]

Bottom line: I'd love to be able to enable the otherwise very well done head shake in DCS and not be forced into sloppy virtual head tilting.

 

There's no reason to keep that component of the headshake at all, but if it could be optional that would be great!

 

This was brought up in 2008, and I have had to have DCS headshake off ever since: http://forums.eagle.ru/showthread.php?t=33832

 

I'd love to turn it on and get the other effects.

 

Please?

 

I'll say thanks by buying the Huey!! :joystick:

Edited May 20, 2013 by TX-EcoDragon

[Robert1983NL]

OK, we mostly agree then! But again - TrackIR should have no place in head tilt, unless we're talking about a guillotine simulator! Imagine the red horizontal line in this forum is the top of the glareshield of your instrument panel, say you want to tilt your head 10 degrees relative to that line, how do you do it? Hint: You don't need to turn on your TrackIR!

 

See what I'm getting at?? When you tilt your real, actual human head, your eyes perceive that tilt...there's no reason to further augment the tilt. If you do, you end up with this cosmic portal of death:

 

Lol. I have to disagree, I understand you can tilt your head without trackIR as well ofcourse, but here's an example how I tilt my head, flying a helicopter:

 

If I'm in a sharp turn I like to learn forward, to look over the nose and tilt my head at the same time to see as much as possible of the terrain, so if I'd be in the left seat turning left, I would basicly look up through the roof window and get the horizon straight and see as much as possible into the turn. The other way around, left seat and right turn, I need to lean forward, look past the nose and past the upper console to look into the turn. Again, getting the horizon level is more natural and provides the best FOV, since human eyes have wider horizontal view than vertical.