3D Analysis of the Yemen Orb

[Mick West]

This is a more focused spin-off from https://www.metabunk.org/threads/uap-hearing-new-video-yemen-orb.14427/

The Yemen Orb video appears to show a fast-moving object (the UAP) colliding with another object (the missile). It's not immediately clear from the video what is happening.

@Zaine M. has done quite a bit of work on this, and has extracted a fairly high resolution set of data from the on-screen number. In particular we have

• The N indicator, which shows the North relative to the camera drone
• The Azimuth, which shows the camera heading relative to the front of the drone
• The slant range and horizontal range, which gives us:
  • The elevation angle
  • The altitude of the plane (when tracking the ocean)
  • The altitude of the object below the plane (and hence the altitude of the object)

Theoretically, given a drone's speed, this should let us calculate the drone's 3D track (likely fixed altitude and speed but variable heading) and then the target object's 3D track.

From Zaine's data I've extracted Az, El, and Heading. Az and El I can just drop into Sitrec, but I need to write some code to import the drone heading in a generic way (I had something similar for Gimbal, but that's custom code that does not work with the drag-and-drop sitch system.

Zaine has done some spreadsheet and 2D analysis with this data, and posted a video with preliminary results.

Source: https://x.com/ZaineMichael1/status/1967072017695600839

The video used (above) is the version posted on X by Eric Burlison. It's slightly lower quality than the congressional YouTube version, but I'm using it to match Zaine's frame numbers (which I've added). Zaine also did a stabilized version centered on the object, with matching frame numbers

I've been a bit busy with other things, but I hope to have the heading importer done soon, and that should give us more flexibility in examining this data and Zaine's conclusions.

 

[Zaine M.]

Just a note, those figures were derived from the livestream footage, and have been superseded.

I am using this copy of the footage as the basis for the current excel.


Source: https://www.youtube.com/watch?v=__PD0uTbuP0


The data in the attached excel is completed, as best as currently available, and the 2D recreation (to the right when opened), is a work in progress.

As I understand, this thread is about the data and recreations of that data, any items relating to what it is and evidence for that will be in the original thread.

 

[Kyle Ferriter]

Yes this thread should be kept narrowly on the topic of angles and trajectories. I lightly used vpython many years ago and spent a couple hours on Sunday trying to re-learn it in the hope that it could be useful but I kept getting really weird camera field of view stuff. I know a few coworkers in a prior job used ThreeJS and I imagine that is a more modern option, so I think I'll try that.

 

[RobertBillyBob]

Yes this thread should be kept narrowly on the topic of angles and trajectories. I lightly used vpython many years ago and spent a couple hours on Sunday trying to re-learn it in the hope that it could be useful but I kept getting really weird camera field of view stuff. I know a few coworkers in a prior job used ThreeJS and I imagine that is a more modern option, so I think I'll try that.

Another alternative is to setup Blender scenes from python scripts but it can be bothersome if you don't know Blender well enough.

 

[Mick West]

I've done a first pass of @Zaine M.'s numbers in to Sitrec
https://www.metabunk.org/sitrec/?cu...st-2.amazonaws.com/1/Yemen/20250917_202452.js

However, that resultant motion of the ocean is very unlike the video.

Zaine, this is using the number from the first spreadsheet, just camera Az and El, and Plane heading, with plane altitude fixed at 24,540, and the plane speed at 200 knots. No wind.

Would your new numbers change this at all?

 

[JMartJr]

However, that resultant motion of the ocean is very unlike the video.

Zaine, this is using the number from the first spreadsheet, just camera Az and El, and Plane heading, with plane altitude fixed at 24,540, and the plane speed at 200 knots. No wind.

Would your new numbers change this at all?

Would different but reasonable assumptions about wind help? Wind of zero would be somewhat unusual...

 

[Mick West]

Would different but reasonable assumptions about wind help? Wind of zero would be somewhat unusual...

No. The plane's heading has been calculated per frame, so it already factors in any wind. Since the heading does not change much, wind would not change the plane's speed much at all.

What we can change is the plane's speed, set to 200 knots here. a reasonable range would be around 150 to 280 knots. To make the target move down and to the left we'd need a slower speed. But even reducing it to 50 knots, we track is nothing like the video.

So my preliminary assessment is that these extracted numbers are some of the following:

1) Extracted incorrectly by Zaine
2) Used incorrectly by me
3) Too low fidelity to be useful

I suspect #3, but more checking is required.

 

[JMartJr]

No. The plane's heading has been calculated per frame, so it already factors in any wind. Since the heading does not change much, wind would not change the plane's speed much at all.

Understood, you are right. Though it may be worth keeping in the back of your head that wind can vary in direction and speed at different altitudes. Even there, the difference would likely be insignificant compared to the speed of the camera platform. (And if it was indeed a balloon, and not powered or tethered, it would not change speed after being hit, other than adding a "going down" component.)

 

[Zaine M.]

I apologise, I was going to update when i had completed the revision.

The N value DATA is inaccurate.

The value was extracted using this to calculate rotation


Source: https://www.youtube.com/watch?v=4gX0FR-j7IY


So the figures extracted were accurate [edit] - accurate enough generally and I accept full responsibility for this.

However when compared with the magnetic value on screen we can see significant degrees of difference.

By using the Magnetic values, it "tightens" the recreation and I will be back once I have completed the update to the figures.

 

[Mick West]

1) Extracted incorrectly by Zaine
2) Used incorrectly by me
3) Too low fidelity to be useful

I suspect #3, but more checking is required.

It seems like it's all of the above. I accidentally used a manual turn rate instead of the extracted heading. Fixing that improves things.

https://www.metabunk.org/sitrec/?cu...st-2.amazonaws.com/1/Yemen/20250917_232545.js

It does shows the results are VERY sensitive to small inaccuracies, this is always an issue with small noisy angle changes over large distances.

 

[Mick West]

Interestingly, in order to get the background to move at the same speed as the video over the first portion, I had to zoom in so far that the object size was about 80cm across (about 2.5 feet). This might be an artifact of the data.

 

[Mick West]

A much better fit can be got by simply placing a static object roughly in the intersection of the computed LOS.

https://www.metabunk.org/sitrec/?cu....com/1/Yemen Static Object/20250918_001256.js

 

[Kyle Ferriter]

We know the uncropped video exists somewhere and people have access to it. And AARO could get ahold of and release. That video certainly has the heading and altitude readings across the entire video, and probably has other information like camera coordinates. A lot of effort can be put in trying to estimate information that could be put together more accurately in a few minutes with the original video.

 

[InPhaseQuad]

View attachment 84096

Interestingly, in order to get the background to move at the same speed as the video over the first portion, I had to zoom in so far that the object size was about 80cm across (about 2.5 feet). This might be an artifact of the data.

Just for sanity check, rather than making the background fit visually I think you have a couple benchmarks:
at 10 seconds and 29 seconds there's visible sea-level horizontal range as the camera obtains and loses target lock respectively. 6.11nm @10s to 4.50nm @29s panning -13 degrees. With heading unchanged and using an approximate speed of 200knots = 1.05NM traveled, napkin math says about 0.98NM of total ocean pan distance in that period.

Here's the breakdown...

Given from telemetry
AB=6.11NM (At say t=0s immediately before target lock)
CD=4.50NM (At t=19s when target lock first lost)
∠A= -42
∠B= -55
Assuming Speed = 200Knots for all 19 seconds

Calculated:
AC=1.05NM (Distance traveled by drone)
AE=3.84NM, EB= 2.26 NM
CE=3.14NM, ED= 1.36 NM

Therefore, straight line camera panning distance across ocean surface, DB = 0.98NM.

 

[Zaine M.]

I have updated the excel to,
1. reflect the Magnetic bearing for the camera direction (not corrected for Yemen magnetic declination so raw),
2. updated the calculated altitude for the reaper above the object, and above sea level.
3. completed my recreation, still a work in progress.

*strongly note* that due to the loss of the N value, I have been unable to provide accurate camera and reaper heading data until frame 270. (we do see glimpses of a 103 Mag bearing for the camera, however testing with linear and polynomial fits, appeared to give erroneous results when plotting them out, perhaps someone else can attempt to reconstruct this data.
I agree that there is a, fairly consistent, line of sight convergence about 2k feet from the reported location of the object, but as to a better fit, I would be remise if I did not mention that the laser was active with its ranging during the encounter, but that is perhaps a discussion for twitter.

As seen in the image above, I have updated the excel for my version of the recreation, using data from frame 275 onwards, to not only calculate the object speed, preimpact, but to also calculate the object altitude during the encounter. The post impact figures are derived by, taking the distance of the reaper (at later frame location) from the impact location of the object, taking the corresponding elevation (when the cross hair is inline with the object, same with accurate Az frames), and projecting out that (top down) distance to resolve an altitude.

This was done, as I am unaware of any work done to determine the Field of view settings, so a "bush mechanic" solution was cobbled together to extract more information from the data. I am assuming if it were to descend, in very light winds, then its rate of decent would be uniform, Strongly Noting, it is only theoretical at this point and it does NOT include any potential drift modelling. So big grain of salt with that one.

Whilst my results, in the recreation, are still in the 30ish plus knot range for object velocity, I am seeing that 175 knots yields a smoother velocity graph, than the higher speeds like 200 knots. Acknowledging, the recreation is a work in progress, the noise in the data may account for this, its velocity by the second, its a 2D recreation that does NOT capture the motion a 3D recreation has etc. I just thought I would raise this now.

 

[Mick West]

Here I've added the FOV changes and made two versions, a Zaine Data (ZD) version (still using the first set of data, but I don't anticipate a huge difference with the new data), and a Static Object (SO) version
https://www.metabunk.org/sitrec/?cu...com/1/Yemen Zaine with FOV/20250918_165442.js

https://www.metabunk.org/sitrec/?cu....com/1/Yemen Static Object/20250918_165717.js

Note that BOTH versions use ZD for the plane track. We can also have a version with a fixed rate turn (29° over the course of the video), this looks similar, but the ZD heading gives better results as the drone travels in a straight line for a while. The constant rate turn flattens out too early.
https://www.metabunk.org/sitrec/?cu...SO with constatn rate turn/20250918_170240.js

Again, for a visual match on the background (required is the analysis is accurate), I had to zoom in to 0.1° vertical FOV, which is a lot.
And that makes the object small. If the oval is the actual shape, it's just 70cm (28") long and about 25cm (10") wide.

An AGM-114 Hellfire missile only has a 7-inch diameter and is about 64" long.

Also worth noting, in the ZD Heading versions, the shapes of the Az and El curves are quite similar, even though the full ZD has those curves coming from the Az/El data in Zaine's spreadsheet, but in the SO version, the shapes arise simply from the ZD heading curve and the position of the static target.

All this seems very consistent with a near-static small object in light wind that descends relatively slowly after being hit.

 

[Zaine M.]

I admire the speed at which you are processing the data, and there is one impactful point I want to address before coming back in a couple hours with updates and findings.

1. As I mentioned in post #9 "when compared with the magnetic value on screen we can see significant degrees of difference",

With the resultant path, using the N value, of the object being considerably "wide of the mark", when compared to the updated figures relying upon the Magnetic Bearing figure of the camera. Like you I changed from manual turn rate via a bank input, to using the heading figures straight into the calculations, but I will update that back in my next review.

For the avoidance of doubt, in the excel, columns A through Q are the raw data.

A. Frame number,
B. Pod Az displayed value, relative to boresight of reaper,
C. Target info from the footage for slant range,
D. Target info from the footage for horizontal (top down) separation,
E. Target info from the footage for Magnetic heading of the camera direction,
H. Plane heading as a function of, Pod Az plus Mag heading,
I. Elevation calculated from Reaper Alt and horizontal separation,
J. Elevation calculated from Slant Range and horizontal range,
K. Elevation calculated from Reaper Alt and slant range,
L. Elevation complete, based on the above methods, averaged where appropriate,
M. Plane altitude (NM) calculated via slant and horizontal values,
N. Plane altitude (NM) calculated via Elevation and horizontal values,
O. Plane altitude (NM) calculated via Elevation and slant range values,
P. Plane altitude (NM) complete, based on the above methods, averaged where appropriate,
Q. Plane altitude in feet


The heading is based on Az plus Mag values. (needs to be corrected with FOV deviation from target after it zooms out).

*I will take a look at the SITREC for the values you have derived for the FOV.

[edited - 1. to include column breakdown 2. changed points N and O from Reaper Alt, to Elevation as I mis-spoke]

 

[Mick West]

I had to zoom in to 0.1° vertical FOV, which is a lot.

But not inconsistent
https://www.rand.org/content/dam/ra...0/RR276/RAND_RR276.pdf?utm_source=chatgpt.com

Actually that 0.23 in IR, I had seen 0.1 somewhere....

 

[Zaine M.]

1. I agree with your assessment of the size and will come back with with substantive reason and value, instead of napkin math.
2. I do not disagree with your assessment of the FOV, on the basis that, as Kyle in post #13 highlights, we are seeing a cropped version and this version may only be displaying a significant fraction of the full FOV. Considerable work is required to squirrel down "what we see" and how it relates to full figures and vision.

 

[Mick West]

Aha, there's also 2x and 4x "Electronic" (digital) zoom.

Which would be 0.115° for the 2x Ultra Narrow. So that works. I had the three zoom levels as 0.1, 0.2, and 1.2. But now there's no 1.2°. Hmm.

 

[illiniflyer]

We know the uncropped video exists somewhere and people have access to it. And AARO could get ahold of and release. That video certainly has the heading and altitude readings across the entire video, and probably has other information like camera coordinates. A lot of effort can be put in trying to estimate information that could be put together more accurately in a few minutes with the original video.

That's why it's fun watching all of you do analysis with very little information. Imagine what you could do with the full context of the event and not just with what is on a poor quality video recording.

 

[Zaine M.]

I have the following results for size of the object, this is for recreation sizing, otherwise I would of posted this into the other thread.

The "right angle" dimensions, based on the width of the hellfire,

2.13 feet vertical
2.5 feet, rotated to "capture" the total length of the object

Noting that the pod has an elevation of 40.5 degrees, and it may be a perspective factor, I have the perspective lengths, based on the above figures, between,

2.63 and 3.85 feet horizontal,
3.08 and 3.29 feet vertically.

I raise the additional dimensions as, in the zoomed out part, there may be some issues arising when reconciling a *potentially* tumbling object that appears larger than earlier in the video suggests.

 

[Zaine M.]

For the Fov I get the following,

This is based on each of the smaller boxes being 20cm X 20 cm, the larger box is 2m X 2m.

To be more specific, noting the video is cropped, I have these values for the FOV of what can be seen,

Vertical FOV = 0.1357 degrees
Horizontal FOV = 0.2636 degrees

[Edit - updated to be more specific]

 

[Chris Lehto]

Hey all, Chris Lehto here. I appreciate your work and am also interested in finding a final solution for the Yemen UAP. Have you considered the Drone and Target Bullseye in your models? Although the data is sparse, it is ground-stabilized and is another parameter to help nail down the final answer.

In this image, top right 194/32, I believe, is the Drone Bullseye, which changes from 189/XX at the beginning of the video.
Bottom right 190/29 is the Target Bullseye, but remember this will be to the Object until impact, and then to the cursor position in the water after impact.

Hope that helps.
Best,
Chris

 

[JMartJr]

I have the following results for size of the object, this is for recreation sizing, otherwise I would of posted this into the other thread.

The "right angle" dimensions, based on the width of the hellfire,

2.13 feet vertical
2.5 feet, rotated to "capture" the total length of the object

Interesting, but are we seeing the missile, or the heat plume behind it? If the later, might glare make it appear wider than it actually is?

 

[boguesuser]

Interesting, but are we seeing the missile, or the heat plume behind it? If the later, might glare make it appear wider than it actually is?

The hellfire missile's rocket motor only burns for a few seconds so it's unlikely that it's the exhaust ploom.

https://www.dvidshub.net/video/8853...tation-mission-rehearsal-exercise-1-23-b-roll
At -2:19 (the timestamp runs backwards) they light off the motor after they bury it


Since the missle would be coming it at Mach 1, in order to appear so slow on the frame it would need to have a steep trajectory. How much would the size of the missle on screen change if it is actually going at Mach 1?

I'm still not convinced this is a hellfire.

 

[Mick West]

In this image, top right 194/32, I believe, is the Drone Bullseye, which changes from 189/XX at the beginning of the video.
Bottom right 190/29 is the Target Bullseye, but remember this will be to the Object until impact, and then to the cursor position in the water after impact.

Hi Chris. I didn't realize that's what it was.

There's a handy explanation of bullseye references here:
https://codex.uoaf.net/index.php/Bullseye

The BULLSEYE is a fixed reference point on the map that is known to all flights from which the position of an object can be referred to by bearing and range. A position is then referenced like a BRAA, but relative to the bullseye instead of any plane

So 19X/32 means it's 32NM SSW of the bullseye, moving from 198° to 194°.

The target goes from 190°, jumping to 189° when switching from object to ocean, and then tracking the ocean back to 190°

This all seems reasonably consistent with the Sitrec recreation, except right at the end the TGT range goes from 29 to 30, which is in the opposite direction. But at this point it's just the ocean.

These numbers are all very rough. 1° at 30NM is 0.5NM.

 

[InPhaseQuad]

Just wanted to insert a couple points about realistic wind based on some research, and resources for consideration here as I know there's been a lot of "wind"-based skepticism on other platforms since the initial analysis got put out

You can use the Beaufort scale for sea state to visually estimate surface wind speed. White caps as seen in this video indicate a beaufort number of no lower than 3 (7-10 knots) and no higher than 6 (22-27 knots). At category 4 which fits well, wind speed at sea-level can be 11-16 knots.
https://www.weather.gov/media/pqr/beaufort/beaufort.pdf

Second, at higher altitudes, wind speeds rise due to lower friction and density. It varies, but by about 5,000 feet we exit the atmospheric boundary layer and enter geostrophic territory, where pressure gradients and even Coriolis take over. As a rule of thumb, surface level wind speed over the ocean is about 70% of geostrophic speeds. A 15-knot surface (Beaufort number 4) wind is thus estimated to 20 knots exiting the ABL. Speed continues to rise with altitude. By 12,000 feet, it is pretty expected to have 2-3x surface wind speeds, and 20-40 degrees rotated in direction.

All this to say that it is totally reasonable to be considering the object moving 30+ knots with wind at that altitude, using the confirmable conditions in the video.

 

[Mick West]

Interesting, but are we seeing the missile, or the heat plume behind it? If the later, might glare make it appear wider than it actually is?

The hellfire missile's rocket motor only burns for a few seconds so it's unlikely that it's the exhaust ploom.

It will still be hot.

What we are seeing in the video is motion blur, not the actual shape of the object.

Look at two consecutive frames overlaid.

The waves are totally motion-blurred, and there's little to no gap between frames.

The missile overlaps, which I guess means there's some glare making the image bigger than it is

But going back to speed, perpendicular to the camera it's travelling about 2 feet in 1/30th of a second. so 60 feet/second, roughly 40 mph, or 18 m/s (very rough)

Also, very rough, the speed of the missile could be reduced to around 250 m/s or less by impact, so this suggests a pretty small angle between the line of sight and the missile trajectory. Possible the drone taking the video is the same one that fired the missile.

 

[boguesuser]

But going back to speed, perpendicular to the camera it's travelling about 2 feet in 1/30th of a second. so 60 feet/second, roughly 40 mph, or 18 m/s (very rough)

Also, very rough, the speed of the missile could be reduced to around 250 m/s or less by impact, so this suggests a pretty small angle between the line of sight and the missile trajectory. Possible the drone taking the video is the same one that fired the missile.

This is the part I take issue with. Wouldn't we notice a change in the apparent size of the missile when it enters the frame vs it leaving the frame?

Math is hard and these are very rough numbers so bear with me.

If the missile is traveling at 250m/s total, and 18m/s horizontal, then it should be traveling at about 249m/s vertically (relative to the frame). It's in frame for about 2 seconds which means it would travel a little under 500 meters within the frame or 6.75% of the distance to the object (if its 4NM away)

I was under the impression that a hellfire would not go subsonic. At the sound barrier, it would be traveling at 343m/s or 342m/s vertically (relative to the frame) which gives us a travel of 686 meters or 9.25% of the distance to the target.

And to prove my point, oh wait...
The missile does scale slightly. That's awkward.

 

[jdog]

This is the part I take issue with. Wouldn't we notice a change in the apparent size of the missile when it enters the frame vs it leaving the frame?

Math is hard and these are very rough numbers so bear with me.

If the missile is traveling at 250m/s total, and 18m/s horizontal, then it should be traveling at about 249m/s vertically (relative to the frame). It's in frame for about 2 seconds which means it would travel a little under 500 meters within the frame or 6.75% of the distance to the object (if its 4NM away)

I was under the impression that a hellfire would not go subsonic. At the sound barrier, it would be traveling at 343m/s or 342m/s vertically (relative to the frame) which gives us a travel of 686 meters or 9.25% of the distance to the target.

And to prove my point, oh wait...
The missile does scale slightly. That's awkward.
View attachment 84146

Is it worth considering that the missile's heat signature might be shrinking due to cooling effects if, as you noted above, the motor only burns for a few seconds after launch and then it glides?

 

[boguesuser]

Is it worth considering that the missile's heat signature might be shrinking due to cooling effects if, as you noted above, the motor only burns for a few seconds after launch and then it glides?

That is true. The bottom of the missle is dark after the impact.

And from what I can find, yes. It has a breif boost phase then glides into the target.

 

[Andreas]

The bottom of the missle is dark after the impact.

What do you mean by the 'bottom' of the missile? I've interpreted the almost black area below the heat signature as some kind of sharpening effect—but perhaps I'm wrong?

 

[boguesuser]

What do you mean by the 'bottom' of the missile? I've interpreted the almost black area below the heat signature as some kind of sharpening effect—but perhaps I'm wrong?

It could be but it only appears after the impact and seems to reduce as the hellfire pitches down.

 

[Andreas]

It could be but it only appears after the impact and seems to reduce as the hellfire pitches down.

To me, it just doesn't make sense that this would be part of the actual missile—especially since the black area appears instantly. The missile is only visible for about two seconds, and I can't see how there could be any significant changes in temperature during that time.

 

[Mick West]

But going back to speed, perpendicular to the camera it's travelling about 2 feet in 1/30th of a second. so 60 feet/second, roughly 40 mph, or 18 m/s (very rough)

Also, very rough, the speed of the missile could be reduced to around 250 m/s or less by impact, so this suggests a pretty small angle between the line of sight and the missile trajectory. Possible the drone taking the video is the same one that fired the missile.

Here's a video of sidewinder missile hitting an aerial target, filmed from the shooting plane. (originally posted by @Zaine M. on X)

We see a similar slow-looking curved path from this perspective. Of course this one explodes. Just after image there's another missile coming in at a wider angle, so it seems a lot faster.

 

[boguesuser]

To me, it just doesn't make sense that this would be part of the actual missile—especially since the black area appears instantly. The missile is only visible for about two seconds, and I can't see how there could be any significant changes in temperature during that time.

It could be that one of the aerodynamic panels ripped off during impact or it had a significant roll event during impact exposing a colder side. It could also just be an artifact. Hard to tell when it was recorded on the potatotron v0.2

Given this random and possibly untrustworthy source I found, the hellfire uses a tank of compressed air to power the control surfaces, so its not unreasonable to expect some parts to be colder than the rest.

Also, the anti tank Hellfires do use a crush fuse according to that document so if this is a balloon or similar insubstantial object, and the AA version uses the same fuze, that could be why there was no boom. I'm unsure if that's been mentioned yet.

Here's a video of sidewinder missile hitting an aerial target, filmed from the shooting plane. (originally posted by @Zaine M. on X)

Any idea what the target was?

 

[Zaine M.]

I apologise for the delay in updating this, I have been attempting to identify where I am getting a important variance.

1. As noted in post #9 I have found that using the N value provides for an apparent erroneous series of results when calculating the objects path.

2. In private discussion with someone, very knowledgeable, about the symbology we are seeing, they have an educated opinion about the data in the top right corner of the footage being specific to the reaper filming. its reference to bullseye (as Chris Lehto also described), the reaper altitude being fairly consistent with the derived altitude figures. Which is a very similar result to what the heading derived from N Value plus Az originally.

3. I have spent the last day and a half attempting to refine the footage, stabalising on the targeting reticule and reconcile these values noting,
a. that using the Magnetic value for the camera heading results in a smoother plot series,
b. that this smoother plot is more reflective of what the footage shows.

4. Whilst I will endeavour to keep "bashing away" at this, would anyone have an opinion as to a cause?

I will also attach the "new excel" with the additional data such as this heading figure, the observed elevation figure (middle of right-side), references to bullseye for reaper and target.

Reported Reaper details

The two red boxes are Azimuth figure and the reported heading figure for the reaper. This frame was chosen as both values, Magnetic value and reported reaper heading, change on frame 1474.

The blue box is the -38.5 degree result I get whilst using adobe (see image below) and the 39.5 degree value derived by adding reported heading and Az.

The orange box is the variances to Magnetic value of 52 degrees. Reportedly the Magnetic Value is also the direction the camera is pointing.

Many thanks for your time and consideration.

[updated to correct grammar and punctuation.]

 

[Zaine M.]

View attachment 84160

Here's a video of sidewinder missile hitting an aerial target, filmed from the shooting plane. (originally posted by @Zaine M. on X)

We see a similar slow-looking curved path from this perspective. Of course this one explodes. Just after image there's another missile coming in at a wider angle, so it seems a lot faster.

This was obtained from this YouTube video and should be cued up.


Source: https://youtu.be/JQ4RalNSjJk?t=181


Credit to "Wingnut" YouTube channel

 

[Zaine M.]

Please find below the best stabilization on the targeting reticule, I have been able to achieve.


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


As its a hand held device filming there is only so much I am able to achieve.