NYT: GIMBAL Video of U.S. Navy Jet Encounter with Unknown Object

No, there is indeed a correlation. The reason you're not seeing it is because the magnitude and direction of the apparent movement aren't the same as the apparent rotation of the image field window, which is to be expected in a mechanism that uses the compounded movement of multiple axes to correct the orientation. But the noise in the sky, in particular, is perfectly in sync with the apparent magnitude and direction of the rotation of the blob. You can even discern the preferential direction of the glare distortion within the noise.
There’s definitely some streaking in the sky that rotates with the glare. Not obvious at first glance, but watch the videos a couple of times and you’ll see it.
 
There’s definitely some streaking in the sky that rotates with the glare. Not obvious at first glance, but watch the videos a couple of times and you’ll see it.
Very subtle, yes, and also present when the ATFLIR is not even close to the gimbal rotation point. Strong contrast enhancement required to even notice it. My guess is that this is very subtle flare in the optics caused by the object itself, which is a very bright IR source. Or do you really think the Navy would buy an ATFLIR that gives the impression of rotation each time you are on an enemy jet's tail? That would really be confusing...
Most of the apparent background rotation is caused by the jet banking by the way. It mostly banks in the opposite direction of the object's rotation, which enhances the illusion of background rotation in sync with the object.
 
Or do you really think the Navy would buy an ATFLIR that gives the impression of rotation each time you are on an enemy jet's tail? That would really be confusing...
Well they apparently buy FLIR cameras that give the impression of rotation when filming a burning vehicle on the ground. And yes, it seems to have confused a lot of people in the US Navy (apart, perhaps, from the person who saved the file as "Gimbal.wmv").

F-18-Takes-Out-Insurgents-Black-Hot-Flare-Rotating.gif
 
Very subtle, yes, and also present when the ATFLIR is not even close to the gimbal rotation point. Strong contrast enhancement required to even notice it.
You can very easily see it in the GIF you posted in #597. That's how I noticed it. It doesn't require strong contrast enhancement at all.

My guess is that this is very subtle flare in the optics caused by the object itself, which is a very bright IR source.
Or the more likely explanation: it's just ambient scattered light, bouncing around off the same imperfect rotating optical system that is creating the apparent rotation.

Or do you really think the Navy would buy an ATFLIR that gives the impression of rotation each time you are on an enemy jet's tail?
We have an existence proof to demonstrate this proposition.

Most of the apparent background rotation is caused by the jet banking by the way. It mostly banks in the opposite direction of the object's rotation, which enhances the illusion of background rotation in sync with the object.
The rotation you see is a result of the combined movement of the banking jet and the multi-axial rotations of the optical system. It's nonsensical to say that it's "mostly" the jet's movements, when the entire point of the optical system is to compensate for those movements and maintain a track on the object.
 
Well they apparently buy FLIR cameras that give the impression of rotation when filming a burning vehicle on the ground. And yes, it seems to have confused a lot of people in the US Navy (apart, perhaps, from the person who saved the file as "Gimbal.wmv").

F-18-Takes-Out-Insurgents-Black-Hot-Flare-Rotating.gif
Now THAT looks like glare 100%. What we see in the video looks much different.
 
The gimbal object is 2 deg below. It’s tracked from 54 degrees left to approx 5 degrees before the rotation occurs. Also the f18 is in a 20 deg bank.
How is the object tracked ?
If it is a glare caused by lens. How can it start so late ? We should still see artefacts. Even if coelostat mirrors internally do all the tracking initially. They are still moving in relation to outer lens. We would still see some artefacts even before the proposed outer gimbal pod rotation as they converge.
 
Well they apparently buy FLIR cameras that give the impression of rotation when filming a burning vehicle on the ground. And yes, it seems to have confused a lot of people in the US Navy (apart, perhaps, from the person who saved the file as "Gimbal.wmv").

F-18-Takes-Out-Insurgents-Black-Hot-Flare-Rotating.gif
That's not veiling glare, but lens flare. Do you see anything rotating in the scene itself?
https://en.m.wikipedia.org/wiki/Lens_flare
https://en.m.wikipedia.org/wiki/Veiling_glare

The claim is that the object in the gimbal video is a rotating veiling glare of a jet engine viewed from behind. I do think the object itself causes some very subtle veiling glare in the whole image, and this subtle veiling glare rotates with the object.

The ATFLIR image of a jet engine itself, however, is not veiling glare. It is a physical cloud of hot gas that would not rotate with the optics. It only looks like a veiling glare caused by poor quality optics. If the ATFLIR optics was this poor, every twin engine jet you are chasing would start to rotate left and right once you're on its tail. That would be unacceptable.
 
Last edited:
Take this jet for example:
Screenshot_2021-06-05-10-48-33-405.jpegScreenshot_2021-06-05-10-48-17-884.jpeg

If the white blob is a glare from the hot engine nozzles, it would change shape if the engine nozzles are obscured by an object in front of the scene.
If the white blob is a hot gas cloud, however, it would remain its shape under such conditions.
Now watch what happens:
Screenshot_2021-06-05-10-42-04-748.jpegScreenshot_2021-06-05-10-42-20-253.jpegScreenshot_2021-06-05-10-42-36-633.jpegScreenshot_2021-06-05-11-26-47-779.jpegScreenshot_2021-06-05-10-42-51-559.jpegScreenshot_2021-06-05-10-43-06-039.jpegScreenshot_2021-06-05-10-43-23-213.jpegScreenshot_2021-06-05-10-43-35-660.jpeg
 
All three videos seem to just use a passive track - meaning just based on the pixels in the video, not radar.

I watched some more footage of sim videos using SLAVE RADAR and optical tracking it seems that in normal use cases you SLAVE ATLFLIR to the L+S and then always turn on optical tracking as well, I think the SLAVE is mainly to get the ATLFIR pointing and moving in the right direction but the optical tracking is smoother than the RADAR tracking once engaged.

The general idea is to visually id targets that are beyond visual rage that you have RADAR track on, you can't easily point the ATFLIR manually so you point it the RADAR track zoom in and then optical tracking can take over for you. I wonder if leaving SLAVE on means it kicks back in and keeps the ATFLIR on target if you lose optical tracking, i.e. would have avoided the FLIR1/Nimitz situation.

In looking up the SNIPER POD used on the f/16 and finding it is not yet simulated. I realised how much we owe to the the sim makers, without them we'd have no idea of the basics of the ATFLIR. The other thing is looking through the forums that community is well aware of these videos. But the auto-track feature of ATLFIR was not simulated, I think it might be for the other pod (LIGHTNING), and now might also be in the very latest versions of the ATLFIR, but the tracking is a bit of a work around, i.e. there's no full IR simulation and contrast detection. What is also odd to me is the sim pilots and the sim makers seems to know how this stuff works and they know what features are missing from /"hacked" about the ATLFIR simulation. How do they get this information?

https://forums.eagle.ru/topic/266488-atflir-a-a-capability/

1623140976543.png
 
I watched some more footage of sim videos using SLAVE RADAR and optical tracking it seems that in normal use cases you SLAVE ATLFLIR to the L+S and then always turn on optical tracking as well, I think the SLAVE is mainly to get the ATLFIR pointing and moving in the right direction but the optical tracking is smoother than the RADAR tracking once engaged.

The general idea is to visually id targets that are beyond visual rage that you have RADAR track on, you can't easily point the ATFLIR manually so you point it the RADAR track zoom in and then optical tracking can take over for you. I wonder if leaving SLAVE on means it kicks back in and keeps the ATFLIR on target if you lose optical tracking, i.e. would have avoided the FLIR1/Nimitz situation.
Totally agree. This is also my conclusion on why the L+S is not boxed. The use the track to point and then switch to optical track.

In looking up the SNIPER POD used on the f/16 and finding it is not yet simulated. I realised how much we owe to the the sim makers, without them we'd have no idea of the basics of the ATFLIR. The other thing is looking through the forums that community is well aware of these videos. But the auto-track feature of ATLFIR was not simulated, I think it might be for the other pod (LIGHTNING), and now might also be in the very latest versions of the ATLFIR, but the tracking is a bit of a work around, i.e. there's no full IR simulation and contrast detection. What is also odd to me is the sim pilots and the sim makers seems to know how this stuff works and they know what features are missing from /"hacked" about the ATLFIR simulation. How do they get this information?

https://forums.eagle.ru/topic/266488-atflir-a-a-capability/

1623140976543.png
A lot of simmers are pilots/ex-pilots ;-) Also they simulate old versions (such as the 2000s circa era F-18C) so that more info is available on them publicly.
 
My question of how is it tracked is somewhat rhetorical. My point is, the atflir cannot span an area of sky 54 deg to 0 deg - miles away - without something moving or rotating in the process. A glare just suddenly rotating does not seem possibly and is not in line with any of the other glare footage.
I also have doubts that we are comparing the same phenomena. This example we keep using looks incredibly un- ir to me. Some footage can be a fused image which is a combination of standard light and IR image. This could be an example.
Though I’m not convinced it’s IR at all. If you watch the whole video it flips through modes and at this point with the glare, the people running about do not seem to be giving off a visible heat signature and seem to be clearly lit by the fire. This suggests tv mode to me; and the glare being a standard artefact and not IR.
 

Attachments

  • 9F20DEC0-DDE5-4ED3-8473-F809BDD7F557.jpeg
    9F20DEC0-DDE5-4ED3-8473-F809BDD7F557.jpeg
    367.6 KB · Views: 162
Totally agree. This is also my conclusion on why the L+S is not boxed. The use the track to point and then switch to optical track.


A lot of simmers are pilots/ex-pilots ;-) Also they simulate old versions (such as the 2000s circa era F-18C) so that more info is available on them publicly.
That's not quite what I am saying, you do both you SLAVE then use optical, but SLAVE remains boxed, unless you turn it off for some reason.
 
If the ATFLIR optics was this poor, every twin engine jet you are chasing would start to rotate left and right once you're on its tail. That would be unacceptable.
I don't think the optics are poor (actually, it is very high end optics with low stray light). The only optical surface which can never be perfect, is the outside window on the pod. This surface will suffer from all the wind and what not hitting it, causing... micro scratches. These scratches (or dirt/smudge) can even become directional, making the glare elongated because of diffraction effects.
 
I don't think the optics are poor (actually, it is very high end optics with low stray light). The only optical surface which can never be perfect, is the outside window on the pod. This surface will suffer from all the wind and what not hitting it, causing... micro scratches. These scratches (or dirt/smudge) can even become directional, making the glare elongated because of diffraction effects.

Agree that the optics is very high quality (even mirror based if I remember correctly, to prevent any lens artifacts). It's a bit more expensive than the consumer grade lenses used to demonstrate veiling glare in this threat...

Scratches on the window will not fly either, because:
1. This would mean the line-of-sight of the camera to the object happens to constantly go through exactly the same point of the window while the gimbal is rotating the window and the jet is moving in a banking curve. Given the geometry of the ATFLIR the probability of that is zero (note that the field-of-view is probably less than a degree, and the object is even smaller). Try this with your car if you like: put a small dot on the window screen and see if you can keep this dot in between you and an outside object while driving.
2. The camera is focused on the far field (the clouds are in focus). A glare from a stain in the window would be completely out of focus and just show up as a vague cloudy smudge instead of a well delineated one. Pretty much like the veiling glare caused by the object, which rotates with the object. That subtle veiling glare could be caused by small imperfections on the window. The image quality is not hindered very much by this because this veiling glare is completely out of focus.
 
Last edited:
Any optical imperfections on a close non focussing element will generally only show up as glares for bright light sources. The window over the camera here acts a bit like a "UV filter" on a normal camera lens, and having a dirty/scratched UV filter can lead to glaring for strong light sources.
 
1. This would mean the line-of-sight of the camera to the object happens to constantly go through exactly the same point of the window while the gimbal is rotating the window and the jet is moving in a banking curve. Given the geometry of the ATFLIR the probability of that is zero (note that the field-of-view is probably less than a degree, and the object is even smaller). Try this with your car if you like: put a small dot on the window screen and see if you can keep this dot in between you and an outside object while driving.
No. We're not talking about just one or two discreet scratches or imperfections. The situation is more like taking a piece of fine sandpaper and abrading the entire surface. But we don't even need the surface to be degraded that badly, or at all. All optical systems will be plagued by some degree of glare for bright sources. There's no way around it.

2. The camera is focused on the far field (the clouds are in focus). A glare from a stain in the window would be completely out of focus and just show up as a vague cloudy smudge instead of a well delineated one. Pretty much like the veiling glare caused by the object, which rotates with the object. That subtle veiling glare could be caused by small imperfections on the window. The image quality is not hindered very much by this because this veiling glare is completely out of focus.
No. There are multiple examples of "well delineated" glare from known sources in Mick's video, and other places. For example, this one, which we now know without any doubt was a jet airliner:
engine flares banked closeup overlay.jpeg

The glare in GIMBAL was further processed to sharpen it. So it looks exactly the way we would expect glare from the hot engines of a distant aircraft to look.
 
Last edited:
@Buckaroo,

I agree, I was about to write up a similar response, but as you already have penned it nicely, I will leave it there. :)
 
No. We're not talking about just one or two discreet scratches or imperfections. The situation is more like taking a piece of fine sandpaper and abrading the entire surface. But we don't even need the surface to be degraded that badly, or at all. All optical systems will be plagued by some degree of glare for bright sources. There's no way around it.


No. There are multiple examples of "well delineated" glare from known sources in Mick's video, and other places. For example, this one, which we now know without any doubt was a jet airliner:
engine flares banked closeup overlay.jpeg

The glare in GIMBAL was further processed to sharpen it. So it looks exactly the way we would expect glare from the hot engines of a distant aircraft to look.
Sorry, but no... These arguments don't hold.

The image above is a combination of a hot exhaust cloud and bokeh (the image being out of focus just a little bit). It has nothing to do with glare caused by scratches on a window. It's not even glare. You cannot simply fiddle with focus a bit (creating a bokeh effect) until you get a strange blob that vaguely resembles the gimbal object and then jump to the conclusion that a glare caused by a scratched window on the ATFLIR would be in focus. That doesn't make any sense at all.

Concerning your other argument: If you assume that the whole window is scratched, you'll never get such a stable shape across the window area. The glare would constantly change shape dramatically depending on the part of the window you're looking through. Remember that we're talking about a FOV of less than a degree here. I guess the object itself is about a tenth of a degree(!).

As an example: look at the window of your phone lens. It can be pretty smudged and scratched without causing much problems, because your phone will focus on the far field.
 
Last edited:
So, to summarize my arguments:

1. The hazy streaks in the background that rotate with the object are veiling glare caused by the object itself, not proof that the object rotates with the gimbal.

2. The blob you see when trailing a jet with the ATFLIR engaged is not a glare but a hot exhaust cloud (see previous post). Therefore it will not rotate with the optics.

Sorry, guys ...
 
As an example: look at the window of your phone lens. It can be pretty smudged and scratched without causing much problems, because your phone will focus on the far field.

Unless you introduce something glare-inducing into the frame, and then you'll notice the streaking from whatever's on the lens. In particular of the front surface was wiped by something that didn't actually clean the surface, and thus imparted some kind of orientation to the smear. As per the cleaning report above. I just opened my phonecam, pointed it at the window, and there was directional glare that rotated with the phone body exactly as I would have expected. It took all of 3 seconds. Did you even bother to attempt to disprove your own claim before making it with such certainty? Was 3 seconds too much effort?

All of the claims here by various debunkers are fitting remarkably well together and supporting a fairly precise narrative despite being composed from different inputs from different directions. It's almost as if there might be a reason all the physical evidence points towards one conclusion.
 
So, to summarize my arguments:

1. The hazy streaks in the background that rotate with the object are veiling glare caused by the object itself, not proof that the object rotates with the gimbal.

2. The blob you see when trailing a jet with the ATFLIR engaged is not a glare but a hot exhaust cloud (see previous post). Therefore it will not rotate with the optics.

Sorry, guys ...
False claims do not become more true through repetition.
 
Sorry, but no... These arguments don't hold.

The image above is a combination of a hot exhaust cloud and bokeh (the image being out of focus just a little bit). It has nothing to do with glare caused by scratches on a window. It's not even glare. You cannot simply fiddle with focus a bit (creating a bokeh effect) until you get a strange blob that vaguely resembles the gimbal object and then jump to the conclusion that a glare caused by a scratched window on the ATFLIR would be in focus. That doesn't make any sense at all.

Concerning your other argument: If you assume that the whole window is scratched, you'll never get such a stable shape across the window area. The glare would constantly change shape dramatically depending on the part of the window you're looking through. Remember that we're talking about a FOV of less than a degree here. I guess the object itself is about a tenth of a degree(!).

You do. Because the full surface is acting as a stray light source. And as all rays coming from the object fill the FULL aperture (say, lens/mirror diameter), the stray light will cover the full frame/image. When the object would not be hot/bright, this stray light would be hardly visible. But because the light is very bright, it causes this huge glare.

As an example: look at the window of your phone lens. It can be pretty smudged and scratched without causing much problems, because your phone will focus on the far field.

I just did it and I saw a nice directional glare (directional, as I wiped some grease on the lens of my iphone). And this rotates when I rotate my phone.
 
You do. Because the full surface is acting as a stray light source. And as all rays coming from the object fill the FULL aperture (say, lens/mirror diameter), the stray light will cover the full frame/image. When the object would not be hot/bright, this stray light would be hardly visible. But because the light is very bright, it causes this huge glare.
How much of the window an object's rays pass through depends on the position of the window in the optical system and how close it is to a pupil plane. You can't simply say that the full window is utilized by a single object; however, a defect on the window need not be *perfectly* centered on the chief ray of an object's line of sight, depending on the size of the optical footprint on the window.

For example, the iris in my SLR camera's lens is at a pupil and thus the reduction of the iris size impacts all field angles, whereas I could block part of the front of the first camera lens with my hand and only block part of the scene, though my hand would be blurry. A defect on that first lens could cause glare for multiple field positions but not necessarily all of them.
 
zoom-3gun-WMT.png
For example, I randomly grabbed this image off the internet, showing a zoom lens in three configurations. You can see that in the "wide" configuration, different field angles use different parts of that front lens. (glare from blue could be quite different than glare from red), whereas in the "tele" configuration, the different field angles overlap almost completely (glare will be consistent across the field).

Without a detailed raytrace of the optical system being discussed, I don't think we can automatically know the answer.
 
How much of the window an object's rays pass through depends on the position of the window in the optical system and how close it is to a pupil plane. You can't simply say that the full window is utilized by a single object; however, a defect on the window need not be *perfectly* centered on the chief ray of an object's line of sight, depending on the size of the optical footprint on the window.

For example, the iris in my SLR camera's lens is at a pupil and thus the reduction of the iris size impacts all field angles, whereas I could block part of the front of the first camera lens with my hand and only block part of the scene, though my hand would be blurry. A defect on that first lens could cause glare for multiple field positions but not necessarily all of them.

I agree, I just wanted to keep it all a bit more easy to grasp for the "non-optical" folks in this thread.
But yes indeed, it matters where the pupil stop is and what element is stray light affected and where it is wrt the pupil.
 
I agree, I just wanted to keep it all a bit more easy to grasp for the "non-optical" folks in this thread.
But yes indeed, it matters where the pupil stop is and what element is stray light affected and where it is wrt the pupil.
Sure, but we also want to avoid simplifying it so much we get the wrong answer. Some of this stuff is complicated and subtle effects can have a big impact, especially when considering rare instances. I’ve worked with optics for over 25 years (astronomical) and have found that not much of it is as intuitive as I would have guessed. And also the differences between astronomical instrumentation always focused on infinity and other forms of photography have tripped me up many times when thinking of other kinds of optical situations. So I could definitely be wrong about some of what I say here and would gladly be educated by those more experienced than I.
 
Sure, but we also want to avoid simplifying it so much we get the wrong answer. Some of this stuff is complicated and subtle effects can have a big impact, especially when considering rare instances. I’ve worked with optics for over 25 years (astronomical) and have found that not much of it is as intuitive as I would have guessed. And also the differences between astronomical instrumentation always focused on infinity and other forms of photography have tripped me up many times when thinking of other kinds of optical situations. So I could definitely be wrong about some of what I say here and would gladly be educated by those more experienced than I.
You're not wrong. I have been involved in many optical projects related to astronomy (but not only) for two decades (ground based, but later space). I found that there are many different types of optical engineers. Some are more specialised in laser optics, some others more in imaging optics, or perhaps more related to IR imaging. I see that no optical engineer is the same, I guess that is true for most engineering fields.

And indeed, I also like to learn if I was/am wrong. Perhaps sometimes we get fooled by our own "optical gut feeling", don't we?
 
You're not wrong. I have been involved in many optical projects related to astronomy (but not only) for two decades (ground based, but later space). I found that there are many different types of optical engineers. Some are more specialised in laser optics, some others more in imaging optics, or perhaps more related to IR imaging. I see that no optical engineer is the same, I guess that is true for most engineering fields.

And indeed, I also like to learn if I was/am wrong. Perhaps sometimes we get fooled by our own "optical gut feeling", don't we?
I am a scientist so I don’t have any formal training in optics. Just been dealing with it for a long time. So I have mostly learned only what I’ve needed to learn. I definitely feel that there’s no relying on my gut most of the time. It can require careful analysis most of the time.

what we are dealing with in these threads is often not enough information and too much speculation. Filling in the gaps of knowledge with assumed values or rules of thumb may provide *an* answer but it can be difficult to assess if it’s the *correct* answer.
 
Here's the Gimbal footage stabilised on the object, and then overlaid with a crosshair on a static part of the overlay.

I think that this helps to show how the entire image jolts around at the start of each rotation.

 
What you seem to be unaware of concerning the optics of the ATFLIR: the rotating parts only contain mirrors plus the front window. These elements do not integrate light over their surface area to create an image, only lenses do.

All lenses are situated in the fixed, non-rotating part of the ATFLIR, behind the de-rotation mechanism (which is mirror-based as well).

So any glare caused by an integration effect (a.k.a. lens glare) will materialize in the light path after the de-rotation device and hence will be independent of the gimbal rotation since that rotation has by then already been compensated by the de-rotation device

The only possible causes of glare that rotates with the gimbal are the mirrors and the front window in the rotating parts.
The mirrors are very high grade components in a sealed enclosure. They do not produce glare (mirrors in general do not cause any glare because they do not bend and integrate light over their surface like lenses do).
The front window has already been discussed and cannot cause such a consistent glare of a 0,1 degree object over a large area.

Any glare in the ATFLIR image will be produced in the camera, situated in the fixed part of the ATFLIR, looking at the de-rotated image and therefore completely unaffected by gimbal rotation.
 
The mirrors are very high grade components in a sealed enclosure. They do not produce glare (mirrors in general do not cause any glare because they do not bend and integrate light over their surface like lenses do).
that doesn't mean there's no source of disturbance there, see this post from a related thread:
fyi, The cross shaped spikes often seen in star images are caused by the spider holding the secondary mirror is some telescopes.

1623211811107.png


source https://stellafane.org/tm/dob/ota/spider-stalk.html
 
Here's the Gimbal footage stabilised on the object, and then overlaid with a crosshair on a static part of the overlay.

I think that this helps to show how the entire image jolts around at the start of each rotation.

If you look carefully frame by frame you'll see that the jolts are actually small LOS corrections of the ATFLIR to keep the object in its tracking window.

Just prior to each rotation, the object rises a bit. The ATFLIR immediately shifts its LOS to keep the object in its tracking window. These are the jolts you see. Right after that the object starts to rotate. I'll make some screen shots later to show you.
 
Back
Top