Coded my logic above and after three bug fixes (implementation, not logic...), out pops:

Group=AAU, Unit=001, DegStart=-1, DegEnd=-1 (means unit is in Main Body)

Group=AAU, Unit=000, DegStart=0, DegEnd=360 (means unit is assigned to all sectors in ring)

Group=ABS, Unit=003, DegStart=315, DegEnd=45

Group=ABS, Unit=002, DegStart=45, DegEnd=90

Group=ABS, Unit=001, DegStart=270, DegEnd=315

Group=ABS, Unit=000, DegStart=-1, DegEnd=-1

 

It passes the initial smell test. The only logic change was to test directly for all sectors being selected. The sectors are stored as bits flags in a larger integer, so adding up the bit values for each sector gives 255 so I can just directly ask if sectors == 255 then we know the unit is assigned to all 8 sectors.

Say you are a Spru or a Perry on ASW picket duty for a CSG. What do your sprint and drift intervals look like?

 

ex. Sprint for 30 minutes, drift for 10

 

or distance based or ...

Say you are a Spru or a Perry on ASW picket duty for a CSG. What do your sprint and drift intervals look like? ex. Sprint for 30 minutes, drift for 10 or distance based or ...

 

You will want to sprint far enough to make it worth the effort.

 

In other words, when you're creeping and acoustic conditions are at their optimum, your detection range is at its peak.

 

So, you can potentially already hear out to the first CZ (30-33 nm), possibly further.

 

When you sprint, then, you're intending to be in a position (at the end of the sprint) that allows you to hear out to a new, useful distance. I would say at least half of that CZ (15-17 nm).

 

This means, if you're a typical warship sprinting at 25-30 knots, you need to do so for at least 25-30 minutes to move 12-15 nautical miles.

 

When you start the drift, you're trying to once again attain optimum acoustic conditions, both for the sake of your equipment and to try and catch an opponent making too much noise (e.g. trying to catch up, intercept, etc.).

 

And, then, if you get a sniff of a contact, trying to work out a solution without giving your precise position away.

 

Calculating exactly how long this will take (either in terms of time or distance) is a relatively complex physics question, but I think a 10-15 minute drift makes good sense as a rule of thumb.

While drifting it may be a good idea to make a slow 90 degree turn every now and then.

 

The towed array is unable to see contacts in front of your ship. Once you find the contact try to estimate it's course and not heading directly towards at high speed giving your self away.

 

Try and not use your active sonar. Active sonar gives your position away.

 

A sub can hear your sonar pining a long way before you will be able to detect it.

 

If your ship is part of a task group you want to stay 50 to 75 miles in front of the group to reduce the noise that comes from all the ships in the task force.

 

I hope this helps!

While drifting it may be a good idea to make a slow 90 degree turn every now and then. The towed array is unable to see contacts in front of your ship. Once you find the contact try to estimate it's course and not heading directly towards at high speed giving your self away. Try and not use your active sonar. Active sonar gives your position away. A sub can hear your sonar pining a long way before you will be able to detect it. If your ship is part of a task group you want to stay 50 to 75 miles in front of the group to reduce the noise that comes from all the ships in the task force. I hope this helps!

 

Eric, the question is about ship 'sprint and drift' behaviour in the ASW zone of the formation rather than player controlled ASW tactics generally.

 

Btw, a towed array is omnidirectional. It does not suffer the same kind of problem with 'baffles' that a bow mounted sonar would.

Thanks guys, indeed towed arrays in the game can 'see' directly ahead, though I've seen some websites suggest a 30 degree blind spot centered on the towing ship (maybe only for VDS?).

 

What the game does model is sharp turns causing the array to be useless until it has 5 minutes to unbend. Strangely this is triggered when the ship or sub hits a course point and moves on to its next no matter how small or how severe the turn. Currently formation reshuffling doesn't cause this to happen in the game.

Thanks guys, indeed towed arrays in the game can 'see' directly ahead, though I've seen some websites suggest a 30 degree blind spot centered on the towing ship (maybe only for VDS?).

'Towed array' is a very broad term, of course. Linear, passive towed arrays do not suffer from a blind spot.

 

VDS - usually used to describe active sonars - would be more prone to a blind spot, especially when deployed at a shallow depth.

 

What the game does model is sharp turns causing the array to be useless until it has 5 minutes to unbend. Strangely this is triggered when the ship or sub hits a course point and moves on to its next no matter how small or how severe the turn. Currently formation reshuffling doesn't cause this to happen in the game.

Yep, and I like the behaviour. A linear towed array can be thousands of feet long when fully trailed.

Oh crap, MoveUnits (separate from formation-keeping) is too smart and needs to be rewritten to make this better formation-keeping work. Unfortunately MoveUnits does more than just move every unit. It does thinks like check if units are aground, sees if units are slaved in a formation, looks to see if the unit is a plane or not, ...

Oh crap, MoveUnits (separate from formation-keeping) is too smart and needs to be rewritten to make this better formation-keeping work. Unfortunately MoveUnits does more than just move every unit. It does thinks like check if units are aground, sees if units are slaved in a formation, looks to see if the unit is a plane or not, ...

The good news is that it only took about 45 minutes to write a new ExportDLL that does just that.

I took a trip thru the ReleaseNotes.txt just now. I need to go back to 2016.003 for a playable game. That is where gradual altitude changes were introduced. That is a whole lot of good changes ago .

So we shouldn't bother playing with 2017.16 and formation keeping? (which I only just noticed )

So we shouldn't bother playing with 2017.16 and formation keeping? (which I only just noticed )

 

I would be surprised if you get a reasonable game play experience with 2017.016. Units in the formation going all over, likely not where the Group is going on the map. To put it nicely, units drifting out of formation. I've come a long way by adding a couple of ExportDLLs but haven't uploaded them to HG since it isn't all far enough along to share.

 

You are welcome to use 2017.016 but your results may vary

This week brought unexpected progress. Thanks to eeustice this week for testing what were totally untested builds to bring us to 2018.004 where formation-keeping seems to be back to how it was in the olden-days, functional even if it isn't ideal. In any event, I'm feeling comfortable with 2018.004. I'll refine this state for a little while then we can decide where to go from there.