@MaxWaldorf said in AIM 120:

Guys, at what point did you not understand to resume this discussion for U1?

I dont recall anyone saying that was a requirement, just alot of people saying they were going to do that. Also the issue I am concerned with is apparently not considered a bug and Mav isnt even willing to discuss it, so I dont see how waiting for U1 will accomplishing anything if TWS being so inaccurate that it cant tell the difference between jets with 6nm of separating flying in a straight line as not being an error.

@MaxWaldorf said in AIM 120:

@shift8 said in AIM 120:

@MaxWaldorf said in AIM 120:

Guys, at what point did you not understand to resume this discussion for U1?

I dont recall anyone saying that was a requirement, just alot of people saying they were going to do that. Also the issue I am concerned with is apparently not considered a bug and Mav isnt even willing to discuss it, so I dont see how waiting for U1 will accomplishing anything if TWS being so inaccurate that it cant tell the difference between jets with 6nm of separating flying in a straight line as not being an error.

2. Unless you have tangible not classified data (not based on other sims) that can prove that TWS should be more precise, feel free to share.

How about the entire concept of how TWS works? TWS is basically just a narrow volume search with a a set of track correlation algorithims. The individual radar hits are just as precise as they would be in search. Range doppler and angle can be resolved far better than the current BMS implementation suggests.

Where there would be errors is in the the statistical correlation of the targets, where tracks might be confused with other tracks or somehow meneavuer outside the correlation box such that on the next real radar hit the target is outside what the radar considers to be the max possible statistical distance for that track.

In the case of a high speed target, or several, the size of the max statistical distance in the coordinate system would increase in size, which would complicate the ability of the radar to correctly associate tracks. But unless the system is very very primitive, most of the time an error that swaps two tracks in the same correlation box will still result in both targets being tracked, but the missile that is going to them might be swapped.

In any case, this kind of thing is not what we have going on BMS. In BMS the system gets confused by targets flying a moderate speeds in a straight line. Furthrmore, for the system to become confused, it would need to by unable to associate the track in the entire coorinate system, not just one discriminate. For example several targets with similar predicting forward speed and therefore similar predicted position on the next update will generally have different positions in the track coordinate system in azimuth and elevaton, doppler etc.

What I have seen in game, and from other users, is that the deviations are humongous and they do not seem to have much if any bearing on the actual geometry the radar is dealing with. I find it very unlikely that you have replaced the old system however simplistic, with a realistic model of the actual statistical models in a TWS system because doing that under the hood in addition to running the rest of the game (not to mention for every jet in the game) seems unlikely. More likely, there is some kind of RNG system that has been applied to the radar in TWS, and it does not seem to care at all about what the targets are actually doing and how that would actually probabalistically affect such as system.

@MaxWaldorf said in AIM 120:

@shift8 said in AIM 120:

@MaxWaldorf said in AIM 120:

Guys, at what point did you not understand to resume this discussion for U1?

I dont recall anyone saying that was a requirement, just alot of people saying they were going to do that. Also the issue I am concerned with is apparently not considered a bug and Mav isnt even willing to discuss it, so I dont see how waiting for U1 will accomplishing anything if TWS being so inaccurate that it cant tell the difference between jets with 6nm of separating flying in a straight line as not being an error.

2. Unless you have tangible not classified data (not based on other sims) that can prove that TWS should be more precise, feel free to share.

How about the entire concept of how TWS works? TWS is basically just a narrow volume search with a a set of track correlation algorithims. The individual radar hits are just as precise as they would be in search. Range doppler and angle can be resolved far better than the current BMS implementation suggests.

Where there would be errors is in the the statistical correlation of the targets, where tracks might be confused with other tracks or somehow meneavuer outside the correlation box such that on the next real radar hit the target is outside what the radar considers to be the max possible statistical distance for that track.

In the case of a high speed target, or several, the size of the max statistical distance in the coordinate system would increase in size, which would complicate the ability of the radar to correctly associate tracks. But unless the system is very very primitive, most of the time an error that swaps two tracks in the same correlation box will still result in both targets being tracked, but the missile that is going to them might be swapped.

In any case, this kind of thing is not what we have going on BMS. In BMS the system gets confused by targets flying a moderate speeds in a straight line. Furthrmore, for the system to become confused, it would need to by unable to associate the track in the entire coorinate system, not just one discriminate. For example several targets with similar predicting forward speed and therefore similar predicted position on the next update will generally have different positions in the track coordinate system in azimuth and elevaton, doppler etc.

What I have seen in game, and from other users, is that the deviations are humongous and they do not seem to have much if any bearing on the actual geometry the radar is dealing with. I find it very unlikely that you have replaced the old system however simplistic, with a realistic model of the actual statistical models in a TWS system because doing that under the hood in addition to running the rest of the game (not to mention for every jet in the game) seems unlikely. More likely, there is some kind of RNG system that has been applied to the radar in TWS, and it does not seem to care at all about what the targets are actually doing and how that would actually probabalistically affect such as system.

@MaxWaldorf said in AIM 120:

Guys, at what point did you not understand to resume this discussion for U1?

I dont recall anyone saying that was a requirement, just alot of people saying they were going to do that. Also the issue I am concerned with is apparently not considered a bug and Mav isnt even willing to discuss it, so I dont see how waiting for U1 will accomplishing anything if TWS being so inaccurate that it cant tell the difference between jets with 6nm of separating flying in a straight line as not being an error.

@Mav-jp said in AIM 120:

@shift8 said in AIM 120:

@Mav-jp said in AIM 120:

I won’t comment about the TWS va STT Probability of Guidance as there is nothing to comment about

A 6nm error isnt worth commenting about?

Absolutely not

So I guess your content with it being wrong then

@Mav-jp said in AIM 120:

I won’t comment about the TWS va STT Probability of Guidance as there is nothing to comment about

A 6nm error isnt worth commenting about?

@Mav-jp said in AIM 120:

Because the aim120 are not guiding to the targets but to a position where the FCR believes that the target is.
The accuracy of this FCR estimation is very dépendant on FCR submode
On TWS which is the worst , the accuracy on position and speeds are very poor. When the aim120 then activated his seeker , he will look at the position where it believes the target is. But in reality this target can be miles out. If other targets are within the seeker it will pick up those ones.
The aim120 can not check if the target it’s seeker picks up is the same of the original
From FCR
This is why having q clear avenue of
Fire is crucial
If you want to maximize your chance to pick up the correct target , fire in STt…and even then it tWo contacts are very close there will be no garantee that the aim120 will pick the original one

No doubt that STT is more precise than TWS, but this example makes it clear that something is off. TWS tracks would not have an error of 6nm. If they did, they would not even be able to build and maintain the track files, because there would be far too much uncertainty. If the radar is confusing the tracks at 4nm for the ones at 10nm, than the correlation algorithims would be going nuts because they would be seeing 4 targets in the same zone. Range, doppler, and angle accuracy in TWS is going to be the same as that in search for the actual radar hits. Even HPRF radars that must rely are far less accurate FM ranging have much better range accuracy than that (and thats talking true range, not apparent). And there is more than enough dwell time during the raster scan to obtain sufficiently accurate apparent range, doppler, and angle information even of closely space targets unless they are so close that they are inside the same res cell, and even then there are ways of resolving that problem.

In this particular example, we have a MPRF only radar and the first set of targets in virtually certainly inside the first interval of Ru, and the second set of targets is right on the edge. There is not possibility of overlapping in apparent range. The targets probably also have different doppler, and the have different angles. All of these things go into the TWS logic.

What appears to have actually happened here is that there is something missing from the datalink logic and when the missile went pitbull, almost immediately after coming off the rail given the range, it was not told to ignore the closer in targets.

Ah ok. I thought it was supposed to say centered on the bugged target. Was this recently added? I have never seen it before.

Starting noticing this behavior post 4.33. Anyone know what this is? Also have been having an issue with TWS stopping its scan. The Azimuth sweep just sticks in the center of the azimuth lines and wont start again unless I go RWS. Dont have footage of that yet but will post it when I do.

@Blu3wolf:

The “goalposts”? There’s no goal here.

There would be a goal, but the participants have differences of opinion which are axiomatic in nature. This is why the discussion is circular…

https://en.wikipedia.org/wiki/Moving_the_goalposts

This discussion is not axiomatic in nature. It is factual. The discussion is circular because of one end of the argument holds the position that you can substitute irrelevant historical statistics as if they were physics themselves, instead of discussing the actual technical details of the issue.

I like how the goalposts of this discussion have shifted from “can a flare defeat the 9x” to “but look! it could be defeated by obsurcation from a smoke counter measure”

incidently, it was stated several times early in this thread that obscuration can defeat a 9x the same way flying behind a mountain can. Or using your wingman as a flare.

@Blu3wolf:

Your linked image even explains how the flare works against an imaging seeker…

If you spent as much time reading about the subject as you did writing the post, you would have skipped writing the post XD

Are you referring to the bit I already referenced in the previous posts? Where the flare works simply because it obscures LOS? I would not consider this the flare working. You could just as easily have a smoke generator.