jarlrmai
Senior Member
No I mean in Go Fast the range is from RADAR using correlation with the ATLFLIR LOS and a RADAR track, this feature is not simulated DCS, DCS does not have a full simulation of the systems the F/18 has regarding sensor fusion.
GoFast : is the background moving relatively to the movement of the jet ? Or the camera only ?
- Thread starter dimebag2
- Start date
MclachlanM
Active Member
I'm not sure I follow here, how do we know that the range is from radar and not the same thing that DCS is showing?
It's just struck me that we are using different manuals as I can't find this, I have this https://www.mudspike.com/chucks-guides-dcs-f-a-18c-hornet/ . Do you mind giving me a link as I can't seem to find yours?
I did a little demonstration of what is confusing me about the GoFast footage:
First I lock the target on radar and you can see the range Vc etc. displayed as normal on the HUD.
Next I start trying to autolock which gives the TGT range given on the LDDI which doesn't seem to have any relation to the radar range, even after getting the autotrack.
I switch off the radar, the target information on the HUD is lost yet the LDDI keeps the TGT it has.
Losing the autotrack doesn't have an effect on TGT.
This hurts my head. The manual says:
And Also:
If we can get two 'target ranges' which are completely off at the same time then maybe what Chris says about the range in GoFast being wrong is correct?
I think you've also got to consider the possibility that DCS might have bugs.
MclachlanM
Active Member
This is true, but seeing as how it finds the range is a bit of a mystery and this bug coincidently supports Lehto's claim I think it's best to get to the bottom of this.
It seems that the notation TGT is only used when dealing with air-ground bombing. I can't get it in a A/A mode or find any examples of TGT appearing for an air target.
jarlrmai
Senior Member
In the manuals when RNG is used on displays it always seems to refer to a range to a RADAR track.
When TGT is used it refers to a range to a point on the ground or a nav point.
In Go Fast RNG appears when the target track is acquired as does VC the format is similar to the way RADAR targets are shown, it's possible these figures are 'donated' to the DDI by sensor fusion, ie correlation of the ATFLIR track with a RADAR track on the MSI computer. Or even that the ATFLIR track is driving the RADAR to designate that as the target.
It is also possible there is some sort of passive ranging but no one seems to know exactly how this would work.
These feature is not yet simulated in DCS F18 but they are mentioned in sims of other aircraft.
When TGT is used it refers to a range to a point on the ground or a nav point.
In Go Fast RNG appears when the target track is acquired as does VC the format is similar to the way RADAR targets are shown, it's possible these figures are 'donated' to the DDI by sensor fusion, ie correlation of the ATFLIR track with a RADAR track on the MSI computer. Or even that the ATFLIR track is driving the RADAR to designate that as the target.
It is also possible there is some sort of passive ranging but no one seems to know exactly how this would work.
These feature is not yet simulated in DCS F18 but they are mentioned in sims of other aircraft.
Last edited:
MclachlanM
Active Member
Ahhh, yeah you're spot on. I've just sat playing with the TGT and it is based on the coords of ground targets designated for bombing, very accurate for the ground, gives wrong numbers for planes. At some point I forgot that the original video said RNG instead of TGT
.You've cleared that up for me cheers. At least this might give some insight into where the differing pilot analysis is coming from (Lehto saying it's just wrong and Lemoine giving the weird trig explanation), simply a confusion between TGT and RNG? It doesn't help that the manuals refer to them both as target range despite being different in where they're used, how they're obtained and what distance they measure.
Yeah fair enough, perhaps I should say I'm no wind expert either
Anyway my main point was that the ocean speed would not be the main culprit responsible for this discrepancy, and I'm not convinced it can even be extracted as a separate variable from the wind.
jarlrmai
Senior Member
Ahhh, yeah you're spot on. I've just sat playing with the TGT and it is based on the coords of ground targets designated for bombing, very accurate for the ground, gives wrong numbers for planes. At some point I forgot that the original video said RNG instead of TGT
You've cleared that up for me cheers. At least this might give some insight into where the differing pilot analysis is coming from (Lehto saying it's just wrong and Lemoine giving the weird trig explanation), simply a confusion between TGT and RNG? It doesn't help that the manuals refer to them both as target range despite being different in where they're used, how they're obtained and what distance they measure.
Yeah I think you're possibly right Lehto not having used ATFLIR specifically making some assumptions about it being in AG mode and RNG being TGT. The possibility still remains that its angle rate ranging that I can find some references to but no full description of. One thing that is interesting is that Gimbal is from the same aircraft/crew during the same flight and the Gimbal object does not display a RNG when tracked by ATFLIR.
MclachlanM
Active Member
This might be more evidence to the RNG coming from the radar. I had a hard time getting a radar lock on the F-18 in my recreation of 'gimbal' and had to make the other fighter a friendly to pick it up easily using the information donor contact through datalink (the symbol on the radar tells us ).
If my recreation is close to what is actually happening then it would be possible for the WSO to spot a stealthy fighter with the ATFLIR which the radar cannot pick up. Although the audio in the video about whether or not the object seen is the L&S target is confusing, as well as what we here about "A hole fleet of them".
Is it possible that what we see is a stealth fighter like in my recreation, not using datalink and actively spoofing the F-18s radar?
https://blog.bliley.com/radar-jamming-deception-electronic-warfare
I don't know how unlikely or possible this is though as I'm being speculative now and the ECM measures for fighters are obviously very hard to get info on.
MclachlanM
Active Member
Just the default RWS, you would only use TWS for multiple targets as it is less precise than STT.
jarlrmai
Senior Member
Did you have a weapon selected?
MclachlanM
Active Member
I wasn't armed but think I had a sidewinder selected. The fact that I struggled says more about my skills in DCS rather than what the aircraft is capable of, my point is that if the other aircraft was something like an F22, F35 or the mysterious NGAD then it would be practically invisible anyway.
For example here is an F22 that is invisible to both the radars of a Su27 + AWACS and can start flying formation without either detecting it (in DCS).
Source: https://youtu.be/DnUTPwfuJHE?t=251
MclachlanM
Active Member
In fact the USS Theodore Roosevelt was testing a low RCS design at around this time period: https://en.wikipedia.org/wiki/Northrop_Grumman_X-47B. Surely a perfect fit for the type of thing that could be seen by ATFLIR at 15nmi but not radar?
It's a drone, which might explain the pilot dismissing it as a drone. Has this been ruled out?
It's a drone, which might explain the pilot dismissing it as a drone. Has this been ruled out?
DavidB66
Senior Member
Surely that is far too large. Even if the object is down by the water, it can't be more than a few meters across, or smaller if Edward Current's simulation, linked at #8 above, is correct.In fact the USS Theodore Roosevelt was testing a low RCS design at around this time period: https://en.wikipedia.org/wiki/Northrop_Grumman_X-47B. Surely a perfect fit for the type of thing that could be seen by ATFLIR at 15nmi but not radar?
It's a drone, which might explain the pilot dismissing it as a drone. Has this been ruled out?
jarlrmai
Senior Member
I think Go Fast and Gimbal are getting mixed up again. "It is a drone bro" is from Gimbal.
jarlrmai
Senior Member
There's nothing to indicate that the Go Fast object is anything other than balloon possibly with a radar reflector, you might be mixing up gimbal again.
The RADARs settings/range etc in RWS are determined somewhat by the weapon you have selected.
MclachlanM
Active Member
Apologies, I was talking about this, probably the wrong thread to start getting into detail about the Gimbal video
MclachlanM
Active Member
In the 0.7 seconds where the operator is not slewing, you see that the downward angle goes from -22 to -23 degrees, to me this looks like the INR mode (i.e. ATFLIR is pointed at a fixed azimuth and elevation with respect to earth rather than the plane.) and I agree not 'snowplough' mode.
By the way, this is also a confusion between litening and ATFLIR pods on my part. Worth making clear also that inertial LOS and Snowplough are completely different.
Having a play in DCS I can't tell any difference between the snowplough and INR modes and can't get anything that looks like what's in GoFast so I'm going to assume INR isn't modelled, not sure if that is a valid assumption though?
For reference here is a comparison between the slew in DCS and GoFast, note that the water (barely visible but it's there) is travelling much faster in DCS:
Decided to do the maths properly, assuming that the ATFLIR in gofast is looking at a fixed point, modelling 0.7 FOV and the az/el angles in CAD then projecting the length that a point moves in 0.7 seconds to get the speed:
I'm really unhappy with this because it seems far too fast for it to be wind although If GIMBAL is recorded from the same flight then this number isn't completely ridiculous as they quote a wind speed of 120knots to the west and we know that this isn't the jets' speed as it would be much faster (360 knots).
How does moving the ground texture in the simulation 150knots, 170degrees to the right from the initial velocity vector effect the accuracy to the original?
jarlrmai
Senior Member
This is the description of the pointing modes from the FSX F/18E manual at:
https://forums.vrsimulations.com/su...ing_Infrared_(FLIR)#A.2FA_FLIR_Pointing_Modes
3.2 Inertial LOS Mode
When the FLIR is in the Inertial LOS Mode, the MC maintains the FLIR LOS at a constant pointing angle with respect to the inertial reference plane. TDC slewing with TDC priority to the FLIR causes the FLIR LOS to be slewed accordingly. The Inertial LOS pointing mode may also commanded via the Az/Ez format (FLIR as the selected sensor), by depressing the TDC ENT in open space to invoke the scan centering cross, followed by a second depression to select the LOS.
I think maybe we are looking at some slight version/mode differences between what we see in the Navy videos and what the simulators show/have. The box selection grab method we see in Go Fast is not mentioned in either sim, so possibly it has a pointing mode that supports that box selection.
We deducted that the figure in the bottom middle of the DDI is the time in seconds since mission start or some other point.
In Go Fast it is ~4251 at the end and GIMBAL is ~5306 so around 17-18 mins between videos so at
https://forums.vrsimulations.com/su...ing_Infrared_(FLIR)#A.2FA_FLIR_Pointing_Modes
3.2 Inertial LOS Mode
When the FLIR is in the Inertial LOS Mode, the MC maintains the FLIR LOS at a constant pointing angle with respect to the inertial reference plane. TDC slewing with TDC priority to the FLIR causes the FLIR LOS to be slewed accordingly. The Inertial LOS pointing mode may also commanded via the Az/Ez format (FLIR as the selected sensor), by depressing the TDC ENT in open space to invoke the scan centering cross, followed by a second depression to select the LOS.
I think maybe we are looking at some slight version/mode differences between what we see in the Navy videos and what the simulators show/have. The box selection grab method we see in Go Fast is not mentioned in either sim, so possibly it has a pointing mode that supports that box selection.
We deducted that the figure in the bottom middle of the DDI is the time in seconds since mission start or some other point.
In Go Fast it is ~4251 at the end and GIMBAL is ~5306 so around 17-18 mins between videos so at
MclachlanM
Active Member
Did a quick job in blender to try and see what the 150 knots I calculated looked like, just using @dimebag2 s geometry https://www.metabunk.org/threads/go...-the-jet-or-the-camera-only.11812/post-251755
But I've added the third vector for the water plane, here's what it looks like:
I think better but still not fast enough?
But I've added the third vector for the water plane, here's what it looks like:
I think better but still not fast enough?
jarlrmai
Senior Member
MclachlanM
Active Member
Absolutely, the more I compare the videos the more I'm convinced my mind is playing tricks on me.
Before the render I thought it was going too fast, after the render it looks too slow:
Try looking at it frame by frame: pick a wave/structure/feature/bunch-of-pixels, and count how many frames it takes to get to the other side of the image (or traverse the whole field of view). Then you can really compare with numbers and not rely on your eyes.
Edit to add: make sure your animation has the same frame rate than the video
MclachlanM
Active Member
dimebag2
Active Member
Creating a velocity vector for the background in my 3D model, the speed of the background should be ~300 knots if the object is at 12000 ft and ~4Nm from the plane (basically the speed of the plane minus the speed of the object going in the same direction). If the object was near the surface, its speed would be ~200-300 knots opposite to the plane, and the speed of the background would be more of the order of 550-650 knots. Which one fits the video better ?
If I try a rough calculation, with a FOV of 0.7deg, and an initial distance of ~9.4Nm for the background (derived from the model), there is 0.7*9.4*pi/180 = 0.115 Nm in one frame. It takes roughly ~2sec for a feature in the background to cross its diagonal (i.e. sqrt(2)*0.115=0.162Nm
-> The speed of the background is about 0.162/2*3600= 292 knots.
It's clearly more in the range of the high altitude object.
If I try a rough calculation, with a FOV of 0.7deg, and an initial distance of ~9.4Nm for the background (derived from the model), there is 0.7*9.4*pi/180 = 0.115 Nm in one frame. It takes roughly ~2sec for a feature in the background to cross its diagonal (i.e. sqrt(2)*0.115=0.162Nm
-> The speed of the background is about 0.162/2*3600= 292 knots.
It's clearly more in the range of the high altitude object.
MclachlanM
Active Member
The area of the ocean we see is not equal to the FOV of the camera where it meets the ground?
We agree on the horizontal distance (roughly) but as the camera is pointing down at a shallow angle it is picking up more of the ocean in the vertical axis, I end up getting the similar numbers to whatever magic Derick was doing here (trig hurts my head);
782ft (= 238m)
1845ft (= 562m)
I get a diagonal distance of 647m, Derick's numbers give 610m so say 630m:
630/2*3600 = 1,134 km/h =612 knots!
MclachlanM
Active Member
The problem is that the ground speed of jet - object (360 - 57 = 300 knots) can't seem to get close to the ground speed seen.
I though that the ground speed relative to air might be amplified by parallax but even at the seemingly high estimate of 150knots (from assuming that the camera is in INR mode and that the brief period at around 2 seconds in when the operator doesn't slew shows the velocity of the ground) it doesn't seem to add up.
What else could be happening here that I'm missing?
MclachlanM
Active Member
Sorry, 300knots TAS, corrected to ground speed by the 150 knots wind/current calculated, if we're looking at a ground speed in the video of 600knots you would need a 300knot wind which sounds silly
dimebag2
Active Member
Yes that's right I did not account for the inclination of the background in the FOV. Well if your numbers are correct then the object would have to be close to the ocean. Is it possible that the instruments estimated a distance from the camera angles (because the given RNG match the lines of potential position for the object), but somewhat underestimated it ?
jarlrmai
Senior Member
What instrument other than the RADAR can provide a range to a an object in the air?
MclachlanM
Active Member
I don't think so, jarlrmai has convinced me that the ATFLIR is in A/A mode using the RADAR. If range is coming from something else (i.e. coords for ground targets) then the F-18 seems to use different notation to make that clear. At least in the manuals RNG only ever refers to target range obtained by a RADAR track
Range and closure speed from RADAR nomenclature used in A/A:
Target range obtained by the navigation computer / ATFLIR used for A/G:
Despite Chris Lehto never having used ATFLIR and not being able to tell it apart from the Litening pod, what he says about range still bothers me a bit though.
So this is meant to be the ground speed at what point exactly?
At the beginning when there is no track and the water is slowly panning I got 133knots.
But if you're doing the calculation after remember the distance across and down will change as the camera does, it gets lower.
MclachlanM
Active Member
With no track at the beginning I got 150 knots:
https://www.metabunk.org/threads/go...-the-jet-or-the-camera-only.11812/post-251902
I was responding to dimebag2's rough estimate so it is just a ballpark number to show that the velocity of the ground is well above what we expect.
Did the calculations for 30seconds in and got 423 knots ocean speed
This doesn't seem that far off the near 400knots the plane is doing and could easily be explained by wind?
Edit: well the speed of ocean and jet don't need to match of course, that's only if the object is directly in the middle.
Minus the speed of the object X 2 if it is in the middle.
Last edited:
Similar threads
