Some Refinements to the Gimbal Sim

[Mick West]

I have no dog in this race, it's just what I think is an objective statement of the facts, given what we see in the video, and also LBF and Zaine angle measurements. Can the rotating glare theory survive without its head (the 1st observable)?

The first observable is the decoupling of the horizon rotation and the glare rotation during banking, or, as I say in the video: "no rotation while banking." That seems to remain, as I said three years ago:

The initial slight continuous rotation does not change the fact when the jet banks, the horizon rotates, but the object does not.

And as you said

This part is true, it does not follow artificial horizon as you would expect from a well-defined object outline. I'm not saying the 1st observable came out of nowhere, there is a real aspect to it.

I'm quite happy with the position that the totality of the evidence I outlined in my original video shows that it's a rotating glare, even if there are some unanswered questions about some aspects of the rotation. I don't think they are significant to the overall argument. Frankly, I think you can tell it's a glare just by looking at it, with the motion, the background light pattern movement, and the diffraction spikes. The observables are just more measurable.

I'm open to being proven wrong. I see Marik is tweeting about me panicking as my life's work collapses around me, which is a bit odd. I largely lost interest in this years ago, as I think it went about as far as it could. It's fun to revisit from time to time, and I may at some point shift it all over to the more general Sitrec framework for some more experiments as the available tools improve.

But right now, I think if you want to declare it to not be a glare, you've got a significant challenge in explaining your determination to anyone who has not spent years looking at it. Zaine, I admire your persistence, but I think you keep getting things wrong in very complicated ways. I just don't think it's worth my spending many hours trying to unravel and confirm or rebut that when nobody else understands what it is I'd be confirming or rebutting.

I anticipate having more free time around April, maybe I'll revisit it then. But, as I said three years ago:

No rush though

 

[Zaine M.]

Zaine, I admire your persistence, but I think you keep getting things wrong in very complicated ways

I admire you Mick, I've said it privately and publicly, i think you make everyone work harder, the way its meant to be, but as for getting things wrong in a complicated way.

Just level the video on the artificial horizon, then take off the camera tilt due to plane pitch. Or at least let me know how we can distinguish between

a. this is background motion due to elevation changes or
b. the pod needs to have more rotation.

I didnt create the orientation for that horizon, its the result, and a result that doesnt need the complicated secondary rotation device, continual tweaking of an algorithm, more graphs and plots.

We can agree to disagree, but just like the thought experiment of a small 15 foot object exceeding the energy output of an F-18 in full afterburner from 300 feet away, I would of loved to see you on Rogan. "doesnt mean its aliens, DARPAs doing some crazy stuff."

 

[Mick West]

See my previous answer.

 

[Mendel]

Even if there were small variations in the stitched footage, the angular motion indicative of pod elevation change, like in go fast, is clearly there

Ok, first, some aviation terminology. An artificial horizon is a spinning gyroscope on the dash of an aircraft that is powered by compressed air. What we have in GIMBAL is an attitude indicator (AI). It indicates bank of the aircraft, and is only indirectly related to the actual horizon in the ATFLIR camera feed, since the ATFLIR is not looking straight ahead.

"Angular motion" means rotation. You are actually referring to linear motion, i.e. the cloud motion vector in the video is not aligned with the bank indicator of the HUD superimposed on the video.

There are two potential explanations for this:
a) the ATFLIR horizon angle does not match the bank angle
b) there is contiuous downward panning of the camera
We have evidence for a), but not for b).
We also know that the derotation device in the ATFLIR can rotate the imge any amount.

So, no, the cloud motion direction does not allow us to conclude that there is pod elevation change.

 

[Mendel]

One can say it's subjective or noisy but Zaine has measured the angles, it's somewhere in the thread, the objects rotates counterclockwise by a few degrees, versus the clouds by 8-10 degrees clockwise. Seems pretty clear to me.

I understand it's not clear to everyone why this matters. But it's a direct test for the hypothesis that the dero is responsible for clouds realignment in the first 20 seconds (cf 1st post of the thread). The object and clouds are supposed to rotate in concert, because dero without roll is supposed to rotate the whole image. Dero alone does not make a glare or other artifact to rotate one way, and the rest of the image the other way. Dero with pod roll yes, because pod roll changes orientation of the background versus glare, and dero rotating the whole image (glare+background) creates this decoupling between glare and the background. But here it's supposed to be without pod roll (because the pod is supposed to roll in steps, that's the basis of Mick's step-pod theory).

We can take it as given that the glare rotation is an in-camera effect, as we see it happening in the second part of the video as the camera rotates to avoid gimbal lock. So the question then is, why does the camera window/lens rotate with respect to the camera horizon/background when the pod is not rolling?
a) as the camera pans toward the front, the effect of pitch on the horizon shrinks
b) as the aircraft changes bank angle, the pitch angle also changes
These effects are minor compared to pod roll, though.

 

[Zaine M.]

Calm down, I am about to make life easier for you. It even has an option so you can account for camera tilt due to pitch.

Figures of bank, frustum roll, pitch for level flight, current pitch value to compensate for bank.

I have a second de-rotation box there for LBFs formula, but it is so specific to gimbal and subjective take of the glare angle and cloudline, i literally can not use it on any other videos (hint to me about its usefulness, personal opinion).

 

[Zaine M.]

"Angular motion" means rotation. You are actually referring to linear motion, i.e. the cloud motion vector in the video is not aligned with the bank indicator of the HUD superimposed on the video.

You know what I mean so we're good with me using angular motion

 

[TheCholla]

But right now, I think if you want to declare it to not be a glare, you've got a significant challenge in explaining your determination to anyone who has not spent years looking at it.

Well largely because of your videos and TV/media appearances, that present questionable analyses like the 1st observable, and completely ignore the context of the incident. Like for everything, it takes 10 min to put some wrong idea in people's mind with incomplete analyses, much more to deconstruct it and explain what the data really shows.

The first observable is the decoupling of the horizon rotation and the glare rotation during banking, or, as I say in the video: "no rotation while banking." That seems to remain

Except when it doesn't.

There are two potential explanations for this:
a) the ATFLIR horizon angle does not match the bank angle
b) there is contiuous downward panning of the camera
We have evidence for a), but not for b).
We also know that the derotation device in the ATFLIR can rotate the imge any amount.

So, no, the cloud motion direction does not allow us to conclude that there is pod elevation change.

There are three explanations for the mismatch between the clouds and artificial horizon (or attitude indicator if you prefer, but people have called it artificial horizon for years in the discussions).

a) the ATFLIR horizon angle does not match the bank angle -> it does at Az=0 when there is no ambiguity on how the horizons should align. so unless it derailed between the beginning and end of Gimbal, it indicates bank.

b) derotation is responsible for misalignment between real and artificial horizons-> then the object is not a glare, or we would see at least a hint of progressive CW rotation with the clouds

c) the camera shows accurate bank and there is no added dero component, the misalignement is inherent to the scene and how it is filmed -> assuming that the cloud line is a perfect indicator of elevation (almost constant at 2 degree) is wrong

a is unsupported, b goes against the rotating glare hypothesis, c goes against a distant aircraft. So there is no scenario here that has a rotating glare from a distant plane. Which means at the minimum any presentation of the rotating glare theory should mention the close path, which changes everything about the interpretation of the event.

a) as the camera pans toward the front, the effect of pitch on the horizon shrinks

As shown earlier, pitch decreases the slant in the clouds, removing it increases the mismatch. So unless pitch is negative this is not it. And I don't think it's completely off the hook as fighter jets can do maneuvers others planes cannot do. But negative pitch completely destroys Mick's Gimbal sim and the glare hypothesis as the pod would roll very differently.

Zaine, I admire your persistence, but I think you keep getting things wrong in very complicated ways.

I think the same about the ensemble of your Gimbal analyses. It's a textbook example of what not to do when investigating such a case, imo.

EDIT: typo

 

[Mendel]

There are three explanations for the mismatch between the clouds and artificial horizon (or attitude indicator if you prefer, but people have called it artificial horizon for years in the discussions).

a) the ATFLIR horizon angle does not match the bank angle -> it does at Az=0 when there is no ambiguity on how the horizons should align. so unless it derailed between the beginning and end of Gimbal, it indicates bank.

No.
When the camera and the aircraft are pointed in the same direction, they are banked the same.
But the more the camera heading is away from 0⁰, the more camera bank deviates from aircraft bank.
I don't think the HUD attitude indicator does not indicate camera bank. There's a small aircraft in it, not a camera.

b) derotation is responsible for misalignment between real and artificial horizons-> then the object is not a glare, or we would see at least a hint of progressive CW rotation with the clouds

Derotation does nothing to the HUD?
You said we do see that?

progressive clockwise rotation of the clouds in the leveled version of the video

--

c) the camera shows accurate bank and there is no added dero component, the misalignement is inherent to the scene and how it is filmed -> assuming that the cloud line is a perfect indicator of elevation (almost constant at 2 degree) is wrong

it's not nessary to add both negations to this alternative, and then obscure in your explanation what is actually happening geometrically

a is unsupported

no

, b goes against the rotating glare hypothesis,

no

c goes against a distant aircraft.

no

So there is no scenario here that has a rotating glare from a distant plane. Which means at the minimum any presentation of the rotating glare theory should mention the close path, which changes everything about the interpretation of the event.

I cannot fathom how you end up at rotating glare when my comment was aimed solely at the cloud motion vector.
It's the opposite of clear thinking.

As shown earlier, pitch decreases the slant in the clouds, removing it increases the mismatch. So unless pitch is negative this is not it.

too many pronouns, this is incomprehensible and unsupported. and completely fails to address my comment, despite purporting to be a reply to it

 

[TheCholla]

@Mendel sorry but you're truly lost about the analyses and you just add some noise here. I think Mick and LBF understand what we discuss, you don't.

 

[TheCholla]

It's interesting you reply on X but not here @Mick West

Source: https://x.com/MickWest/status/2001397932877291954


What misunderstanding? It's based on two very basic things:
- glare should rotate with frame edges in a leveled video (because glare is said to be fixed in the original video)
- the whole image is derotated by the derotation mechanism (including glare)

Which one is false? Super easy and short to answer, faster than replying this on X even, so no need to wait for April.

 

[jarlrmai]

Maybe you can demonstrate what you mean by videoing a glare with rotation of the camera and then apply a de rotate to it and see what it does to the glare that might help us understand what you mean

 

[TheCholla]

Maybe you can demonstrate what you mean by videoing a glare with rotation of the camera and then apply a de rotate to it and see what it does to the glare that might help us understand what you mean

It would rotate. Mick has made many examples of the mechanism.
https://www.metabunk.org/threads/th...t-lights-or-ir-heat-sources.10596/post-229959

In Gimbal it's the opposite, the glare is supposed to be fixed over the first 20sec because there is no roll.

Then if you stabilize on artificial horizon, it is supposed to rotate with the image frame (frame edges), an idea that was explained here a long time ago:
https://www.metabunk.org/threads/do...bunk-the-claim-that-the-object-rotates.12068/

So in the following stabilized video, it should rotate with frame edges:

Source: https://www.youtube.com/watch?v=-KVvebg4cXc&t=34s


It does a little bit but this is certainly not a 1° to 1° relationship, as we see in the close-ups.
See also LBF angle measurements in this post, it's not what you expect from a fixed glare:
https://www.metabunk.org/threads/some-refinements-to-the-gimbal-sim.12590/post-359311

First problem with the 1st observable.

The 2nd is that the derotation mechanism is proposed to explain slow clockwise rotation (realignement) of the clouds with artificial horizon. I'm not going to explain it further, it really is the topic of the thread. But because derotation rotates the whole image, including glare, the glare should follow the clouds realignment (slow CW rotation) , except when there is bank when it should rotate CCW with the frame edges.

If you combine the two effects, you expect this rotating motion of the object over the first 20 sec:

In other words the object should wobble with an amplitude of 3-4° each direction, being dragged CCW by bank, then CW by dero, in the stabilized video. Or wobble the same in the original video, but back and forth from its original position.

Because there is no discernible wobble, I'm pointing that there is a problem with the 1st observable, and that dero cannot be invoked to explain cloud realignement with artificial horizon.

This is a direct test for the glare hypothesis, based on two simple things:
- glare should rotate with frame edges in a leveled video (because glare is said to be fixed in the original video)
- the whole image is derotated by the derotation mechanism (including glare)

That are well-established principles of what a glare should do. These things have been discussed for years here so I do not see where the misunderstanding is.

 

[Mick West]

It's interesting you reply on X but not here @Mick West

Source: https://x.com/MickWest/status/2001397932877291954


What misunderstanding? It's based on two very basic things:
- glare should rotate with frame edges in a leveled video (because glare is said to be fixed in the original video)
- the whole image is derotated by the derotation mechanism (including glare)

Which one is false? Super easy and short to answer, faster than replying this on X even, so no need to wait for April.

Neither of those are false. What I think is wrong here is the degree of physical rotation you think is needed. Your graph there seems to be related to an expanded inverted version of the white line here.

https://www.metabunk.org/gimbal/

I've never been particularly happy with this since the start of this thread, but I've also never really been motivated enough to dive back in, as I know it will take many hours to figure out, and I've got better things to do. Hence April. Sorry.

 

[TheCholla]

What I think is wrong here is the degree of physical rotation you think is needed. Your graph there seems to be related to an expanded inverted version of the white line here.

You'll have to develop because I do not follow what you mean by "expanded inverted version of the white line here" and which degrees of physical rotation you are talking about. CCW? CW? Both?

 

[Zaine M.]

What I think is wrong here is the degree of physical rotation you think is needed.

I disagree, what is wrong is assuming that when presented with motion consistent with an elevation change, the response is "We have to rotate the camera".

Simply put, if it were glare, it would match the 11.5 degrees of camera rotation due to bank that you have. But that isnt what happened.

We have an allegation of "software/ hardware" extra rotation occurring, invalidating the artificial horizon, and the "actual" horizon. To compensate for a glare rotation that isnt there.

The way to test this, is to use both methods side by side and measure the results, which has been done on gofast as a control, and this extra rotation model, removed the background motion changes that we expect to see.

 

[Zaine M.]

Is LBFs formula in this camera software?


Source: https://www.youtube.com/watch?v=RQ401Jqr6is


The light patterns are just going crazy. Put another way, the light patterns are responsive to the orientation of the fire stick

OR

Its like Falch said, should be cued up.


Source: https://youtu.be/L9bBXgZBhsk?t=272


"It all comes down to rotation, is it a glare rotating or is the thermal signature rotating"

So to recap,
1. any standard way to level the footage, results in it not operating like glare,
2. It requires the addition of a secret sauce that is only applicable to gimbal, and fails at anything else.

But, for some reason its definitely glare and the horizon isn't showing angular motion, using elevation, its camera rotation?

How was it determined that its not elevation but camera rotation?

[edit grammar]

 

[Mick West]

You'll have to develop because I do not follow what you mean by "expanded inverted version of the white line here" and which degrees of physical rotation you are talking about. CCW? CW? Both?

Well I don't follow what your graph shows, so I guess we're even.

Seriously, I'm not interested in digging into this right now.

 

[Zaine M.]

Seriously, I'm not interested in digging into this right now.

I am, this needs to end

 

[Mendel]

@Mendel sorry but you're truly lost about the analyses and you just add some noise here. I think Mick and LBF understand what we discuss, you don't.

Yeah, I'm lost, and that's why I've been asking you guys to be more concise and more explicit, but you've ncot done that, and also ignored all of my clarification questions ("correct?"). In my experience, if you can't explain yourself to an interested audience, you don't have a good understanding of the matter, either.

 

[TheCholla]

Well I don't follow what your graph shows

I'm sure you do, it's glare 101.


Source: https://x.com/MickWest/status/1502110125682610177?s=20

 

[Mendel]

In Gimbal it's the opposite, the glare is supposed to be fixed over the first 20sec because there is no roll.

It doesn't require pod roll for the camera to rotate against the background (although pod roll will do it).
Whenever the camera rotates against the background, the glare shape also rotates, and vice versa.
The relation of the glare to the background is not affected by the derotation mechanism in the ATFLIR, that is a red herring.

I hope I understood correctly that the glare is rotating through a different angle than the clouds in the first 20+ seconds.

 

[TheCholla]

"Dero rotates the entire image" (Mick West, optics 101)

"The progressive realignment of the clouds is caused by dero realigning horizons" (Mick West, LBF, this thread).

What do we expect the object, glare, thing, to do when the clouds progressively realign?

I'll check back in April, you have 5 months.

 

[Zaine M.]

Yeah, I'm lost, and that's why I've been asking you guys to be more concise and more explicit,

There has been zero explanation or justification for the formula LBF is using except to keep "glare" in some particular orientation, from what i have gathered.

No one can say why,

when we remove, plane bank and camera tilt, that we all agree is in the footage,

how its determined that its not an elevation change? (pod elevation change of 0.37 degrees, stays with the -2 degree region, between -1.5 and -2.49)

But is "clearly" an additional camera rotation system in use, that has had to have many tweaks and refinements, I think LBF is still making tweaks according to this thread.

 

[TheCholla]

It doesn't require pod roll for the camera to rotate against the background (although pod roll will do it).
Whenever the camera rotates against the background, the glare shape also rotates, and vice versa.

Yes!! The camera rotates vs background due to bank (pod roll is fixed) -> background tilts in the image, not the glare. It's fixed.

Now, stabilize the video to remove bank, the glare rotates!

 

[Mick West]

There has been zero explanation or justification for the formula LBF is using except to keep "glare" in some particular orientation, from what i have gathered.

It's explained in the first post in this thread. Three years ago. Did you not read the explanation?

 

[Zaine M.]

The first issue is discussed here:
https://www.metabunk.org/threads/gimbal-derotated-video-using-clouds-as-the-horizon.12552/
The short version is that by measuring the direction the clouds moved in (parallel to the horizon), it was clear that this was at a shallower angle than the artificial horizon, most so when looking left.

The reason for this is that the horizon angle in the ATFLIR image should match what the pilot sees when they look in the same direction. I was previously assuming the horizon angle would be the same as the jet's bank angle (i.e. the same as the artificial horizon). But that's only true when looking forward. When looking more to the side then the bank angle contribution is diminished, and you get some more of the pitch angle. The precise math for calculating the desired horizon angle is found here: https://www.metabunk.org/threads/gi...using-clouds-as-the-horizon.12552/post-276183 and explained:

1. Yes, camera tilt plays a part, and we can calculate that.
2. This results in the clouds still not going through level.
3. ???? (maybe Raytheon is using an additional rotation device)
4. Logic bear has some code that makes the clouds go through level "parallel to the horizon"

That thread only talks about rotating the footage, nothing to do with the angle the clouds move through as a function of elevation.


Source: https://x.com/MvonRen/status/2001403816202383689


Hence, it has only recently been bought up, and "has been ruled out???

We all agree those factors are impacting the footage (bank/ camera tilt)" but, allegedly, its "definitely a camera rotation feature from the software/ hardware"

Just let me know the reason for that determination, please.

 

[Zaine M.]

That is pod elevation change, image above.



Somehow thats NOT pod elevation change but camera rotation needed, image above.



This is pod elevation change, even though the elevation figure stays at -27, image above.

How can you tell the difference between

Pod elevation change, and

Camera needs rotating?

 

[Mick West]

I wish you well in your investigation. I don't think it will change anything substantial, but who knows. Let's regroup next year. Happy holidays.

@Zaine M. , you need to explain things better. I suggest you go back and forth with @TheCholla in private to make sure you are on the same page. Start with the panoramas.

 

[Zaine M.]

Wish you well also. Merry Christmas, happy holidays to you and your family

[edit to discuss in private]

 

[Mendel]

Yes!! The camera rotates vs background due to bank (pod roll is fixed) -> background tilts in the image, not the glare. It's fixed.

Now, stabilize the video to remove bank, the glare rotates!

Yes. Because the camera also rotates sideways, which affects the horizon.

 

[TheCholla]

And how far away do you think those clouds are? In Sitrec I have them starting about 70 miles away, with the cloud horizon (the blue line) calculated at about 120 miles.

I also have them as a wide fairly flat layer

View attachment 87050

I want to be sure I understand correctly: the closest clouds that we see in the recreation are at 70 miles, and the furthest ones (cloud-sky line) are at 120 miles?

 

[Mick West]

I want to be sure I understand correctly: the closest clouds that we see in the recreation are at 70 miles, and the furthest ones (cloud-sky line) are at 120 miles?

I've changed it so the cloud layer comes all the way to the camera. Individual clouds are quite large and are offset by +/-500 feet from an MSL altitude. You can see this better if you increase ambien intensity.

The dark blue line (sorry it's a bit thick, experimenting with some things) is the calculated horizon of the top of the clouds (i.e. a standard horizon calculation for Earth radius plus cloud altitude, viewed from the lookCamera - the jet). The clouds move with wind.

Their altitude (which i think is the altitude of the average TOP of the clouds) 11740 feet, and configurable in Effects/Jet Tweaks.

https://www.metabunk.org/sitrec/?mo...amazonaws.com/1/gimbal_mod/20251219_165716.js

Side view. The elevation angle does not change the cloud horizon distance, just if it's visible in the view.

You seem to be suggesting a much closer, oddly slanted set of clouds. I encourage you to model that.

 

[Mick West]

Another missing factor is here refraction, which could change the relative apparent vertical positions of relatively near and very far object by a small amount by the horizon. I think this was discussed a few years ago, and it wasn't huge.

 

[TheCholla]

Ok thanks. My question was more: what's the distance between the closest clouds and the furthest clouds seen in the FOV, in this configuration?

 

[Mick West]

Ok thanks. My question was more: what's the distance between the closest clouds and the furthest clouds seen in the FOV, in this configuration?

Hard to say, a few miles?

 

[TheCholla]

You seem to be suggesting a much closer, oddly slanted set of clouds. I encourage you to model that.

There is an arc, or rotational motion, in the clouds in the sim, that is absent or less pronounced in the video where the clouds cross the FOV at an angle but in a straighter way. I had noticed that in Eddie Current recreation too. That's noticeable at the beginning of the video.

The only way I can reproduce that straighter motion in your sim is with a decrease in elevation angle that "pulls" the clouds up and straighten the arc. But then in your sim it makes the cloud-sky line go up and be visible behind the object, unlike the video. It could be due to two things, as far as I can tell now:
- in the video the cloud layer is not infinite and more complex that just infinite uniformly flat clouds
- there is some IR attenuation in the video and we do not see as far as in your sim, in which clouds at 70 or 120 miles are seen the same, in terms of brightness and definition. Hence my question.

But the straighter versus arched motion of the clouds is another indicator for El change, and it aligns with @Zaine M. 's stitch.

 

[TheCholla]

And while I'm at it beforeI forget (April is far!), your sim confirms that the change in El angle estimated by @Zaine M. completely absorbs the altitude increase for the close path.

I just had to change rise in El angle to -0.4 (Zaine's estimate), start from -1.6 instead of -2, and very slightly tweak heading of the object more like I have it in our best-guess scenario, to find zero altitude increase.

Saved setup is here : https://www.metabunk.org/sitrec/?mo...onaws.com/1/gimbalnear_mod/20251218_175247.js

Note the air track in blue is just a slight left turn at slow speed after deceleration, that turns into an abrupt stop/reverse on radar (SA page, ground track) with the effect of the strong westerly wind facing the object.

 

[Mick West]

And while I'm at it beforeI forget (April is far!), your sim confirms that the change in El angle estimated by @Zaine M. completely absorbs the altitude increase for the close path.

Now you've just got to get the clouds right.

I might get back into it sooner. April's just an estimate based on my life right now.

 

[TheCholla]

To get them right we'd need to know how they were first. They aren't right in your default config, in terms of the arching motion especially, but not only.

The close path described on the range fouler is here, does what Graves said they saw on SA (slow mover, turns on a dime), and it's not even breaking physics, just having some interesting aerodynamics. It's flat when accounting for cloud slant/estimated change in elevation angle (another coincidence?). It rises a little at the end as shown in our Fig. 9 and seen in @Zaine M. 's stitch, when overtaken by the wind.

Anecdotal evidence but the aircrew reported that it wobbled after the cutout, which seems very inconsistent with the rotating glare. That could be hallucinating or lying but until now, what they said in the range fouler report and through Graves could be verified, so I'd be cautious about making that assumption.

Happy Holidays!