Jump to content
Sign in to follow this  
hailstorm

Parallax effect on the HUD

Recommended Posts

The pilot viewpoint is not "a foot in front of the rotational axis"

Yes, you're quite right Suma. Thanks for taking time to consider this issue. I was exaggerating, but it does appear to cause very significant changes in viewpoint - it's still enough to cause big aiming errors (as you've very nicely described may I add)

If you want to get the HUD working as accurate as possible, listen to what RKSL-Rock is talking about. I'll bet my paycheck if there's anyone on this forum you can rely on for accurate information on HUD's he's likely the best source.

Now correct me if I'm wrong Rock, but HUDs will always seem to move independently of the glass it's displayed on - it's a natural effect caused by the design of the device, the same has how a mirror never portrays an image that looks like it's attached to the mirror itself, whereas a TV hooked up to a camera pointing in the same direction as the screen will always have a similar image no matter where the viewer stands relative to it. So from that point, the HUD image will move according to a changing POV.

Where the difference lies (again, correct me if I'm wrong) is that the HUD device will not be collimated to the exact point a round will land - it's just not physically possible given ever-changing ranges and the static location of the HUD-image creator on an aircraft. For example, in the video Rock linked, the main HUD image appeared to be projected roughly a metre further behind the HUD glass, but no more. The ways around this are having multiple projected images at different (short) ranges like a rifle iron sight, or having specially designed HUD glass etc. But all of these appear to rely on the pilot having to, to a large degree, keep their point of view in roughly the same location relative to the aiming device, exactly the same as how a soldier has to keep his eye down the rifle sights.

On the other hand,

Personally, I'd prefer that the pilot's viewpoint was completely static in the cockpit when the pilot looks around, and not fixed a foot in front of the rotational axis like it currently is.

The reason I think making a pilot's viewpoint completely static was based on several reasons:

- From a programming perspective, it's a simple, elegant solution to the whole problem. Whether the HUD is collimated to infinity or simply to the HUD glass, it's irrelevant since the device will be accurate. Adding in head sway or eye movement while looking around might add that slight feeling of realism, but that is quickly destroyed when gun sights instantly become inaccurate as a result - which causes more time and effort to find a working HUD solution etc. It's just the simple, relatively real solution.

- While a pilot's eyes will move around when a they turns their head, the pilot will most likely place his eyes and POV in the exact same spot every time they use the HUD for targeting. But since there is no way a computer program can guess when a player wants to do this when the avatar's head is in the wrong position, it's a much better solution to assume that the player will want to look accurately through the HUD at all times - hence, keep the POV in the same position while he/she looks around.

- From someone who has an fixed-wing acrobatics rating, one of the most important things I learned to do is to keep the head as still as possible while maneuvering - even though my planes do not have HUDs. Besides the fact that keeping the HUD accurate is a reason to keep the head still, a static head also prevents disorientation for such dizzying effects like losing the ability to balance that the inner ear provides. Just today I fell victim to this when I looked down to check my map the same second my Co-pilot initiated a climbing left turn - the perceived backwards and tilting motion (as opposed to the expected motions of the maneuver) on my head made me instantly sick, and the only way to fix it was to keep my head still and my eyes fixed on the horizon for the next five minutes, hence why I fully agree with this in regards to the new added head sway:

In my opinion the camera movement under G is too dramatic which is making it worse.

Watch this cockpit video of an L-39 experience flight - and notice how much the guy's head moves while doing 2g and 3g turns.

He pretty much doesn't move at all. I'd be very pleased if the current head movement was toned down to less than a third of what it is now, if not removed altogether - it's just detrimental to fighting from aircraft in general.

Share this post


Link to post
Share on other sites

Personally I hope BI can reach a compromise without having to resort to a static viewpoint as the head movement really adds to the immersion of flying in A2.

@Hailstorm, are you really saying that you can keep your head immobile when pulling off high G maneuvers?

For me personally as long as the pilots head re-centres shortly after the initial high G move I will be happy.

Disclaimer: I know nothing about the science of head up displays. I just know I love the new head movement when flying in what is after all a game and not a hardcore flight simulator.

Edited by stun

Share this post


Link to post
Share on other sites

while yes it may not be realistic considering how much the player's head moves during flgiht , but it is probably there to simulate the feeling - not the movement

Share this post


Link to post
Share on other sites
...

Now correct me if I'm wrong Rock, but HUDs will always seem to move independently of the glass it's displayed on - it's a natural effect caused by the design of the device, the same has how a mirror never portrays an image that looks like it's attached to the mirror itself...

Correct, but only within the "cone of projection".

Where the difference lies (again, correct me if I'm wrong) is that the HUD device will not be collimated to the exact point a round will land - it's just not physically possible given ever-changing ranges and the static location of the HUD-image creator on an aircraft.

Exactly.

But there are more modern setups coming along soon that use head position tracking systems to compensate. We are diverging from the the real topic here but its kind of relevant since this technology is now being used in some attack helicopters and the latest generation of aircraft.

The F-35 with its dependance on Helmet Mounted Displays is a prime example. The system tracks the pilots head position relative to 3 or more fixed points in the cockpit and adjusts the display in his visor to compensate. The same goes for the latest generation of HMD Helmets that EADs and others are developing. The Eurofighter will have it at some point as will quite a few others: Eurocopter Tiger, AW129, CH47F, AW159 Lynx Wildcat, AH-64D and others.

For example, in the video Rock linked, the main HUD image appeared to be projected roughly a metre further behind the HUD glass, but no more.

The distance depends on the projector Lens. In theory the further its "projected" the more accurate the alignment should be.

The ways around this are having multiple projected images at different (short) ranges like a rifle iron sight, or having specially designed HUD glass etc. But all of these appear to rely on the pilot having to, to a large degree, keep their point of view in roughly the same location relative to the aiming device, exactly the same as how a soldier has to keep his eye down the rifle sights.

What you are describing is the way the Harrier HUD works, actually quite a lot of older generation HUDs. Its called a Dual combiner holographic HUD. It uses two reflecting plates with different refractive indexes to make a composite image.

http://www.richardfisher.com/images/design_examples/full_size/dual_combiner_hud.gif

And its the same basic principle of a truly 3D holographic HUD. True 3D displays are still being developed but the more modern HUDs like that of the Typhoon use different layers projected to different depths to give an even better accuracy of alignment rather than use multiple plates.

- From someone who has an fixed-wing acrobatics rating, one of the most important things I learned to do is to keep the head as still as possible while maneuvering - even though my planes do not have HUDs. Besides the fact that keeping the HUD accurate is a reason to keep the head still, a static head also prevents disorientation for such dizzying effects like losing the ability to balance that the inner ear provides. Just today I fell victim to this when I looked down to check my map the same second my Co-pilot initiated a climbing left turn - the perceived backwards and tilting motion (as opposed to the expected motions of the maneuver) on my head made me instantly sick, and the only way to fix it was to keep my head still and my eyes fixed on the horizon for the next five minutes, hence why I fully agree with this in regards to the new added head sway:

If you fly high G manoeuvres yourself you know how important it is to strap in properly, while you head does move around due to loading, do you ever slide around as much as is represented in ArmA?

Dont get me wrong, im not advocating a return to the fixed view point. As annoying as the movement can be it is at least immersive. But i would like it toned down and corrected.

For example in our Typhoon (I use this as an example because its very noticeable), pull a high G loop and its almost like your head slides down into your body. I've not really done much more than 3-4G in an old C152 Aerobat but I'm pretty sure I didn't duck down so much. Nor move around as much as is shown in game while manoeuvring. But since you are supposedly strapped in and since most ejection seats are designed to support and restrain your head while under G I really don't think the movement should be this extreme.

He pretty much doesn't move at all. I'd be very pleased if the current head movement was toned down to less than a third of what it is now, if not removed altogether - it's just detrimental to fighting from aircraft in general.

Tone it down about 50% and i think it should be fine.

Re 6dof someone mentioned earlier. I'd love it if it was possible. Especially in helicopters. I'd vote for that

Edited by RKSL-Rock
This forum software sucks arse. It keeps deleting the images.

Share this post


Link to post
Share on other sites

Although your head may not move due forces, it's a great and immersive way to portray those forces. Makes the game appear more polished, and provides a more fluid experience.

Now if we could only have the same system for dampening, it would be great. You know, that horrible viewing effect when you're on a bike in rough terrain you're head feels like it's on a stiff stick.

Share this post


Link to post
Share on other sites

Tone it down about 50% and i think it should be fine.

Re 6dof someone mentioned earlier. I'd love it if it was possible. Especially in helicopters. I'd vote for that

It would be better if the effect could be tweaked per vehicle so yo have the less strapped in tight (or none) ground vehicles where you would move about more opposed to a helicopter and moreso a jet.

6dof sounds nice but it would be nice to not have the head tilt motion since in real life your eyes focus on something..simulate the eye not a camera and all that. But you can't exactly lean forward into your seat in ALL helicopters either, the apache for example is very restricting.

Share this post


Link to post
Share on other sites

Yeah it would be nice if we could set the amount of movement. But I guess that would be a bit more difficult to implement.

... But you can't exactly lean forward into your seat in ALL helicopters either, the apache for example is very restricting.

It depends on the harness. Most of the helicopters ive seen use inertia reel systems similar to a car's seat belts. The UK apaches has a limited mobility system. Its essential for the gunner especially. Move too quickly and the harness grabs you.

Fast jets, well anything with an ejection seat has to be far more restrictive. But most aircraft have some sort of inertia belt system. eg:

http://www.conaxfl.com/products/personal-restaints/aircrew-harness-restraint-systems.aspx

Edited by RKSL-Rock
added link

Share this post


Link to post
Share on other sites
- From a programming perspective, it's a simple, elegant solution to the whole problem.

As it is almost impossible to realign the pilot animation so that his eyes stay always on the same spot, I am more and more inclined to implement this. There are potential problems hidden in the implementation, though. The major potential issue is when the player avatar movement is so significant your eyes will be significantly detached from it, and you may end up seeing your body from an obviously impossible angle. Hopefully the effect will be not that bad. Another difficulty is the implementation. Currently when determining camera position we just ask the pilot what his current head position is. This would be no longer possible, and we would have to ask him what his head position would be assuming he was not turning the head. This is not something impossible, but it is a little complication.

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  

×