Some Refinements to the Gimbal Sim

I understand what you are saying, but it comes from,

The clouds are in this particular orientation, and to keep them in this orientation, Raytheon has created a little known de-rotation system that adds in extra rotation.

as opposed to, the pod is working fine,

we remove bank, we remove the camera tilt, and this is the result of, angular motion meaning elevation change.

Because when i applied the two methods side by side, it resulted in background motion going in only one direction for what you are alleging to of occurred.


Source: https://www.youtube.com/watch?v=WFhi_Kq-WEk
 
Zaine M. said:
How did you determine that its not background angular motion and the camera needing rotation?
Click to expand...
If we take your claim at face value, then you've demonstrated that. I believe you're claiming that in order to get a CW rotation that is comparable to what we see then 'el' needs to change by more than we've assume but still only be tenths of a degree. I don't know if that's true, but that is what you're saying, right ? Well, IF that's true then (I'd still need to double check but I'm pretty sure) neither the pod horizon function nor the human horizon function are sensitive enough to 'el' during Gimbal for that to make a difference of several degrees during the first 20 seconds of Gimbal. So the CW rotation will still be accounted for, and if you try to account for it twice with your proposed changes to 'el', then of course that won't match the video.
 
logicbear said:
if you try to account for it twice with your proposed changes to 'el', then of course that won't match the video
Click to expand...

All im saying, and please check everything ive said, for example, attached is my camera tilt and plane bank rotation figures. Formulas are in there, I am not hiding anything.

What I am saying is that with angular motion i get these elevation figures

Screenshot (3841).png


based off this stitched together version

410.jpg


Large blue box is the FOV of 0.35 X 0.35
 

Attachments

Zaine M. said:
I understand what you are saying, but it comes from,

The clouds are in this particular orientation, and to keep them in this orientation
Click to expand...
You showed us that this is not true. Mick's Gimbal video, made prior to these refinements to account for some initial CW rotation, already included the principle that the dero should reproduce what would be seen outside the window. We just later discovered that implementing that principle more accurately also happens to reproduce the cloud angles during Gimbal more closely. The pod isn't broken, and you haven't shown that there is anything 'novel' about the very simple software controlled derotation method that is proposed.
Zaine M. said:
Because when i applied the two methods side by side
Click to expand...
You applied the SinCos function, but we've already explained a million times that it is not the method that is actually used in Sitrec. You keep sharing this video in which you've mislabeled / misattributed the function that you used as a comparison. Also you seem to be comparing the angle in which the background is moving with the angle of the horizon, but those two are not necessarily the same. I don't know by how much they're supposed to differ during GoFast.
 
Last edited:
Personally I think we would get a long a lot better in DM, refine the chats in there before posting publicly as I know I am guilty of talking past you, but I do have a lot of time for you LBF, I appreciate the time you have taken today, as well as the work you have put into this topic.

There are literally only a handful of people that "get" any of this, and I have said to @Mick West in private and I will say it publicly, I appreciate the time he takes and how approachable he is.

We don't agree on a lot of things, but no one can knock the volumes of work everyone, Cholla especially needs a mention in this, has put in.
 
logicbear said:
In the video Mick says "we want the horizon in the display to look like the horizon out the window", and that the dero rotates it to the correct angle.
Click to expand...

Mick wants that, but you don't know if this is what Raytheon and the military want. What's your basis for this assumption, other than it's what you want?

Having bank reflected in the horizon when looking on the side would not make a damn difference most of the time. C. Spitzer on X had done tests with DCS, and he didn't get horizontal horizon when banking and looking at 90° Az. You can say DCS is not the real thing, but as is Sitrec, which is a less advanced version of a fighter jet sim if you ask me. I remember that the situation is so unpractical (tracking an object with large bank and large Az) that he had a hard time to even recreate the situation.

logicbear said:
The current model assumes that the pod doesn't roll initially, but I also tried it without that assumption. It doesn't make any significant difference whether it is avoiding roll or not, whether el is off by tenths of a degree, you *still* get around the same initial degrees of CW rotation if you just assume that Raytheon didn't do a half assed job here, if they just implemented a dead simple function to get the horizon exactly right.
Click to expand...

The stabilized video below clearly shows the clouds realigning with artificial horizon, from an initial slant of about 10°. Your argument is that this is dero acting without pod roll and derotating the whole image (thing/glare+clouds). But the thing doesn't move, which shows it actually follows bank, while the clouds rotate CW. In other words, there is a decoupling between the thing and the clouds here, like at the end during the step rotation. Which would correspond to gradual CCW rotation of the pod while the dero is going the other way, at odd with the proposed step rotation mechanism, and which would have the dero going in the opposite direction to roll.

This is otherwise consistent with perspective change you expect from closing on a object with a distant background, seen from the side then from the front, as @Zaine M. has illustrated it in this thread.

Zaine M. said:
I can see three degrees of rotation, from the start to the first major rotation of target, but at this stage, I can NOT say that is definitive, i will have to work on a more stable version.


Source: https://www.youtube.com/watch?v=-KVvebg4cXc
Click to expand...


Source: https://www.youtube.com/watch?v=-KVvebg4cXc
 
Last edited:
Sorry to bother, just with @logicbear *appearing* to agree that there is background motion of the clouds, and not a camera rotation/ software thing.
logicbear said:
You showed us that this is not true.
Click to expand...

Can anyone tell me which version of the gimbal roll simulator is the one to use?

The first one or the second?

Mick West said:
There are two issues with the Gimbal simulation, which I think I've now resolved.

1 - The difference between the cloud horizon and the artificial horizon over the first 22 seconds
2 - The possible slight apparent continuous clockwise rotation of the object over the first 22 seconds

Both these issues only affect the first 22 seconds and not a lot. They don't make any difference to the four observable that demonstrate we are looking at a rotating glare, except in the sense that it's now both more accurate in terms of the physics and in the end results.

The original Gimbal sim: https://www.metabunk.org/gimbal1/
And the latest one https://www.metabunk.org/gimbal/
Click to expand...
I ask because the original version "gimbal1" doesn't have that software de-rotation that was mentioned and implemented in V2 "gimbal"

Many thanks for your time and consideration
 
And as a follow up, we have camera tilt and plane bank to account for, with only the de-rotation mirror in operation correct? Or is there *potentially* a software de-rotation to compensate while the pod isn't using roll (to remove camera tilt maybe, and if you could let me know what its doing, I would appreciate that)?

I ask because if there is that would cause a mis-match between software de-rotation, and the actual mirror de-rotation, and is the position that *maybe* the bumps are it "re-syncing" them both?

Again many thanks

[edited for clarity ]
 
TheCholla said:
which shows it actually follows bank, while the clouds rotate CW.
Click to expand...
Over the first 20 or so seconds the horizon rotates ~19 degrees, the jet banks 12 degrees, but the thing only rotates around 4-8 degrees. It's clearly not following bank. If it were a real object it'd have to be constantly rotating, throughout the entire video, at just the right rate to fool us into thinking that it's glare.
Zaine M. said:
I ask because the original version "gimbal1" doesn't have that software de-rotation that was mentioned and implemented in V2 "gimbal"
Click to expand...
It's a bit misleading to call it "software de-rotation". It's still most likely just a hardware device that optically de-rotates the image. It's just that some controller software might be issuing commands to that hardware device to make it rotate the image to a specific angle. The newer refinement introduced in this thread is of course the one that is used in Sitrec. It involves calculations that just seem like they would be easier to implement in software (with just a few calls to simple functions they likely already had available) instead of some electronic or mechanical controller solution, although Raytheon's ingenuity could surprise us. The prior version of the sim did not imply anything about whether it's software or hardware that's controlling the dero.

The ATFLIR manual mentions that while the dero itself has to be inside the EOSU, it appears to be controlled by electronics modules that are in the back behind the EOSU:
1729056611464.png


1729056534746.png

1729051637730.png

1729056455656.png

1729056793148.png

It doesn't say exactly how it controls the dero, but if you already have a software controller for the derotation angle, then you might as well make it as intuitive to the pilots as possible, even if the benefit in doing so were marginal.
 
Last edited:
logicbear said:
Over the first 20 or so seconds the horizon rotates ~19 degrees, the jet banks 12 degrees, but the thing only rotates around 4-8 degrees.
Click to expand...
is this based on my de-rotation method? Where are you getting this 19 degrees, 12 bank and 4-8 for the object/ target/ glare/ thing being filmed?

Can i get a link to that manual or can you post the explainer for 39 to 41 please

Just so i get it right in my head, no pod roll, bank compensates so there's no roll, so no mirror de-rotation?


[edited to be more specific]
 
Last edited:
Again, I apologise, I'm trying to work out, that there's now background angular motion, and how this would now relate to whats occurring in the gimbal roll sims, and any impact, those extra degrees of rotation that arent now needed all factor in to each other.
 
My understanding is the ATFLIR de-rotation algorithm is not simply trying to keep the horizon straight based purely on jet roll it's trying to provide an intuitive angle for the pilot to match the out the cockpit view in a way which is not disorientating so is potentially influenced by a number of factors including jet bank, direction of look, relative elevation etc

In some situations that might just be influenced by bank, but in most it might a combination of factors that combine to produce the eventual de-rotation amount
 
jarlrmai said:
ATFLIR de-rotation algorithm is not simply trying to keep the horizon straight
Click to expand...
I *believe* that we, Logic bear (LBF), cholla and I have seen that there is angular background motion, so horizon not level, * sorry, clouds not level, but horizon in de-rotated footage horizon is level*


logicbear said:
You showed us that this is not true.
Click to expand...

and im trying appreciative to get lbfs input, as the architect of the formula, on how that impacts what is occurring. I truly appreciate any and all feedback clearing this up, so everyone (me) isn't rushing to any conclusions.

[edited for accuracy, I am so used to typing lbf not logic bear]
 
Last edited:
Zaine M. said:
I *believe* that we, Logic bear (LBF), cholla and I have seen that there is angular background motion, so horizon not level, * sorry, clouds not level, but horizon in de-rotated footage horizon is level*
Click to expand...
I don't get the way you use language, I can't understand this.
which video, gimbal or gofast?
what background, clouds or sea?
"angular motion" means rotation?
are you saying the clouds in gimbal rotate?
I have never seen clouds rotate, how does that work? (actually I have, on satellite video time lapse, but it is very slow)
I have seen water spiral down the drain, but gofast looks nothing like that.

so what is it that you actually mean?

and what does "horizon in de-rotated footage horizon is level" mean? are you saying that the visual horizon in gimbal is parallel to the edge of the video?
 
*Adding- I am not sure how familiar @jarlrmai was with the remarks and i was being polite, not confrontational by spamming a bunch of previous posts and not trying to put words into other peoples mouths is all.

This is for the gimbal video, clouds have angular motion - they move through the FOV at an angle due to elevation change, no I'm not saying the clouds rotate.

Mendel said:
and what does "horizon in de-rotated footage horizon is level" mean? are you saying that the visual horizon in gimbal is parallel to the edge of the video?
Click to expand...

in the levelled video,

Source: https://www.youtube.com/watch?v=-KVvebg4cXc


the horizon is level, but the clouds do not move through level, they move through at an angle.


I hope this clears everything up


[edited for additional context]
 
Last edited:
Are you talking about the difference in the artificial horizon and the "real" horizon as displayed by the ATFLIR on the MFD view?
 
Zaine M. said:
*appearing* to agree that there is background motion of the clouds, and not a camera rotation/ software thing.
Click to expand...
No, sorry if that was a unclear. It seemed like you were once again implying some impropriety in the way this refinement was introduced. You previously used phrases like "working backwards" to describe that. I disagreed ("this is not true") and noted that something you recently brought up showed ("you showed us") why there was no such impropriety. It was not "working backwards", just "to keep [the clouds in some] orientation", because actually the latest refinement follows from the same principle that was already set out in the Gimbal analysis video.
 
logicbear said:
No, sorry if that was a unclear. It seemed like you were once again implying some impropriety in the way this refinement was introduced.
Click to expand...
I wasn't implying any impropriety, I thought you did agree that when we account for tilt, bank, the clouds do move through at an angle.


logicbear said:
"working backwards"
Click to expand...

Im not sure how else to put it, the clouds are in this orientation and this formula keeps them in that orientation, I had asked a few times about what the deciding factor was in angular background motion V camera needs rotation, I had read the thread and there was no mention of a function of elevation.

So I will ask directly, are the clouds level throughout or do we see angular motion. (my latest questions were based on how angular motion impacts it), I apologise if i misunderstood you. All I am doing is trying to get my head around what you are saying and being on the same page.

Zaine M. said:
I think we would get a long a lot better in DM, refine the chats in there before posting publicly as I know I am guilty of talking past you,
Click to expand...
I am well aware we dont talk to the same points, im just asking for clarity



jarlrmai said:
Are you talking about the difference in the artificial horizon and the "real" horizon as displayed by the ATFLIR on the MFD view?
Click to expand...
My initial remarks were primarily addressing the effect that clouds moving through at an angle has on the formula, software rotation/ de-rotation etc. As ive mentioned throughout the thread, I dont see the cloud line being the actual horizon
 
logicbear said:
Over the first 20 or so seconds the horizon rotates ~19 degrees, the jet banks 12 degrees, but the thing only rotates around 4-8 degrees. It's clearly not following bank. If it were a real object it'd have to be constantly rotating, throughout the entire video, at just the right rate to fool us into thinking that it's glare.
Click to expand...
Can you show us the section where dero without pod roll rotates both the clouds and the thing CW?

If thing=glare:
#1 Dero with roll-> thing rotates relative to the clouds (final step rotation with your step pod)
#2 Dero without roll (that you say happen in the 1st 20sec)->thing and clouds both rotate (by ~7° you say).

Where is #2 here?


Source: https://www.youtube.com/watch?v=-KVvebg4cXc
 
Zaine M. said:
in the levelled video,

Source: https://www.youtube.com/watch?v=-KVvebg4cXc


the horizon is level, but the clouds do not move through level, they move through at an angle.
Click to expand...
ok, so this has nothing to do with ATFLIR derotation?

You are comparing (a) the horizon indication on the bank angle indicator in the center of the screen, and (b) the direction of apparent motion of the clouds, caused by the camera panning and moving, correct? (the horizon indicator still tilted a bit in your video)

You have been advised that the bank angle indicator represents the angle of the aircraft with respect to the axis of the body of the aircraft. The camera pans from -54⁰L to 6⁰R, and we can see that the cloud motion starts to match the horizon indicator as the camera swings toward 0⁰. This suggests that the discrepancy is caused by the pitch-up attitude of the aircraft, which tilts the wing-mounted camera sideways, the more the camera turns to the side.

I'm not the first person to mention this, so clearly you think this explanation is insufficient, correct? Why?
 
Last edited:
Mendel said:
You have been advised that the bank angle indicator represents the angle of the aircraft with respect to the axis of the body of the aircraft. The camera pans from -54⁰L to 6⁰R, and we can see that the cloud motion starts to match the horizon indicator as the camera swings toward 0⁰. This suggests that the discrepancy is caused by the pitch-up attitude of the aircraft, which tilts the wing-mounted camera sideways, the more the camera turns to the side.

I'm not the first person to mention this, so clearly you think this explanation is insufficient, correct? Why?
Click to expand...

Pitch tilts the clouds the other way (unless pitch is negative).

See @Zaine M. 's post #73 , he details this step by step:
https://www.metabunk.org/threads/some-refinements-to-the-gimbal-sim.12590/post-358791

Also verifiable here: https://www.metabunk.org/gimbal/
 
What I am trying to nail down is this part
Screenshot (3945).png

Screenshot (3946).png


That is the first gimbal sim, the latest version has that angle change more severe and i dont see the glare/ thing we see orientate with those angle changes.

So that isnt horizon but the roll of the camera that the dero is *meant* to compensate for but isnt because the pod isnt rolling, this is the planes bank responsible for orientation and where this software de-rotation claim comes into it all.

So even though the glare stays fixed to that orientation, and we would see 14-19 degrees of rotation of the object, which we dont.

[edited for more context]
 
Last edited:
the V2 from gimbal pod roll sim
Screenshot (3947).png


Screenshot (3948).png



Now the object stays orientated to the horizon, not to that orientation. Which means its a real object and not glare
 
Last edited:
Mendel said:
This suggests that the discrepancy is caused by the pitch-up attitude of the aircraft,
Click to expand...
I have already corrected for that, which comes down to, the additional angle the clouds are at is due to,
1. angular motion of the clouds due to elevation (pod elevation, eg -2 degrees down to -2.37 degrees down)
2. A Raytheon software rotation device, that needs the clouds to be level with the horizon as a starting point

Which is why I have been asking


Zaine M. said:
Two questions, just to be clear,

1. How did you determine that the pod wasn't impacted by elevation changes (that would result in my above stitched together gimbal image)
2. How did you determine that the camera needed rotating, if you weren't using a clouds are perfectly levelled position to start with?
Click to expand...
[edited for additional context]
 
Mick West said:
"Using this, I have stitched the footage together,"
Click to expand...
By matching background features. Same as per Bellingcat with the mosul footage.

This post

https://www.metabunk.org/threads/some-refinements-to-the-gimbal-sim.12590/post-358786


Source: https://www.youtube.com/watch?app=desktop&v=PpJ6ppruH70

But even if the stitched together footage has slight variations, the background angular motion is visible in the above version, so theres pod elevation going down in value

In adobe terms, to match the features/ background angle of the clouds we need to move the frame left and then down, which results in why its curved.

[edited for additional example]
 
Last edited:
Just adding clarity to my naming convention,

Wide means wide angle of the FOV, correct as in correct sequence size.

When I put it together, i didnt know how large the frame would be, so i guessed and it was smaller than anticipated.


Source: https://www.youtube.com/watch?v=CktCA0U7wyU


As opposed to

"stitch tight" which is a tight view of the object, and not the fuller frame.


Source: https://www.youtube.com/watch?v=avzhx1BPIRw


and "stitch tight correct" is the same but the correct sequence size.


Source: https://www.youtube.com/watch?v=SPm60J23QIM


I provided the full frame version as I wasnt hiding anything
 
Zaine M. said:
No in premiere
Click to expand...
So I still don't understand what you are trying to say/show here. What determined the rotation of each frame in your panorama? If it's the direction of the motion of the clouds, then it should be flat. If it's the artificial horizon, then that should be fixed. But neither is true.
 
Mick West said:
What determined the rotation of each frame in your panorama?
Click to expand...
Plane bank and camera tilt due to planes pitch.

Zaine M. said:
1. Remove bank @ 1:1 (i say this because when we level on JUST the artificial horizon, any banking is cancelled out and we dont see any appreciable change in the cloud line, IE it would be derotated @ 1:1 not as per Sin or Cos of bank)
2. Account for planes pitch increase due to bank, over the standard 3.6 degree pitch for level flight, apply that value to determine how much camera tilt there is. (As we all agree the camera is parallel to the planes fuselage, and any pitch the plane has, is reflected in how much the horizon is offset from the planes artificial horizon).
3. Use those figures to derotate the footage, to put the footage into a global state where up is up for everyone.
Click to expand...
 
Mick West said:
If it's the direction of the motion of the clouds, then it should be flat
Click to expand...
If its flat then there needs to be some additional rotation, but leads to the larger questions,


Zaine M. said:
Two questions, just to be clear,

1. How did you determine that the pod wasn't impacted by elevation changes (that would result in my above stitched together gimbal image)
2. How did you determine that the camera needed rotating, if you weren't using a clouds are perfectly levelled position to start with?
Click to expand...
 
You must log in or register to reply here.

Similar threads

Zaine M.
GImbal Glare, Rotation, Clouds, and Angles
2 3
Replies
86
Views
3K
Mick West
Mick West
Mick West
3D Analysis of the Yemen Orb
2
Replies
75
Views
11K
Mendel
Mendel
A
Mick vs Marik (rotation glare gimbal)
Replies
19
Views
4K
jackfrostvc
jackfrostvc
J
Hank Green vs. Luis Elizondo, absent summaries, and the AI rabbit hole
Replies
15
Views
3K
Minus0
Minus0
TheCholla
Gimbal derotated video, using clouds as the horizon
2
Replies
57
Views
13K
logicbear
L

Latest posts

Trending content

Back
Top