active sonar doctrine question

[SeaQueen]

Originally Posted by Molon Labe

The problem with some of the more popular autocrew is that while the autocrew is freeing up your time (which is good sometimes), it is using information that cannot be known to the player...some autocrew are more than just time-management, they are cheats.

Has anyone like Jamie ever said that? I've often been suspicious of this claim on the basis that most people are also AWFUL at TMA. If you made a computer program that automatically did Ekelund ranging, Speiss ranging, etc. then OF COURSE it'd do better than someone who was basically making guesses.

[SeaQueen]

Originally Posted by LuftWolf

.. there is a fair amount of skill in making any specific AI behavior meaningful in a tactical sense. In other words, we can make it ping, but that doesn't necessarily make it a better OPFOR unless there is other programming done along with to make incorporate the behavior into tactically effective action.

I totally second that. This parallels an issue we run into in professional wargames as well. Something we do frequently is to vary different tactics for a given scenario and see how sensitive the outcome is. Sometimes it matters a little bit, sometimes it matters a lot, sometimes it doesn't matter at all.

Honestly, I think the idea of a kind of universal doctrine for the AI to follow is a little bit ambitious. It amazes me that DW does as well as it does. So many of these things are driven by all kinds of things, which there's just no way to a computer program that's written as generally as DW is, to really take into account. I've been pretty happy with what they got. It beats me sometimes...

[Molon Labe]

Originally Posted by SeaQueen

Has anyone like Jamie ever said that? I've often been suspicious of this claim on the basis that most people are also AWFUL at TMA. If you made a computer program that automatically did Ekelund ranging, Speiss ranging, etc. then OF COURSE it'd do better than someone who was basically making guesses.

Renzie said as much, way back in the day.
The TMA has been improved (which is to say has had more error added to it) in the 1.03 patch, so this has been partially addressed.

[SeaQueen]

Originally Posted by Molon Labe

Renzie said as much, way back in the day.
The TMA has been improved (which is to say has had more error added to it) in the 1.03 patch, so this has been partially addressed.

It surprises me that they did it the way they did, then. Speiss and Ekelund both have a certain amount of error built in. I guess the hard thing is getting it to know what ranging techniques to use when. *shrug*

[jason taylor]

Originally Posted by SeaQueen

Yeah.. but in real life the stations demand more of your attention too. The way I see it, realism wise, is that they're both "realistic" in some sense. I think it's more of a matter of personal preference.

I *like* doing TMA and fussing over bearing rates, so I do it.

The missions I like to play are frequently very long compared to what a lot of other people play too. It's not always possible for me to stare at a sonar screen all day, so I let my autocrew fill in, mostly so I won't miss something. Honestly, though, by this point, based on the mathematics of things, I usually have a pretty good idea of about when I need to start paying attention anyhow. If I know about how long it takes my Pd to reach about 0.5 given the assumptions I've made about my speed and what not, then I start worrying around the median time to detect. It seems to work pretty well. I call it a victory for Koopman. :-)

____________________________________________
By the way there is a story in the RN about a Corvette captain who when he went out on maneuvers clapped a headphones on and waited in the CIC for days, sleeping and eating with the headphones on. During the maneuvers he suddenly woke up from sleep saying, "thats a submarine". When questioned he explained, "it had a metallic sound."

[jason taylor]

Originally Posted by Black_Dingo

I dont profess to know anything about how the doctrines work, but ive been pondering this one for a long time.

AI submarines and active sonar. in DW they basically dont use it. True, that there are very few occasions where a submarine would need to or want to use it. But, I feel there are certain instances where it would be used.

There are two instances where I feel active sonar would be used in a tactical situation:

1. Rapid detection of a time-sensitive target. in other words, if you were a sub searching for a submerged target and needed to find it quickly and destroy it before it could do something you wanted to prevent it from doing, such as launching an SLBM, sending a radio message, deploying commandoes. You might find the need to send out a blast or two of active sonar.

2. Counter-fire snapshot. If by chance youve been made by the enemy and he's got fish in the water after you. you might want to fire snapshots down the bearing of incoming fish and later send out a blast of active sonar to firm up your solution just before you turn tail and evade his fish. This is a common soviet tactic from the cold-war days when the soviets felt that NATO boats would more often than not get to shoot first. The russian tactic would be to fire snapshots and hammer away with active sonar before taking evasive action.

Any way to implement this into the game for AI subs? like I said before, I know nothing about editing in this game, im just curious and wanting a new experience.

____________________________________________
Active has become more popular with the advent of the Akula. It is wondrously quiet and often needs active to find.
Active is also sometimes used to make the final targetting as you implied.
And it is a way of saying "gotcha". As most of modern naval experience is cold rather then hot war, a means of doing this is advantageous. Obviously that is not the primary purpose but it is a use.