Maverick through clouds

[Aginor]

Hey guys!

Not sure whether this is the right section, please move if necessary.

 

Me and another user over at lockonforum.de noticed that the IR Maverick (D-version) locks targets through clouds quite nicely, I would even say too well. Is that a bug or intended behaviour?

I am not an expert for infrared sensors, but if I recall correctly you can't see through thick clouds (like when you set clouds to 10/10 in the editor, 100 meters thick), much less lock on a target.

The sensor locks on contrast, on the same info I can see on the MFCD. It is completely white, no contrast visible that I could lock on. So it shouldn't hit, I guess.

 

Or am I wrong?

 

You can test it by just turning labels on: Aim for the labels and lock on, it works.

I will do some more precise tests if you wish.

[SDsc0rch]

yeah, there should definitely be a line-of-sight test

 

true for a lot of things in DCS actually..

 

not just for visual wavelengths but IR as well

[Aginor]

Yeah, line of sight would be cool for a lot of things. For example those damned AAA and tanks seeing and shooting right through trees and stuff. Even houses sometimes.

[Flagrum]

It could be that the IR Mav actually can see through the clouds as IR waves might be able to penetrate them to some extend. And we might not see this in the cockpit because the IR representation is only "faked" (different contrast, etc.) afaik.

 

A lot of "if"s here ... but perhaps some deeper investiation is necessary here. Have you tried the same with a CCD Mav?

[Buzpilot]

Hey guys!

Me and another user over at lockonforum.de noticed that the IR Maverick (D-version) locks targets through clouds quite nicely, I would even say too well. Is that a bug or intended behaviour?

I am not an expert for infrared sensors, but if I recall correctly you can't see through thick clouds (like when you set clouds to 10/10 in the editor, 100 meters thick), much less lock on a target.

The sensor locks on contrast, on the same info I can see on the MFCD. It is completely white, no contrast visible that I could lock on. So it shouldn't hit, I guess.

 

Or am I wrong?

 

I think the IR is getting blocked at +200m, if I remember it right.

Edited October 28, 2014 by Buzpilot

[Aginor]

I didn't have the time to test it yet, but I will do that in the next few days.

 

Of course IR should be less prone to bad weather than others, that's one of the reasons why IR is used after all.

I just wonder how good it is in real life. 100m of clouds is a lot, I don't have a clue whether that would work in reality.

 

And then of course, like Flagrum said: Maybe the missile has the capability, and the picture on the MFCD is wrong, it should also provide a look through the clouds, that's possible.

[EtherealN]

Note: there are line-of-sight tests already in place. (Example: note how your radio sucks if you're flying in canyons.)

 

Trees is a separate issue: there the line-of-sight test results in PASS in the old engine due to the fact that trees are not collision-modeled. (Houses that are not collision modeled are probably bugs, so if you can reliably replicate this please report the location in the bugs forum and the Test Team will look into it.) This was due to a limitation of the old terrain engine, where such collision modeling would basically require either very blunt "blocks" as collision models - meaning you might collide with "thin air" when flying close to trees - or just die because of the level of computation required for the old method.

 

The issue with detection is basically an offshoot of that issue.

 

A theoretically possible fix would have been to "instance" the same visual detection routines used for players (painting a 3D scene and detecting contrast), but everyone would regret doing that when 20 tanks do it at the same time and all of them ask your computer (in case of singleplayer) or the server (in case of multiplayer) to handle this computation in realtime, effectively asking your computer to render an additional 20 3D scenes...

[Aginor]

I think most of us are aware that things like that are always tradeoffs, and it has to be like that because of limited performance. But maybe we can help to find a way to make those techniques a little less obvious.

 

I guess we won't know until release of DCSW 2.0, but some of those problems may even be solved already. We'll see. :)

 

As for the original topic: Does anyone have a reliable source what can be done by sensors such as those used in a Maverick? Maybe we can come up with a mathematically simple approximation what should be possible. Maybe we can even find a way how the sim could handle it.

 

Some basic paper about the topic: http://www.deltagearinc.com/library/OpticsFacts/EO.pdf

 

And this paper talks specifically about cirrus clouds. Cool math involved.

http://web.atmos.ucla.edu/~liougst/Group_Papers/Liou_AO_29_1990.pdf

[Aginor]

I did a few tests. There is definitely something fishy.

 

base of clouds at 2000m

overcast 10/10

 

altitude of flight: 10,000 ft.

cloud thickness: 200m

AGM-65D (IR) locks on at 7 miles slant range.

AGM-65K (CCD) locks on at 6 miles slant range.

 

cloud thickness: 500m

AGM-65D (IR) locks on at 6 miles slant range.

AGM-65K (CCD) locks on at 5 miles slant range.

 

cloud thickness: 1000m

AGM-65D (IR) locks on at 6 miles slant range.

AGM-65K (CCD) locks on at 5 miles slant range.

 

cloud thickness: 2000m

altitude 13,000 ft., to get out of the clouds

AGM-65D (IR) locks on at 6 miles slant range.

AGM-65K (CCD) locks on at 5 miles slant range.

 

 

I marked the targets with the TGP, but if I am not mistaken that should not allow the Maverick to lock on.

 

btw: Sometimes the TGP can look through the clouds as if they aren't there...

Edited October 29, 2014 by Aginor

[aceviper]

The infra red sensor locks on a heat source....not visible to the human eye.

From wiki.........." Maverick D replaced the electro-optical guidance with an imaging infrared system which doubled the practical firing distance and allowed for its use at night and during bad weather. A reduced smoke rocket engine was also introduced in this model. It achieved its initial operation capability in 1983."

Edited October 29, 2014 by aceviper
.

[Aginor]

I know, aceviper, but the picture on the MFCD shows exactly what the seeker head sees, not what the human eye would see. So if you can't see a contrast on that screen, the Maverick cannot lock on.

And IR does not allow you to see through several hundred meters of clouds. That's what RADAR is for.

 

Clouds block infrared light just like visible light, maybe not equally well but they certainly do.

And as you see the CCD version also locks on through 2000m of clouds. And that one works in the visible spectrum.

[vicx]

This is an issue even if it is refused classification as a bug. This is another fundamental failing in the simulator.

 

It is curious how some pilots are very selective when it comes to acknowledging issues in the sim. Some pilots will happily raise realism issues to give themselves more of an edge but will oppose discussing any issues in the sim that might see an absurd super-power taken away.

 

This type of pilot-pleasing skews DCS into being more of an ego-shooter than some would like to admit.

 

IR sensors are an optical sensor. Optical wavelengths get scattered by clouds and other mediums. 10/10 clouds are meant to represent a pea-soup visual environment of near zero visibility. How far you can see when flying through 10/10 clouds in DCS. It is less than 5m.

 

There is no way any near-optical wavelength sensor can see through 10m of pea-soup cloud let alone 500m.

[Flagrum]

This is an issue even if it is refused classification as a bug. This is another fundamental failing in the simulator.

 

It is curious how some pilots are very selective when it comes to acknowledging issues in the sim. Some pilots will happily raise realism issues to give themselves more of an edge but will oppose discussing any issues in the sim that might see an absurd super-power taken away.

 

This type of pilot-pleasing skews DCS into being more of an ego-shooter than some would like to admit.

 

IR sensors are an optical sensor. Optical wavelengths get scattered by clouds and other mediums. 10/10 clouds are meant to represent a pea-soup visual environment of near zero visibility. How far you can see when flying through 10/10 clouds in DCS. It is less than 5m.

 

There is no way any near-optical wavelength sensor can see through 10m of pea-soup cloud let alone 500m.

Huh!?

 

Who refuses to see this as a bug and who is selective in his perception here and opposes discussion the issue?