Aim-9X way too sensitive to flares

[Pavlin_33]

If you look at the seeker's point of view, you can see that on two occasions it goes for the plume of the afterburner, just aft of the engines.

I am not sure if there was any image recognition in those cases. It looked like the seeker was going for the hotest thing it could find.

[Nealius]

I'd like this to be looked into a bit more as well. Recently I've employed 13 AIM-9X shots on AI and their seemingly unlimited flares, with only four shots connecting. That's an abysmal 30% hit rate.

[ruxtmp]

All air to air missiles seem to have a terrible Pk in single player for me. Does not matter if it is player or AI launched at best any air to air missile is a 50% chance of hit or less. Online against human targets that are somewhat aware the Pk goes down to about 20%

Edited March 9 by ruxtmp

[GGTharos]

On 12/1/2023 at 6:45 PM, Pavlin_33 said:

If you look at the seeker's point of view, you can see that on two occasions it goes for the plume of the afterburner, just aft of the engines.

I am not sure if there was any image recognition in those cases. It looked like the seeker was going for the hotest thing it could find.

Image recognition yes, it just might not mean exactly what you think it means.  It can classify a thing as a target (not this is an airplane, or whatever, just target) and 'these are parts of the target' in some respect and then classify other things as 'not my target' or 'false target'.  Generally without very modern countermeasures a 9X should not be getting decoyed if the decoy isn't making the target IMHO.

[Mad_Shell]

I allow myself to up this report as I've had no answer that is has been looked at or aknowledged by ED.

[Mad_Shell]

@Chizh Hi. Since you're onto the IRCM/IRCCM topic, could you take a look here too please? Thanks.

[Creampie]

No matter which way you want to put it, No one actually knows the capabilities of the AIM9x, R-73.

Everything related to that in the sim has to be done based on speculation. The information that any developer releases on their missiles needs to be taken with a grain of salt, Some boast about them, other release diluted results for national security reasons. Truth is, we do not & will not know. 

 

 

[Mad_Shell]

1 hour ago, Creampie said:

No matter which way you want to put it, No one actually knows the capabilities of the AIM9x, R-73.

Everything related to that in the sim has to be done based on speculation. The information that any developer releases on their missiles needs to be taken with a grain of salt, Some boast about them, other release diluted results for national security reasons. Truth is, we do not & will not know. 

Indeed, we'll never know the exact capabilities. However, if we try to model a missile, we have to take into account the clues we can find to begin somewhere and try to be somewhat close to reality. That's why I did some standardized tests, and found a big difference with the real life demo tests, even if the conditions are close but not exactly the same. 

Edited September 15 by Mad_Shell

[Creampie]

2 hours ago, Mad_Shell said:

Indeed, we'll never know the exact capabilities. However, if we try to model a missile, we have to take into account the clues we can find to begin somewhere and try to be somewhat close to reality. That's why I did some standardized tests, and found a big difference with the real life demo tests, even if the conditions are close but not exactly the same. 

 

That's an absolute fair point. 

But you're talking about a missile with the confirmed kill of a balloon.

Where as for example, The AIM7, depending on the variant have a confirmed hit rate of anywhere between 9%-66?% depending on the variant. Something that is long documented confirmed and public information. I bring this up only as reference to a missile that is being replaced and isn't exactly a contender is must keep secrets.

Another example of this exact issue is the SD-10/120C fiasco. We just do not know and can only speculate and the same goes for ED. 

I get the point you are trying to make, At times the 9X seems a little underperforming. But that's just the RNG roll we get, Is this missile going to hit this after burning & countermeasure dispensing flanker or not? It's just dice rolling on every heater in the game. I've personally had perfect parameters of a 9X & R73 miss jets in burner.

 

[bies]

Even though specific data is classified, some obvious limitations can be estimated. There is an interview with Lt. Col. Fred "Spanky" Clifton, experienced aggressor pilot, with 500 hours in MiG-29 alone and way more than 1000 hours in different NATO jets. He was one of the members of MiG-29 evaluation program. According to him, during the tests in 1990s, US flares proven to be effective against AIM-9M, but not really against R-73 - when Soviet MiG-29 flares proven effective against R-73, but not against AIM-9M.

If 1980/1990s Soviet flares (what we have in DCS MiG-29, Su-27, Su-25 etc) were not effective against 1980s AIM-9M sensor, such flares would be basically useless against 2000s Focal Plane Array, creating a detailed, high-resolution "thermal image" of the target, not just relying on a single "hot spot" like the exhaust or the flare. Especially closer than ~20km.

The focal plane array imaging seeker can distinguish between different parts of the target (engine, wings, fuselage, flare) and it's resilient to countermeasures such as flares because it can maintain lock on the overall shape and heat distribution of the target rather than just focusing on a single source of heat.

[Pavlin_33]

On 9/16/2024 at 9:35 AM, bies said:

Even though specific data is classified, some obvious limitations can be estimated. There is an interview with Lt. Col. Fred "Spanky" Clifton, experienced aggressor pilot, with 500 hours in MiG-29 alone and way more than 1000 hours in different NATO jets. He was one of the members of MiG-29 evaluation program. According to him, during the tests in 1990s, US flares proven to be effective against AIM-9M, but not really against R-73 - when Soviet MiG-29 flares proven effective against R-73, but not against AIM-9M.

If 1980/1990s Soviet flares (what we have in DCS MiG-29, Su-27, Su-25 etc) were not effective against 1980s AIM-9M sensor, such flares would be basically useless against 2000s Focal Plane Array, creating a detailed, high-resolution "thermal image" of the target, not just relying on a single "hot spot" like the exhaust or the flare. Especially closer than ~20km.

The focal plane array imaging seeker can distinguish between different parts of the target (engine, wings, fuselage, flare) and it's resilient to countermeasures such as flares because it can maintain lock on the overall shape and heat distribution of the target rather than just focusing on a single source of heat.

You can't really model any of this in DCS, 'cause flares are just a random generator.

Even if you decrease flare sensitivity, if the target has enough flares one will eventually work.

[bies]

1 hour ago, Pavlin_33 said:

You can't really model any of this in DCS, 'cause flares are just a random generator.

Even if you decrease flare sensitivity, if the target has enough flares one will eventually work.

Yes and no. Yes it can't be modeled using current, extremely simplified countermeasures model, probably not changed much since 1999 Flanker 2.0, developed to run on todays hand calculator CPU.

On the other hand the whole countermeasures system should definitely be rewritten from scratch as even most bascic simulators nowadays use more complex model, it's high time. Flares, chaff, IRST, electronic warfare. I believe ED is working on it.

[Hobel]

vor 9 Stunden schrieb Pavlin_33:

You can't really model any of this in DCS, 'cause flares are just a random generator.

Even if you decrease flare sensitivity, if the target has enough flares one will eventually work.

I wouldn't necessarily say that it's just that anymore, there are a lot more factors that have an influence, otherwise what you see in the videos I present wouldn't be possible if it was just a "random generator". until recently i thought so too, but there are clearly more factors to be seen  

if you go to the missile view (F6) you can see how the flares compete with each other and against the heat source of the aircraft.
it also seems to be the case that the flares seem to cool down, but this is only a vague assertion based on observation.

 

 

 

[Nealius]

In recent missions on the current DCS iteration the 9X seems stronger against flares, at least against Fishbeds. 

[Pavlin_33]

On 9/29/2024 at 5:31 PM, Hobel said:

I wouldn't necessarily say that it's just that anymore, there are a lot more factors that have an influence, otherwise what you see in the videos I present wouldn't be possible if it was just a "random generator". until recently i thought so too, but there are clearly more factors to be seen  

if you go to the missile view (F6) you can see how the flares compete with each other and against the heat source of the aircraft.
it also seems to be the case that the flares seem to cool down, but this is only a vague assertion based on observation.

 

 

 

I don't have any insight about what is actually going on in the background, but if you dump enough flares no IR will hit you. I haven't been hit, in the flanker, by an IR missile for 2 years now, except im cases when I did not know they're coming.