TrackIR change in 1.2.6

[SharpeXB]

Did the profile or feel of the TrackIR change in 1.2.6?

It feel different. Faster and less ability to move back in the cockpit

[Harrysound]

Now you mention it......

[TurboHog]

They did add a proper neck recently. Now your head is not moving as if it is twisting around on a pole but you actually look over your shoulder!

 

Maybe this is what you experienced (as well).

[Lange_666]

I used 1.2.6 only briefly until today but the first thing i noticed was that it seemed that TrackIR was a lot less smooth in movement. Like when you turn your head, there is some sort of stutter to be seen. It's no real stutter but i get the feeling that the terrain movement has a more skippy movement then it used to be.

Honestly, i hate this.

I have serious problems with things like headshake or headroll and now this kind of movement. I really hate it when a computer tries to simulate some sort of movement of my head/eyes that i don't really experience live at that moment.

In real life, your brain detects this and compensates this somehow which it doesn't when i'm in front of a screen because i don't sense the motion. I can sit in a car going on a dirt track with my head moving all over the place without a problem but i can't sit in front of my computer driving a simulated car "trying" to give me the same experience. I end the game mostly after 30 secs, i just can't stand it.

Same goes for headroll/shake in flightsims. These are turned off by default because they make me sick. If i can't do that, the sim isn't played. And now this.

Does it mean i have to stop playing because of some "simulated implementation of head movement"? Crap!!

 

Please give us at least some option to turn it off.

Edited September 14, 2013 by Lange_666

[Wolf Rider]

@ Lange...

 

Totally agree - you're not alone

[Harrysound]

Well that's why I keep messing with the graphics settings because it just doesn't seem smooth as before....

[Brisse]

They did add a proper neck recently. Now your head is not moving as if it is twisting around on a pole but you actually look over your shoulder!

 

Maybe this is what you experienced (as well).

 

That was added in 1.2.5

[TurboHog]

That was added in 1.2.5

 

Yea thought so.

 

I'm using facetracknoir which is recognized as TrackIR input so it's essentially the same except for how movement of my head is determined. I have no such troubles...

[Conanliquid]

I agree. I have taken to no TrackIR at the moment which is ok I guess but not best. Fix would be appreciated. It certainly seems less fluid.

[Lange_666]

That was added in 1.2.5

i can't confirm that, skipped 1.2.5 due to real life.

[Lange_666]

Imho, what seems to be simulated are ones eye-movements when looking sideways at fast moving terrain. If you look at someone who looks at fast passing terrain you see his/her eyes quickly and constantly move from left to right (or vice versa). This should result is some stuttery motion of the passing terrain/landscape. But in real life your brain compensates this eye movement and gives you a fluid motion picture. The computer doesn't and because you look at your screen with your eyes hold still (you look straight forward at a sideways moving terrain) you don't get compensation for the computed jittery motion. Just a plain horrible experience imo.

 

Before reading this i was already tweaking my TrackIR profile and i also turned the smoothness up to max. This helped a bit but only a bit so far.

 

If this is the future it will be bye bye DCS for me. I'm already holding out on buying the Mi-8 add-on due to this.

[Flagrum]

Imho, what seems to be simulated are ones eye-movements when looking sideways at fast moving terrain. If you look at someone who looks at fast passing terrain you see his/her eyes quickly and constantly move from left to right (or vice versa). This should result is some stuttery motion of the passing terrain/landscape. But in real life your brain compensates this eye movement and gives you a fluid motion picture. The computer doesn't and because you look at your screen with your eyes hold still (you look straight forward at a sideways moving terrain) you don't get compensation for the computed jittery motion. Just a plain horrible experience imo.

 

Before reading this i was already tweaking my TrackIR profile and i also turned the smoothness up to max. This helped a bit but only a bit so far.

 

If this is the future it will be bye bye DCS for me. I'm already holding out on buying the Mi-8 add-on due to this.

That seems unlikely to me. The eye movement you mentioned is because you try to focus at a point. But as that point is moving away, you quickly find an other point near the position at the old point where you can focus at. And so on. But it makes no difference if the moving scenery is outside your car window or your monitor.

[PeterP]

.

 

If this is the future it will be bye bye DCS for me. I'm already holding out on buying the Mi-8 add-on due to this.

 

 

Lange,

Whatever it is you experiencing...

"Imho, what seems to be simulated are ones eye-movements when looking sideways at fast moving terrain. ..."

I can tell you that it's not caused by DCS . And the only thing what's new , as already pointed out , is the 'SchoulderSize' function.

 

It simulates a 15-20cm head shift to the side when looking further back as 90° - and that's a very close representation what happens in real live when reaching the 'gimbals-limits' of your neck.

 

Here is a 1:1 comparison of the new 'ShoulderSize' functionality. - both sides use the exact same Track-Ir input via a Track replay.

 

 

And nothing like a 'jitter' or whatever is added. - This must happen at your end and has to do something how your Tracjk-IR sees your tracking dots or you system is processing your movement to slow.

 

Examine the "Camera" window in your Track-Ir software to see what the camera sees.

I would suggest to change change your "Light Filter Threshold" setting of your Track-Ir camera and/or move it closer to your head somehow.

So you have a more reliable and smother tracking.

 

Here is a way to cancel out any inputs/interpretation of your head movement of the DCS.

Download the server.lua and place it in your saved-games folder as instructed >>> Reworked Cockpit Views with proper Neck.

But don't use it as t is - please make these edits (red) to cancel out any direct influences of the sim to your head movement - so you have it 1:1 as Track-Ir passes the info to DCS:

--/////////////////////////////////////// Global Pilot View parameters 1.2.5 - Start - PeterP ////////////////////////////////

 

local GlobalEyePoint = {0, 0, 0} -- {front/back , up/down , left/right} <- 'neck dimension' in meters

local GlobalCameraViewAngleLimits = {20,160} -- {min horiz. FOV in °, max horiz. FOV in °} <- aka Zoom

local GlobalCameraAngleLimits = {190,-75,115} -- {Max turning to back, max looking down , max looking up} <- in °

local GlobalShoulderSize = 0 -- move body when azimuth value more then 90 degrees

local GlobalShoulderSize_for_not_full_3D_Cockpit = 0 -- 0 = OFF - set to > GlobalShoulderSize = 0.15 or 0.2 < to turn it ON in the non fully 3D cockpits of FC3. Keep in mind that the 2D seat texture will look strange.

 

--/////////////////////////////////////// Global Pilot View parameters 1.2.5 - End - PeterP //////////////////////////////////

If afterwards you still have the impression that there is something like "stuttery motion of the passing terrain/landscape"

-it's definitive not inside of DCS - It's your system and how it's set-up.

Edited September 14, 2013 by PeterP

[_Dredd]

I have to disagree here somewhat Peter, something has been changed with 1.2.6, I'm seeing the same loss of smooth movement with trackir now compared to 1.2.5, it's not a bad thing just a slight nuisance.

 

However I am very pleased that frame stutter seems to have been fixed when watching moving terrain down low :thumbup: and my suspicions were that these are related. I have my trackir on smoothness 50, and I thought maybe somehow fps management or something vsync in game related was the cause of it, so I lowered smoothness to 30 and it is much better but still there. I don't know if this is coincidence, I do run through low frames of 28 to mid 50 and up.

 

EDIT: aside from initial experience, tnite I'll try default trackir profile, and see what effect different curves, settings have.

Edited September 14, 2013 by _Dredd

[Lange_666]

I can confirm that. It's not the Shouldersize Function that i have a problem with. I think is a great improvement. It's like _Dredd and other users say, it's the less smooth motion when looking sideways on (fast) moving scenery when your head is moving.

To be a bit more precise, it occurs only when you turn your head. The SchoulderSize should not have something to do with it since this only comes into play when looking over your shoulder. It occurs directly when looking sideways. The moment you keep your head still (or switch from TrackIR to fixed view) it all goes smooth again.

 

PS: And it's definitive not my system and/or how it is set up. It's not 100% setup for this sim alone, i still do other things with this computer. I still have a old standalone A-10C version installed and the difference is clear to be seen between the two.

[Wolf Rider]

Lange,

Whatever it is you experiencing...

"Imho, what seems to be simulated are ones eye-movements when looking sideways at fast moving terrain. ..."

I can tell you that it's not caused by DCS . And the only thing what's new , as already pointed out , is the 'SchoulderSize' function.

 

It simulates a 15-20cm head shift to the side when looking further back as 90° - and that's a very close representation what happens in real live when reaching the 'gimbals-limits' of your neck.

 

Here is a 1:1 comparison of the new 'ShoulderSize' functionality. - both sides use the exact same Track-Ir input via a Track replay.

 

 

And nothing like a 'jitter' or whatever is added. - This must happen at your end and has to do something how your Tracjk-IR sees your tracking dots or you system is processing your movement to slow.

 

Examine the "Camera" window in your Track-Ir software to see what the camera sees.

I would suggest to change change your "Light Filter Threshold" setting of your Track-Ir camera and/or move it closer to your head somehow.

So you have a more reliable and smother tracking.

 

Here is a way to cancel out any inputs/interpretation of your head movement of the DCS.

Download the server.lua and place it in your saved-games folder as instructed >>> Reworked Cockpit Views with proper Neck.

But don't use it as t is - please make these edits (red) to cancel out any direct influences of the sim to your head movement - so you have it 1:1 as Track-Ir passes the info to DCS:

If afterwards you still have the impression that there is something like "stuttery motion of the passing terrain/landscape"

-it's definitive not inside of DCS - It's your system and how it's set-up.

 

 

 

err... your "video" (Shoulder size on) shows the same stutter effect

 

 

Listen... what is trying to be recreated is the use of RL peripheral vision, and until a two point pivot, with a goose neck (for clarity) between the two points, it ain't a "proper neck", and that's in no way, no shape and no form. You don't like what I'm saying? fair enough... just call it a View Point mod and be honest about it, because that's all it is.

 

Now.. if what is a "true" Over the Shoulder Look , the view would tilt slightly (headroll).... the eyes don't stay level when looking behind with moving the head/ shoulder/ spine to see... the eyes are then on about a 15 - 20 degree angle.

Edited September 14, 2013 by Wolf Rider

[Flagrum]

Could it be that it indirectly is caused by the introduction of the shoulder? What I am thinking about - as total layman in this matter - is: without shoulder the eye point just turns around. The 3d scene does not change, only your pov. But with shoulder, the eye point actually moves and thus all angles also change for the new pov - i.e. has to be calculated new. So I would conclude that moving your eye point with shoulders causes more calculations and therefore also cause more strain on the cpu(?)/gpu(?)

[PeterP]

err... your "video" (Shoulder size on) shows the same stutter effect

 

errr... , video is recorded 2x times to play it side-by side on 15frames a sec and add the compression on the top.....

I wouldn't call it a true representation of smoothness in the sim. All what the video has to show (and it does it pretty good I think) is the changes in the shouldersize. - beside this - it provides no other meaningful data.

 

 

Could it be that it indirectly is caused by the introduction of the shoulder? What I am thinking about - as total layman in this matter - is: without shoulder the eye point just turns around. The 3d scene does not change, only your pov. But with shoulder, the eye point actually moves and thus all angles also change for the new pov - i.e. has to be calculated new. So I would conclude that moving your eye point with shoulders causes more calculations and therefore also cause more strain on the cpu(?)/gpu(?)

 

Before we get into thinking and believing...

(I prefer to do such things only in a church ;).. - and I have to admit that I hardly can recall when it last happened.)

 

Have you made the tests I suggested?

- I made them, and I had no change of any 'jitter' (if there at all) that is new or increased or wasn't there before the shulderzise was introduced.

[Wolf Rider]

errr... , video is recorded 2x times to play it side-by side on 15frames a sec and add the compression on the top.....

I wouldn't call it a true representation of smoothness in the sim. All what the video has to show (and it does it pretty good I think) is the changes in the shouldersize. - beside this - it provides no other meaningful data.

 

 

 

 

Before we get into thinking and believing...

(I prefer to do such things only in a church ;).. - and I have to admit that I hardly can recall when it last happened.)

 

Have you made the tests I suggested?

- I made them, and I had no change of any 'jitter' (if there at all) that is new or increased or wasn't there before the shulderzise was introduced.

 

 

Smooth as silk, this side, before hand

 

put up the "side by side', in real time, without any modifications...

you might "re-think" the Church comment though, as it leaves you wide open ;)

Edited September 14, 2013 by Wolf Rider

[_Dredd]

Have I missed something here, isn't the shoulder view the same as it was in 1.2.5?

 

I should add I'm not seeing this when looking back towards the shoulder view. I'm seeing this stutter when looking 15-20 degrees each side of the HUD, this at the moment is where it is most noticeable.

Edited September 14, 2013 by _Dredd

[zaelu]

I think this stutter is not a feature introduced with 1.2.6 but merely a bug. I have seen this stutter when turning head in many other games and it was always fixed later on.

 

The headtracking interface surely had some change and it was not for the best... that's why we the freetrack facetracknoir users had no tracking at start of 1.2.6 and had to do a mombo jombo with the headtracking.dll

[_Dredd]

I think this stutter is not a feature introduced with 1.2.6 but merely a bug. I have seen this stutter when turning head in many other games and it was always fixed later on.

 

The headtracking interface surely had some change and it was not for the best... that's why we the freetrack facetracknoir users had no tracking at start of 1.2.6 and had to do a mombo jombo with the headtracking.dll

 

Meaning a head tracking bug in the 1.2.6 code?

[Flagrum]

Before we get into thinking and believing...

(I prefer to do such things only in a church ;).. - and I have to admit that I hardly can recall when it last happened.)

 

Have you made the tests I suggested?

- I made them, and I had no change of any 'jitter' (if there at all) that is new or increased or wasn't there before the shulderzise was introduced.

 

Well, I try not to limit my thinking to church visits ...

 

Anyways, I did not do the tests you suggested - I was only trying to add an other approach vector to the discussion of the problem.

[PeterP]

Well, I try not to limit my thinking to church visits ...

 

sorry, this was not my intention to say exactly this - anyway- I "think" you got what I was talking about :).

 

I think this stutter is not a feature introduced with 1.2.6 but merely a bug. I have seen this stutter when turning head in many other games and it was always fixed later on.

By what it got fixed and which games?

 

The headtracking interface surely had some change and it was not for the best... that's why we the freetrack facetracknoir users had no tracking at start of 1.2.6 and had to do a mombo jombo with the headtracking.dll

 

Do you have a Trackir 4/5 running or any evidence to say that TrackIR isn't working as it should/can in 1.2.6 ?

[Wolf Rider]

How's the (as it comes off the roll, so to speak) side by side coming along?