1. Mick West

    Mick West Administrator Staff Member

    Not really. It looks like a bright IR flare you'd expect from any engine type heat source. What exactly are you comparing it against?
  2. Robert Mcintyre

    Robert Mcintyre New Member

    At the risk of being off topic I don’t think the FLIR footage is that interesting. But the fact that a leader of a navy fighter squadron described engaging with this object and tried to do a combat manoeuvre to get round it and it countered him is pretty hard to explain conventionally. They have to report all the time and he reported this straight away just like he would an encounter with, e.g a Chinese plane. But that’s eyewitness testimony of course.
  3. deirdre

    deirdre Moderator Staff Member

    well, just to clarify.. Fravor did not engage with "THIS" object. (FLIR footage). This footage came from youtube a few years back. There really is nothing to (at all) to verify this footage is even from that day.

    Fravor says he engaged with "an object", not "this object" that doesn't read right, as Fravor does seem to push this footage is of what he saw. so..

    so... Fravor's story is about "an object", not necessarily "this object".
  4. Robert Mcintyre

    Robert Mcintyre New Member

    The jets were circling around the area not crossing over it though?
  5. Robert Mcintyre

    Robert Mcintyre New Member

    Well there's a claim of chain of custody, but obviously being "top secret" we haven't seen this
  6. Mick West

    Mick West Administrator Staff Member

    That makes no really difference. At 400+ feet per second a circle would be a few miles wide, and just a gentle curve instead of a straight line.
  7. Kaen

    Kaen Member

    I recently watched some youtube videos taken above sea from an F18 cockpit. It surprised me how wide and clear the view is and how calm the ocean looks. On a bright day in calm weather I think it is actually quite easy to spot an object, keep it in view while descending, and to assess its appearance from the cockpit.
  8. Kaen

    Kaen Member

    One of the questions in this case is whether the object in the FLIR1 video is the same object as seen earlier by Fravor and Slaight.

    The object does seem to have the ‘oblong pill shaped’ appearance as reported by Fravor and Slaight. It also doesn’t show any heat signatures you would expect from an engine.

    So I tried to extract some more information from the video, using an analysis of the object’s angular size over time.

    That analysis shows that the video is consistent with:

    > A closure speed of 400 kts
    > Towards a 40 feet long object
    > At a distance of 10 nautical miles (NM)

    Note that 400 kts True Air Speed (TAS) is consistent with the jet’s Calibrated Air Speed (CAS) of 250 kts shown in the ATFLIR display, and the 40 feet object size is consistent with the size estimates made by the pilots in an earlier sighting of the object.

    The caveat is that the consistency remains if you multiply these three values by the same factor. So 800 kts closure speed to an 80 feet object at 20 NM also works. And 200 kts closure speed to a 20 feet object at 5 NM works as well.

    But at least the values are not inconsistent with the object’s size estimate given by Fravor and Slaight.

    My calculations are based on the 20 second segment in the video where the ATFLIR is in TV mode and the zoom level is 2.

    I took a snapshot at 1 s intervals and calculated the angular size of the object for each snapshot.

    I calculated the object’s angular size by computing the ratio between the visual object size and the total FOV size for each snapshot (separately in x and y since the video is slightly asymmetrical). Knowing the actual angular FOV of the ATFLIR (0.35 degrees in both x and y at zoom level 2) I could calculate the actual angular size of the object for each snapshot by simply multiplying this ratio by 0.35 degrees.

    Below is a collection of snippets from the snapshots, with the time in seconds for each snippet (t = 0 is the time in the middle of the 20 s video fragment). The yellow lines are the angular sizes I calculated. There is quite some image noise, but because of the many (22) data points I could still extract some valid data (more below).


    Below are the values for the actual angular size of the object I calculated for each snapshot, based on the length of the yellow line compared to the total size of the display’s FOV.


    Now for the calculations:

    If a constant closure speed Vc is assumed for the whole 20 s video segment, the angular size α of the object as a function of time will be:

    α(t) = S/d(t), where S is the actual object size and d(t) is the distance to the object as a function of time.

    This means: 1/α(t) = d(t)/S

    Since d(t) = d(0) + Vc.t, where d(0) is the distance to the object in the middle of the 20 s video fragment, we can rewrite this as:

    1/α(t) = d(0)/S + (Vc/S).t

    Since d(0), S, and Vc are all constant, this essentially is the formula for a straight line.

    So under the assumption of a constant closure speed, the graph of 1/α against time will yield a straight line with a directional coefficient of Vc/S and a value of d(0)/S at the intersection with the y-axis.

    Therefore I plotted 1/α against time and made a linear regression (which looked very nice considering the image noise in the video). From that linear regression I calculated both the directional coefficient and the intersection value with the y-axis. The results are in the graph below.


    Then I put some values of Vc, S, and d(0) in a spreadsheet to find combinations that yield these same values.

    A good combination was Vc = 400 kts, S = 30 feet, and d = 10 Nautical Miles.

    Note that 30 feet is the size of the object image but the object itself is rotated with respect to the FOV plane. So S=30 feet means that the actual object length is approximately 40 feet (roughly 30/(0.5*sqrt(2)) when the object is rotated 45 degrees with respect to the FOV).

    Changing one of these values with a certain factor means that the other values must be changed with that same factor to get the same intersection with the Y-axis (d(0)/S) and the same directional coefficient (Vc/S) again.

    The table below gives an impression what that means and basically shows that the values remain in the same ball park:


    So this analysis shows that the 20 second video fragment in TV mode at zoom level 2 is consistent with:

    > A closure speed of 400 kts
    > Towards a 40 feet long object*
    > At a distance of 10 NM

    *30 feet apparent length because the object is rotated w.r.t. the FOV plane

    400 kts True Air Speed (TAS) is consistent with the jet’s Calibrated Air Speed (CAS) of 250 kts shown in the ATFLIR display, and the 40 feet object size is consistent with the size estimates made by the pilots in an earlier sighting of the object.
  9. Mick West

    Mick West Administrator Staff Member

    That's not the issue here. The issue was the description of the object being at 1000-3000 feet and "crossing" an area that was 80 feet wide.
  10. marrowmonkey

    marrowmonkey Member

    How do you mean they are consistent? (If I understand you correctly Vc is how quickly the jet is closing in on the object.)

    How did you come up with the angle 45° ?

    The angular size as a function of t is
    a(t) = 2 * arctan(S / (2 d(t))

    I believe you should calculate the angle as
    a = 2 * arctan( g * tan(FOV/2))
    where g is the ratio between object lengt over total screen length, and FOV the field of view?

    2 * arctan( g * tan(FOV/2)) = 2 * arctan(S / (2 d(t)) => g * tan(FOV/2) = S / (2 d(t))
    => g (FOV/2) ~= S / (2 d(t)) [small angle approximation]
    => g*FOV ~= S/d(t)
    So I guess it works anyway.
    Last edited: Jun 14, 2018
  11. Robert Mcintyre

    Robert Mcintyre New Member

    Fravor has said the water disturbance was the size of a 737 (see video below @ 2m 35s) which has a gross area of 335 square feet (see http://www.b737.org.uk/techspecsoriginals.htm) and also that the object was going back and forth acoss it not just crossing it once. But yes that's not what "Source" said but isn't that person in the 2nd plane that remains at higher altitude and is also a less experienced pilot? Also, the Source account is 2nd hand reporting which I have seen being dismissed as unreliable earlier in this thread, whereas Fravor is 1st hand.

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

    The backgound music alone in the above video has convinced me of the veracity of his account
    Last edited: Jun 14, 2018
  12. Kaen

    Kaen Member

    I used this TAS calculator: http://indoavis.co.id/main/tas.html and filled in the altitude (20.000 feet) and CAS (250 kts) from the ATFLIR display. This yields a TAS of 385 kts.


    Yes, but allegedly the object was hovering when this video was recorded. If the closure speed is in the same ballpark as the jet's True Air Speed this is an indication that the object was indeed hovering. It does move slowly from right to left with respect to the jet but this can be due to a very slight banking maneuver by the jet (on the ATFLIR you can see the jet slightly tilting to the right with respect to the horizon).

    It’s just an approximation based on what I see on the ATFLIR screen, see figure below. The exact angle doesn’t change much to the end result since the whole exercise will only give you approximate values. That’s why I rounded the numbers for the object size to 30 and 40.

    Yes, the ATFLIR FOV is just 0,35 degrees in this case, so a ‘small angle approximation’ is allowed here.
  13. deirdre

    deirdre Moderator Staff Member

    what? what am I missing here? Fravor didn't take that video. At the very least, if we assume it was actually taken the same day, it was 3 or 4 hours later and the pilots who took the FLIR (according to Fravor) didn't see it visually.

    And since according to Fravor all the guys on the ship were making fun of him, isn't it more probable the next crew took FLIR footage of one of their own planes, zoomed in to get the 'fly off fast' effect and presented it to Fravor as a joke? We've never heard 'boo' from the pilots who took that footage.
    • Agree Agree x 1
  14. marrowmonkey

    marrowmonkey Member

    OK, but the closure speed would be the difference of the jet speed and the objects speed: Vc = Vobject - Vjet .

    TAS is only a measure of the jet speed relative to the air, if the object is moving the two are unrelated. Even if the object is stationary (relative to ground) TAS might differ significantly from the speed relative to ground. Only if the object is moving at the same speed as the air where the jet is flying this would be the same value (unlikely).

    Where did you read that? The video was recorded by a second pair of jets/pilots later, after Fravor and Slaight had returned to the ship:
    Source: https://coi.tothestarsacademy.com/nimitz-report/

    I don't see the 45° in the video at all. It could be any angle really. (Based on the examples of FLIR footage from other known cases it seems hard to discern even the basic shape of an object based on only the FLIR footage.)

    Given that it could be rotated any angle relative to the jet, any value for S (the apparent size) between say 5 and 50 feet could be said to be consistent with the story, and that would mean any closure speed between -70 to -700 knot, and distance of 1.6 to 16 NM could be said to be consistent with the story as well. And that is even without considering the uncertainty in the calculated values!
  15. Robert Mcintyre

    Robert Mcintyre New Member

    Fravor is retired and talks about it, maybe the other pilots are still in the service. Joking when back on a ship is one thing but to start faking stuff using military technology to make fun of your squadron leader seems an unlikley level of indiscipline. They were also apparently debriefed by their intelligence officer at the time if the report that's been circulated is genuine under the heading F/A18 F FLIR Tracking:

    https://media.lasvegasnow.com/nxsgl...TIVE REPORT_1526682843046_42960218_ver1.0.pdf
  16. Kaen

    Kaen Member

    Source: page 10 of https://media.lasvegasnow.com/nxsgl...TIVE REPORT_1526682843046_42960218_ver1.0.pdf

    Source: https://fightersweep.com/1460/x-files-edition/

    From where did you get the 3 or 4 hours?
  17. Kaen

    Kaen Member

    Why is that unlikely if they both reside in the same layer of air, at approximately the same altitude, at 10 NM distance from each other?

    The IR part of the footage gives a good impression of the object's rotation with respect to the FOV plane:


    If you measure 30 feet from left to right, the object cannot be smaller than 30 feet as can be seen in the figure below.

  18. marrowmonkey

    marrowmonkey Member

    That is a lot of "if"s to begin with. At the same time you say it was hovering, so then the airspeed relative to ground at that altitude would have had to be zero. And if you read the paragraphs below the one you quoted from the fightersweep article: "it was a few thousand feets below" them. I find it unlikely that the TAS would match the closure speed.

    I disagree. I can't see how the footage gives any indication of the object's rotation. Even if we assume it was roughly pill shaped it could have been rotated at almost any angle relative to the jet.

    The change in apparent size based on rotation also gives another source of error here; the object could have rotated slightly while being recorded. So the apparent change in object size might be because of changing rotation as well as changing distance.

    Well, yes, if the apparent size is X then the true size must be equal to or larger than X.

    It's a nice idea, but unfortunately the uncertainties means the result is consistent with most things they could have filmed, so it doesn't tell us much.

    Also noticed this snippet from the fightersweep article (my highlighting)
    The author of that article is also an ex fighter pilot, and apparently he thought the FLIR recording, in itself, was unremarkable. So again: the only evidence we have indicating something remarkable happened is Fravor's and Slaight's eyewitness accounts.
    Last edited: Jun 15, 2018
  19. Kaen

    Kaen Member

    Why? Do you have any data that shows that wind speeds at sea, at 20.000 feet, on a bright and sunny day, vary considerably over a distance of 10 NM and an altitude difference of 1 NM? Or do you just ‘find it unlikely’ that they don’t?

    To my eyes it is clearly rotated but maybe your eyes see something different.
    Maybe the composite picture from Ian’s combination of IR and TV helps:

    But I don’t care that much about the exact rotation. Even in the minimum size case, i.e. if the object is perfectly parallel to the FOV plane, it would still be 30 feet, which is still close to the pilot’s estimates (that vary from ’25 to 30 feet’ in the navy event log to ‘about the size of an F18’ – an F18 is 55 feet - in the recollections of Fravor).

    Yes, you can invent other narratives that are also consistent with the video, I already pointed that out in my analysis. But why would you assume that a self-invented narrative that contradicts the original narrative is closer to the truth?
    The original narrative tells that an object was approached that did not show any apparent movement, and this is not contradicted by the video. Only in case of clear contradictions I would be inclined to change the story.
  20. deirdre

    deirdre Moderator Staff Member

    I forget where I originally saw it. But it's also in the latest TTSA "document" released.

    https://media.lasvegasnow.com/nxsgl...TIVE REPORT_1526682843046_42960218_ver1.0.pdf


  21. deirdre

    deirdre Moderator Staff Member

    Thanks, yea I saw that unsourced document by TTSA, but your wording made it sound like the pilot said it was hovering.
  22. marrowmonkey

    marrowmonkey Member

    The closure speed is Vc = Vobject - Vjet

    You have calculated an estimate of Vc based on your measurements of the apparent size of the object seen in the FLIR video. But you don't know the speed of Vobject, nor of Vjet. The best we have is an estimate of true air speed (TAS) for the jet, but that is not the same as the speed over ground (it can vary at least as much as +/-200 knot). And we don't know what Vobject is at all.

    I.e. Vc is not the same as TAS of the jet.

    I really only see a fuzzy elongated (horizontally) blob. I don't understand how you get 45 degree from that, not even if you assume it is tic-tac shaped (and that blob looks too thin to be a tic tac btw). Yes, depending on the angle and the real size the apparent size could be 5-50 feet (smallest if it is pointing right at you, largest if it is pointing 90 degree to the line of sight).

    I.e. Fravors estimated the tic tac to be about X feet long, but the apparent size could be much smaller than that based on the angle. This is another large unknown.

    I try not to assume anything.

    The problem is that the result you found is consistent with both the story being true and not being true, so it doesn't really tell us anything. The uncertainties are so large it would be consistent with almost anything mundane.
    If you want to distinguish between e.g. three theories, a test that is consistent with all the theories doesn't help us single out which one is correct.

    I.e. this doesn't really tell us anything new because the uncertainties are so large.
  23. Mick West

    Mick West Administrator Staff Member

    You also have to account for the angle. V denotes a velocity - i.e. a speed in a specific direction in the air. The speed of the jet in the reference frame of the object and the closing speed of the jet towards the object are different things if the camera angle is non-zero.
  24. Kaen

    Kaen Member

    You already solved that one yourself:

    But you haven’t answered my question yet:

    Why is that unlikely? Do you have any data that shows the wind speeds at sea, at 20.000 feet, on a bright and sunny day, vary considerably over a distance of 10 NM and an altitude difference of 1 NM?

    If the apparent size is 30 feet, the object is at least 30 feet along its longest dimension, whatever its rotation w.r.t. the FOV plane might be. This is consistent with the estimations of the pilots.

    If several theories fit the same data, why not pick the one that is consistent with the narrative of the witnesses? They saw an airborne contact 25-50 feet in length which appeared to be capsule shaped, wingless, mobile, and white, with no visible markings and no glass. They later asked a different crew to see if they can get some ATFLIR footage at their CAP point, since this is where it was last seen on radar. The crew came back with footage of an object that is consistent in shape and size, and in its resilience against radar detection.

    This doesn't absolutely prove correctness, but it is an indication of correctness.
    Last edited: Jun 15, 2018
  25. Kaen

    Kaen Member

    We actually have the Vc angles on the ATFLIR display: 0-2 degrees horizontally and 5-6 degrees vertically.

    The effect (the cosine of these angles) is below 1%, so it can safely be neglected.

    Any other speed of the object w.r.t. the jet would be parallel to the FOV plane, but this component is only about 0.25 degrees per second from right to left and can be caused by the jet moving slightly from left to right (the wings are tilted slightly to the right).
  26. Kaen

    Kaen Member

    Yes, at 1200L, an earlier jet apparently came in that was asked at 1100L to investigate an unidentified airborne contact.
    This pilot of this jet only saw the disturbance in the water but picked up other traffic on his radar that was also investigating the area. He thought it was two F18’s.

    (Source: the link you provided)

    The assumption here is that these were the F18’s piloted by Fravor and Slaight, but the time does not fit so these must have been different jets:
    According to the Navy Event Log, Fravor and Slaight first spotted the disturbance in the water at 1430L


    (Source: http://www.nicap.org/reports2/2004_Navy event document 2004 Nov 14.pdf)

    1430L makes more sense, since the ‘ATFLIR flight’ met Fravor’s flight in the paraloft and then launched following Fravor’s flight. The paraloft is where you get your parachutes just before a flight and return them just after a flight. You don't wander around for hours on the ship with a parachute on your back. So Fravor must have just come in and the ‘ATFLIR flight’ was just about to leave if they met in the paraloft.

    (Source: the link you provided)

    The time of the launch of the ‘ATFLIR flight’ was approximately 1500L, which would also be the approximate time of return of Fravor’s flight if they first say the disturbance at 1430L:

    (Source: the link you provided)

    This means the time passed between the two flights investigating the CAP was probably less than 15 minutes, not 3-4 hours.

    Attached Files:

  27. Agent K

    Agent K Active Member

    As I said before
    Also, I posted that the blob looks like an airplane to me.


  28. Agent K

    Agent K Active Member

    Another mistake in the Executive Report: Figure 3 shows the TV mode, but its caption says "in 'Black Hot' mode."
    Also, the report doesn't say that Fravor recorded anything on a gun tape. It says, "CDR Fravor stated that the helmet's recording capability was rarely used therefore he did not think to use it that day."
  29. deirdre

    deirdre Moderator Staff Member

    hhmmm.. the full quote from that alleged "report" is
    so.. now we are to believe Fravor was aware he was looking for a supersonic Tic Tac? He never mentioned that before that I ever heard. Interesting.
  30. Robert Mcintyre

    Robert Mcintyre New Member

    Both flights were out at the same time as one detected the other on radar.
  31. Kaen

    Kaen Member

    It appears that way from the report, but the time at which Lt Col [..] actually reported to the Carrier Intelligence Centre is not mentioned explicitly. He may have reported to the Carrier Intelligence Centre hours later, after Fravor’s flight reported the tic-tac.

    Another possibility is that the two low flying jets saw something and reported it to the Carrier Intelligence Centre but Fravor was unaware of this. Note that these two other jets were both flying low, while during Fravor’s flight both jets came in high, only Fravor flew low and Slaight stayed high so these really must have been different jets.


    I don’t know when this report was made but there is a reference to the year 2007 at the top of page 4.
    I would assume the Navy Event log was made shortly after the incident and is more accurate in the times mentioned.

    In the meantime [Jeremy Corbell] managed to interview one of the radar operators. It’s hard to tell check his identity since he wants to remain anonymous, but for what it’s worth:

    He did the morning shift and kept on seeing anomalous objects on his radar screen during his entire shift. They were not treated very seriously (but apparently - from the report discussed above - some jets were already sent to the area to check it out).

    His shift ended around noon, but a couple of hours later he was invited to watch video footage. According to him it was different footage than the FLIR1 video, much clearer, and what he allegedly saw was quite spectacular.

    Make of that what you will, I don’t know what to think of it, but here it is FYI, the interview starts at 3:10 after a looooong intro:

    Source: https://www.youtube.com/watch?v=-VWVzaww1eU
    Last edited by a moderator: Jun 18, 2018
  32. marrowmonkey

    marrowmonkey Member

    Shouldn't you be the one providing evidence that your assumptions are valid?

    I don't think we should pick one. We have different possibilities and we could assign each a probability. When we get new data we could look at how that data affect the probability of each of the different possibilities and adjust them accordingly. That is hard to do in practice but it can help us reason about the value of new data. If we get new data that affect all the possibilities in the same way, that data won't affect any of the probabilities.

    Anyway, you made me look more carefully at the image:
    I think the object seen in the FLIR video looks too thin to be described as a tic tac. It would have to be more than twice as thick to get the same proportions.
  33. deirdre

    deirdre Moderator Staff Member

    well I take all these 'reports' and articles with a BIG grain of salt.. but the 1200L section sounds like exactly the same story of "Cheeks" in the fighter sweep article.
    and from this new 'document' that has no headers attached or dates.


    so... both accounts (pdf document and article) can't be true. The fighter sweeper article puts Cheeks and Dave together.

    and still we have zero proof this FLIR video was even taken the day of any of these witness statements.
  34. Agent K

    Agent K Active Member

    Well that was out of left field. A radar operator on the Nimitz, not the Princeton? We previously discussed a Reddit post that cited a radar analyst on the Nimitz, but the Executive Report doesn't list the Nimitz among the participants:
    The video he describes sounds like Fravor's UFO chase, but Fravor himself never mentioned that video. Fravor commented to CNN about the FLIR1 video: "When it leaves the screen on the video, it's moving pretty fast."
    • Agree Agree x 1
  35. Kaen

    Kaen Member

    I see no discrepancy between the fightersweep article and the pdf document. Both state that ‘Cheeks’ was advised to stay above 10.000 feet because there were other jets approaching at low altitude. I even suspect that the ‘official report’ may have used some information from the fightersweep article.

    The fightersweep article, written in 2015, states that Fravor’s flight started in the morning:

    Source: https://fightersweep.com/1460/x-files-edition/

    But the navy event log is very clear about the timing of the incident: at 1410L they were vectored to the location and at 1430L they saw the disturbance in the water.

    Source: http://www.nicap.org/reports2/2004_Navy event document 2004 Nov 14.pdf

    Attached Files:

  36. Kaen

    Kaen Member

    The radar anomalies that the Nimitz radar operator reported during his morning shift were basically ignored so no, the Nimitz did not seem to have played an active part in the intercept that afternoon.

    The radar operator was especially impressed about an upward motion, not a sideward motion of the object.

    According to the pilot report published on TTSA’s website, there was another video. Maybe not from Fravor's helmet but from a gun camera mounted on his F18:


    If such footage exists, Fravor might not be allowed to mention it. (The FLIR1 footage was leaked to the internet some years ago, there was never an 'official release' of this footage by the DOD to my knowledge, so he might as well talk about it now.)
  37. deirdre

    deirdre Moderator Staff Member

    you said
    exactly. but the alleged official report had to leave out Fravor's (the star of the "official" "report" show) flight time because they realized it doesn't match the alleged event log. Doesn't sound very "official" does it? They are rewriting the story as they go... which of course is exactly what an Entertainment Company would do.

    Which is why trying so hard to tie any testimony or 'reports' to this FLIR video (the thread topic) is just wishful thinking.
  38. Agent K

    Agent K Active Member

    What other camera could it be, the AN/AAR-55 NAVFLIR? It would have to be a wide enough field of view to see the UFO maneuvering. The radar operator wasn't even asked if it was a FLIR video. At the end of the interview, he said he was excited about the videos being released. I thought he'd be disappointed that the mind-blowing flying saucer video wasn't released, and instead we get lame videos that don't prove anything.
  39. Kaen

    Kaen Member

    Ian Goddard combined the IR and TV image in the video below. He speculated that the top part of the object is hit by the sun’s rays reflecting off the object. For the IR video segment this means that the top part is visible because it reflects the sun’s IR rays and is probably heated by the Sun, too. For the TV video segment this means that the top part is hardly visible against the bright background.

    Source: https://www.youtube.com/watch?v=8m4mq4S2pdQ

    So I enhanced the contrast/brightness of the TV part of the video a bit and went through if frame by frame. To my surprise you can just discern the top part of the object in some of these frames, and the shape that emerges matches a tic-tac very closely. It also matches the shape that emerged from Ian’s combination of TV and IR (the last two pictures).
    I used this video by the way: https://www.metabunk.org/attachments/f4-mpg.29900/


    I don't have any advanced video editing tools so I simply used the Microsoft media player. Maybe some of you can do a better analysis.
    Last edited: Jun 20, 2018
  40. Mick West

    Mick West Administrator Staff Member

    A bit of pareidolia there methinks. Suppose they had said it looks kind of dumbbell shaped, you could go through and find something that more matched that:
    Metabunk 2018-06-20 11-50-28.

    The shape clearly changes though the video, looking like it's turning to the left.

    Indistinct blobs look round. This is an indistinct blob and it's a huge reach to try to fit it to any particular shape.