CMWS detection delay

[Akenar-99]

Hello,

After many deaths, and then many tests, I observed that the detection delay of the CMWS can be quite erratic.

Most of the time, it detects a missile launch immediatly. But sometimes, it takes one second, sometimes more. Which means death each time it happens against a MANPAD shot at 1.5km. It is frutrating, concidering that this system exist to prevent that kind of threat.

I don't know what provoke this delay. It can happen, or not, independently of the configuration (all my tests are in straite level flight at constant speed).

It seems to happen with MANPADS more than with SA13.

 

It is even worth when many missiles are shot successively. It seems that the system doesn't concider the third or fouth shot.

 

What could explain this problem ?

(By the way, it doesn't happen with the Onboard Defense System of the Black Shark 3)

Thanks a lot !

 

 

[Scaley]

There is an intentional delay coded in the lua files. I don't know the reason ED decided to implement that obviously. 

[Akenar-99]

This is quite disturbing.

The Apache has a low IR signature, so you are shot at very close range only and have a short reaction delay available. On the other hand, The CMWS reacts late. Impossible to deploy flares in time. Even in auto deploy setting...

Edited January 8, 2023 by Akenar-99

[Akenar-99]

I would be glad to have some precisions from ED if possible.

I am currently doing some compare with the ODS of the black shark 3 which is way more reactive and gives more situation awareness the the AH 64 CMWS (we know aboute the launch immediatly and  if there is a new shot).

I absoloutly love the job they made on the Apache, but manpads kills me before the warning.

 

[McVittees]

I'd imagine it's not a perfect system in real life and ED just use a random number to model this rather than make some unnecessarily complex algorithm.  See all those KA52s getting shot down in Ukraine? Clearly advanced CMWS and counter measures are not impregnable defences, so don't rely on them.  As important is tactics and intelligence of operational area / target defences.

Edited January 13, 2023 by McVittees

[Scaley]

At the risk of getting more points for encouraging google searches of "things that shall not be googled" - there are quite a few open source, legal, officially release and vetted, references on how the CMWS works. Any cursory inspection of said sources will tell you the system we have in DCS is (deliberately, I assume) different. Given that we can't use or refer to any references for the system (it's too new to be allowed) then it's probably best to work with, discuss, and develop tactics for the system as it's been implements in the DCS Apache by ED, rather than comparing it to how we think it should work in reality. There have already been a few threads about possible changes to the CMWS (on big one started by me) and the upshot of those is that the ED team are pretty happy with the implementation they have now. I don't think we are likely to see any significant changes, so if the Apache system as we have it in DCS has a delay, then I'd just take it as read that the system has a delay. If that renders it less useful then so be it. 

[Akenar-99]

Considering that the system detect the missile exhaust flash (I don't know on which frequency and will obviously never know), and can do it from far away in dcs, I don't understand.

I know what you mean. But still, it can be really slow to react. Not immediate, I can understand, but so late, it is not very useful anymore.

 

Here is a track demonstrating that (Auto-dispense ; steady flight, more than 500ft, shooting distance of about 2 km).

 

It is even worse if you adapt your tactic (understand flying low). They detect you later, shoot from closer, and the CMWS is totally useless.

CMWS Test.trk

Edited January 14, 2023 by Akenar-99

[Akenar-99]

Same test with an SA18 igla with no mods if necessary for track reading

 

CMWS Test 2.trk

[Swift.]

4 hours ago, Akenar-99 said:

Considering that the system detect the missile exhaust flash (I don't know on which frequency and will obviously never know), and can do it from far away in dcs, I don't understand.

I know what you mean. But still, it can be really slow to react. Not immediate, I can understand, but so late, it is not very useful anymore.

 

Here is a track demonstrating that (Auto-dispense ; steady flight, more than 500ft, shooting distance of about 2 km).

 

It is even worse if you adapt your tactic (understand flying low). They detect you later, shoot from closer, and the CMWS is totally useless.

CMWS Test.trk 250.31 kB · 0 downloads

 

I think what Scaley is saying is that whatever we have in DCS is the design intent of DCS. Obviously the way it works IRL is different to DCS. But there is no intent to dig into the accuracy of the CMWS system against real world, because its a very sensitive area. 

So instead of feeling bad that the DCS system delays more than you think it should, we would be much better served by understanding the behaviours and limitations of the DCS system and building our tactics based on that.

Namely, that it has a slight delay, that it can't discriminate between missiles fired at the aircraft and other missiles, etc. What we have is a rudimentary system to detect any launch within X amount of time, and that's all we have.

[Akenar-99]

I know what you mean and honesty, I agree in a way.

The point with th Apache is the fact that ground launched IR missiles are shot really close due to the fact that the helicopter seems to have a very low IR signature (which is nice by the way). But the drawback is that if the CMWS has a delay, you are so close when the missile is fired you are already done when it is detected.

 

PS : On the other hand,they have decided to make the Ka50 Onboard Defense System more reactive even if you are shot at longer distances when flying this aircraft. Which of course doen't save you all the time, but is a little bit more confortable.

 

CMWS Test - Ka50 ODS Compare.trk

[GrEaSeLiTeNiN]

ED has many aspects of the Apache module to model and test. So I suppose if ED tested their model of CMWS as much as an ardent consumer would have, in various surface-to-air scenarios, and feel that getting shot down often by manpads is plausible and that the system is as it should be for DCS World, then that's that. Otherwise, they could perhaps take a second look.