[FIXED][1487] Sonar 'clumps' targets at high latitudes

[Redeye43]

This is an issue that's been reported in the past, but it seems to still be around in 1487 and impairs SOSUS function quite a bit. Peek at SOSUS #3 to see it.

[Raptorx7_slith]

Thanks, logged as 0016367

[Quark73]

It's already logged as 0015060 here.

[Nikel]

Simple curiosity question.

Is the sonar part of the game more difficult to simulate?

And so the sonar bugs or problems created more difficult to fix?

[Quark73]

I assume it is difficult to simulate an array, means a chain or a net of sensors, with a single point sensor. That worked somehow with the previous simplified sonar models, but leads into problems with the more realistic approach since Tiny.

[Nikel]

Thanks.

These are the changes in Tiny searching for Sonar:

Changes to sonar cavitation model:
a) Cavitation is now dependent on desired speed (ie. screw turn rate) rather than current
vessel speed
b) The noise boost from the onset of cavitation is now gradual (instead of all or nothing),
and its magnitude depends on how much above the cavitation threshold the desired
speed is

New sonar sensor type 'vertical flank array' added. This sensor type can improve
tracking ability

You mean the pre-tiny Sonar changes?

http://ftp.matrixgames.com/pub/CommandM ... HANGES.pdf

[Quark73]

Maybe, it was the one were the thermocline layer was not established above 70° North.

[Nikel]

Testing this scenario I see that there are inland detections of Soviet subs, even though they do not last long, should this be impossible?

Also there are no icons for the SOSUS, I see them in your SOSUS test scenario in the other thread you linked.

S.png (129.15 KiB) Viewed 395 times

[Quark73]

Also there are no icons for the SOSUS, I see them in your SOSUS test scenario in the other thread you linked.

The SOSUS arrays are grouped with the main station in Iceland. If you press Ctrl + V they become visible.

[Nikel]

Oh, thanks

And the inland contacts?

[Quark73]

And the inland contacts?

I assume problem with sensor fusion. If a contact is observed by two sensors simultaneously CMO narrows the contact down via cross-bearing estimation based on angular resolution of the sensor. If you have two sensors with large scan intervals CMO doesn't fuse them and just shows the direction for the latest sensor that has "seen" the contact. I am not sure if CMO calculates the range of the contact based on signal strength or just puts it in the middle of possible area. Sometimes it looks like the latter sometimes not. Furthermore CMO does no plausibility check on the estimated position, so it puts them on land.

[thewood1]

Frankly, I personally don't care about the inland contacts. Its an uncertainty zone and as a player, I know enough that its only the water part I need to care about. If players can't bring that much common sense to the game, they need to be playing something else. I really would prefer devs spend their time on real issues that have true gameplay impact.

[Nikel]

I assume problem with sensor fusion. If a contact is observed by two sensors simultaneously CMO narrows the contact down via cross-bearing estimation based on angular resolution of the sensor. If you have two sensors with large scan intervals CMO doesn't fuse them and just shows the direction for the latest sensor that has "seen" the contact. I am not sure if CMO calculates the range of the contact based on signal strength or just puts it in the middle of possible area. Sometimes it looks like the latter sometimes not. Furthermore CMO does no plausibility check on the estimated position, so it puts them on land.

Thanks for the detailed explanation

Frankly, I personally don't care about the inland contacts. Its an uncertainty zone and as a player, I know enough that its only the water part I need to care about. If players can't bring that much common sense to the game, they need to be playing something else. I really would prefer devs spend their time on real issues that have true gameplay impact.

Well, I am sure we all want CMO in the best possible shape. Or at least is the main reason for me to post something "strange".

The devs will decide the best way to get it.

[thewood1]

"The devs will decide the best way to get it."

Really? I didn't know that.

[Quark73]

Frankly, I personally don't care about the inland contacts. Its an uncertainty zone and as a player, I know enough that its only the water part I need to care about. If players can't bring that much common sense to the game, they need to be playing something else. I really would prefer devs spend their time on real issues that have true gameplay impact.

This has game play impact if a unit tries to investigate/attack these contacts. Either if it looks in the desert for submarines, or it's heading jumping between two contact positions that are way off each other or if it nukes your SOSUS because it gave the contact it's own position.

[thewood1]

Two perspecitves here.

The player - Again, common sense

AI- this is what missions and zones are for. If any scenario designer is using ASW assets and they aren't on a mission, well, what can I say.

There all kinds of tools and features to limit the impact of this very easily. I bet its easier for us as players and designers than the devs. Again, if everything was perfect, anyone can have at fixing it. I just want a voice out there that says its not a big deal to the point its close to negligible.

[Dimitris]

Is this behaving better in Build 1544?

[Quark73]

Clumping is gone, inland contacts could still happen.

[Dimitris]

The "inland sub contacts" is an unavoidable byproduct of how sonar AOUs are currently generated, so "known/WAD". Improving this is on our stack but a lower priority than other things in the pipeline.

[Quark73]

Agreed, sometimes you have just a direction and it should be common sense that you won't find a submarine in the middle of Greenland. There are ways in CMO to deal with that.