Head Tracking Help

[Leg2ion]

Hi All - after some advice/assistance if possible?

 

Complete novice/newbie with headtracking - and have just purchased DelanClip (awesome turnaround by the way - ordered Wednesday night - received Saturday morning) and have set the camera up and installed OpenTrack. 

 

Am now following this guide (A complete guide to set up Head-tracking (Opentrack) - Hardware, Software and Controllers - IL-2 Sturmovik Forum) to calibrate and set curves etc, but also notice you can set curves within DCS.

 

Question is - what is the better way to set up for DCS, set the curves in OpenTrack, curves in DCS or a combination of both?

[MAXsenna]

I have the Delan clip, and I use OpenTrack with it. I don't use ANY curves in DCS. Didn't enten know I could.
Not on my computer know, but I will be happy to post my settings later.
I admit, that it was a little intimidating at first, fiddling with curves etc. Some use dead ones, I doubt think I do. But it all came together when I got used to it, and stopped bobbing my head around.
Cheers!


Sent from my MAR-LX1A using Tapatalk

[Leg2ion]

Thanks for that @MAXsenna. 

 

So far tuned 'Yaw' - what I have noticed is that at max 'look back' the image flickers in game - as if I'm on the edge of what I can view - so not sure how to stop that happening?

[MAXsenna]

True, if you lean backwards too much. If it bothers you, just reduce it. With my settings I can peak into the cargo compartment of the Mi-8.

Sent from my MAR-LX1A using Tapatalk

[Leg2ion]

6 hours ago, MAXsenna said:

I have the Delan clip, and I use OpenTrack with it. I don't use ANY curves in DCS. Didn't enten know I could.

OK - so stopped faffing with the OpenTrack curves. Reset them all back to default, re-calibrated the Delanclip and adjusted the curves in the axis tuning settings in DCS. 

 

Things seem to be ok(ish) - just need to fine tune/tweak it now, and re-learn how to land with tracking active!!!

[MAXsenna]

This new forum is strange... The text disappeared.

 

Anyway, what I tried to write...

 

Sorry mate, completely forgot.
Don't remember how I ended up with these curves, but I guess I just tuned and tuned.
I used to have different profiles for different aircraft, but now I only use this one.

 

Cheers!

 

 

 

 

 

 

 

 

 

 

 

 

 

Edited March 28, 2021 by MAXsenna
Missing text

[Hiob]

I don't use opentrack/delanclip but trackir - but I can say that I always tuned my headtracking in the trackir-software and left the curves in dcs alone.

I think of it as calibration. You would/should always first calibrate your Joystick before adjusting curves in DCS. Same for headtracking.

[Leg2ion]

Hi @MAXsenna - many thanks for that - as a test will 'default' the in game settings and have a ganders at yours and see how it compares. 

 

@Hiob - completely agree - I would rather set the pre-game software and leave DCS all alone - but just seemed easier to tune DCS.

 

Next question - but what does the asymmetric tracking do compared with the top graph curve - are they just different views of the same thing?

 

 

[Rudel_chw]

33 minutes ago, Leg2ion said:

Next question - but what does the asymmetric tracking do compared with the top graph curve - are they just different views of the same thing?

 

All these headtracking axes have the zero value on its mid position .. for example the axis that corresponds to your head's "pitch", 0 would be looking straight ahead, and it can go 90º up or 90º down ... on this case you might want an asymetric curve, since looking up (for bandits) is very different from looking down, where you rarely hit straight down, instead looking to instruments and consoles.

[MAXsenna]

Hi @MAXsenna - many thanks for that - as a test will 'default' the in game settings and have a ganders at yours and see how it compares. 
 
@Hiob - completely agree - I would rather set the pre-game software and leave DCS all alone - but just seemed easier to tune DCS.
 
Next question - but what does the asymmetric tracking do compared with the top graph curve - are they just different views of the same thing?
 
 

Exactly like@Rudel_chw says.
I wanted a smoother motion to look down at my instruments, compared to looking out of the cockpit.


Sent from my MAR-LX1A using Tapatalk

[Leg2ion]

Oh my! Not sure how that translates into the curves seen above - what differentiates between looking down slowly and looking up quicker?

[Rudel_chw]

15 minutes ago, Leg2ion said:

Oh my! Not sure how that translates into the curves seen above - what differentiates between looking down slowly and looking up quicker?

 

You should really give a try to adjust the curves yourself .. all setups are different: we have different monitors sizes, different distance to the monitor, different tastes .. noting replaces your own raw experience, you can't just simply copy someone elses curves and hope they will be good for you. Here is an old video where I show the trial-error process of adjusting the headtracker curves:

 

 

I don't use the same headtracking software as you, but the concepts are the same ... try it, it isnt hard 

[MAXsenna]

Oh my! Not sure how that translates into the curves seen above - what differentiates between looking down slowly and looking up quicker?

That would be pitch. But when I look at it now, it seems strange. The order of the screen shots got mixed up.
When I set it up first time years ago, I watched a Grim Reapers YT, and then took it from there.
As you can see, there are actually 6 axis. Hence six degrees of freedom. 6DoF.
Pitch, roll and yaw are like your stick/rudders. The xyz are forward/backwards, side to side, straight up/down.
The linear/non-linear curves I just happen to make after playing with it, until it felt natural for me, like I use my head while driving a car, sort of. Also the non-linear will help you look backwards without yawing your head to much.
Also pay attention to the different degrees, as they will determin "how far", the travel of movement can be. If that makes sense.
Cheers!


Sent from my MAR-LX1A using Tapatalk

[Leg2ion]

10 hours ago, Rudel_chw said:

you can't just simply copy someone elses curves and hope they will be good for you. 

 

I don't use the same headtracking software as you, but the concepts are the same ... try it, it isnt hard 

 

Thanks @Rudel_chw. Understand individuals settings are, well, individual, and would fully expect to 'tweak' - but curious to see others input settings to see how things are affected. The curves created in DCS seem ideal, so (as an example) yaw movement can be tuned so angular movement is fairly slow moving across the instrument panel, but then increases as you move towards looking at the edge of the monitor - hence trying to figure out how this is recreated using Opentrack. I am guessing whereas OT has the centre position bottom left of the graph, DCS has its central position in the centre of the graphic - so thinking if I replicate the movement I see in DCS into something similar in OT, then reset the DCS curves I should get roughly the same movement.

 

 

[ruddy122]

Like all these fine posters said

There is no quick and easy model

Each setup is completely different

Just know what those numbers mean

kinda like joysticks you just have to tweak to your liking


Sent from my iPhone using Tapatalk

[Leg2ion]

2 hours ago, ruddy122 said:

There is no quick and easy model
 

 

As I am learning! So - defaulted/reset all DCS settings, and went into the OT settings. Wow - what a difference - a horrible difference!!! Very jerky/laggy. So - nulled everything, left OT on its defaults - set my original curves in DCS - smooth. 

[cj43g3r]

I would recommend the following opentrack mapping curves based on the following for added realism:

• you are strapped tightly into the ejection seat for high G maneuvers so there will be little to no X, Y, Z axis movement
• typical camera on top of display can detect approximately +/-30deg face yaw and +/-30deg face pitch as input
• a human neck can do approximately +/-20deg roll
• a human neck can do approximately +/-75deg pitch with 5 degrees added for vertical eye movement. But you may be able to look +90deg in an inclined seat. And may require -90deg for cockpit controls.
• a human neck can do approximately +/-90deg yaw, but with some torso rotation and horizontal eye movement can reach approximately +/-160deg

Y and Z mappings are the same as X

null

null

nullnull

[SharpeXB]

1 hour ago, cj43g3r said:

you are strapped tightly into the ejection seat for high G maneuvers so there will be little to no X, Y, Z axis movement

That’s not realistic. (also you’re replying to a 4 year old thread )

 

[cj43g3r]

@SharpeXB Notice how the pilot's straps are very loose that they're constantly falling off.

You don't want to do that if you don't want neck/back issues. But to each their own...

An average human head is around 11 lbs. At 7.5G's, 82.5lbs. At 9G's, 99lbs. I think you get my point.

And I know I'm replying to a 4 year old thread because there are DCS users that still use head tracking.

I see a lot of YouTube videos using head tracking unrealistically with 180deg yaw or pitch.

Edited April 2 by cj43g3r

[MAXsenna]

@SharpeXB Notice how the pilot's straps are very loose that they're constantly falling off.
You don't want to do that if you don't want neck/back issues. But to each their own...
An average human head is around 11 lbs. At 7.5G's, 82.5lbs. At 9G's, 99lbs. I think you get my point.
And I know I'm replying to a 4 year old thread because there are DCS users that still use head tracking.
I see a lot of YouTube videos using head tracking unrealistically with 180deg yaw or pitch.

Well, then they tend to make a new thread, and even so, this thread was never about realism. The topic was to get headtracking working.

Sent from my SM-A536B using Tapatalk

[SharpeXB]

8 hours ago, cj43g3r said:

Notice how the pilot's straps are very loose that they're constantly falling off.

Right. The idea that pilots are strapped so tightly in their seats is a flight sim myth. 

8 hours ago, cj43g3r said:

You don't want to do that if you don't want neck/back issues. But to each their own...

Well the alternative is getting shot down. Which is worse for your neck/back.

[cj43g3r]

@SharpeXB Then why don't you post more realistic opentrack mappings ?

At 9G's it's not the straps but the G's holding your head against the headrest.

I see lots YouTube DCS videos where the pilot's head tracking is easily moving in all 6 degrees despite pulling high G's. And current head tracking software doesn't use DCS telemetry to scale the mappings.

[SharpeXB]

22 minutes ago, cj43g3r said:

Then why don't you post more realistic opentrack mappings ?

I use TrackIR but I’ve never found it necessary to create custom profiles, it works fine by default. The Smooth preset is the right one for flight sims where you want to look at 6:00

23 minutes ago, cj43g3r said:

I see lots YouTube DCS videos where the pilot's head tracking is easily moving in all 6 degrees despite pulling high G's.

No doubt it requires a lot of effort to move your head and body when pulling Gs. This can’t be simulated in a game though. Another CFS game simulates exhaustion effects and even the benefit of G suits. It’s rather a good effect actually. 

[MAXsenna]

9 hours ago, cj43g3r said:

@SharpeXB Then why don't you post more realistic opentrack mappings ?

Why should he? 

This is the original question in the OP:

"Question is - what is the better way to set up for DCS, set the curves in OpenTrack, curves in DCS or a combination of both?" 

You're completely off topic in a ≈four year old thread. I encourage you to make your own thread like "Realistic TrackIR curves". Too bad they can't be adaptive, so it will be close to impossible to perform any cold start without a full fidelity simpit or VoiceAttack/Macro keyboards. 

Mods, @NineLine @BIGNEWY, I believe this thread has run it's course. 

Cheers all! 

[cj43g3r]

@MAXsenna The title says Head Tracking Help, so I believe this applies.

 

When I was researching head tracking for DCS, I didn't even know this forum existed. And I only recently found out about this thread involving head tracking.

I couldn't find any opentrack mapping templates. So I researched various medical websites, NIST, and took measurements of many cockpit videos freeze frames to come up with the numbers i provided. You can match up the numbers with the axis of the various mappings. It will help new opentrack users to understand what they mean.

I used a linear mapping since it's easier to control.

There is also a deadband when looking straight ahead to avoid jitter from micromovements.

I would also recommend using any of the opentrack filters to reduce jitter when moving.

If you are flying warbirds instead of jets, it would be applicable to have linear curves for the X, Y, and Z axis.

These are just recommendations after all and the user can adjust them to personal preferences.