Question on navy search

[Arkham]

I understand the basics of search, in that each aircraft will do a ten degree arc of the defined search radius and fly both in the morning and afternoon. If I have more aircraft assigned than ten degree segments it will double up aircraft per ten degree segment (So 6 aircraft assigned to a 30 degree arc will have two aircraft fly 270-280, two flying 280-290 and two flying 290-300).

My question is, if I set my search radius to be pretty short, say only three hexes while the aircraft is capable of flying 9 hexes of search, will the same aircraft fly the pattern multiple times in each phase? IE, 3 searches along that 3 hex arc since it could do 9 hexes?

[RangerJoe]

I understand the basics of search, in that each aircraft will do a ten degree arc of the defined search radius and fly both in the morning and afternoon. If I have more aircraft assigned than ten degree segments it will double up aircraft per ten degree segment (So 6 aircraft assigned to a 30 degree arc will have two aircraft fly 270-280, two flying 280-290 and two flying 290-300).

My question is, if I set my search radius to be pretty short, say only three hexes while the aircraft is capable of flying 9 hexes of search, will the same aircraft fly the pattern multiple times in each phase? IE, 3 searches along that 3 hex arc since it could do 9 hexes?

No matter the range that is set, all search aircraft have a chance to spot any ships up to 4 hexes away.

There are also morning and afternoon search phases nor to mention the night search. So if you double up the aircraft so there are two aircraft per search arc, then the morning and the afternoon searches will be performed in those arcs.

I presume that a shorter range will allow a more intense search but I do not know if that is true.

[Arkham]

No matter the range that is set, all search aircraft have a chance to spot any ships up to 4 hexes away.

There are also morning and afternoon search phases nor to mention the night search. So if you double up the aircraft so there are two aircraft per search arc, then the morning and the afternoon searches will be performed in those arcs.

I presume that a shorter range will allow a more intense search but I do not know if that is true.

Navy search can only detect ships up to 4 hexes away from the base but not further? The manual says they have reduced chances after 300 miles from their base, so thats what, six hexes? (Each hex is 48ish miles, right?) Are you saying that only four hexes work any any search after that isn't going to succeed no matter what?

If that's the case then why would I want to run a mission past four hexes? If that's not actually the case then is there any benefit to NOT maxing out the range past 300 miles?

[btd64]

Depending on the situation I will put a 12 plane squadron on 50 % rest and search up to 60 degrees. Range depends on the location of the the squadron and the situation....GP

[RangerJoe]

No matter the range that is set, all search aircraft have a chance to spot any ships up to 4 hexes away.

There are also morning and afternoon search phases nor to mention the night search. So if you double up the aircraft so there are two aircraft per search arc, then the morning and the afternoon searches will be performed in those arcs.

I presume that a shorter range will allow a more intense search but I do not know if that is true.

Navy search can only detect ships up to 4 hexes away from the base but not further? The manual says they have reduced chances after 300 miles from their base, so thats what, six hexes? (Each hex is 48ish miles, right?) Are you saying that only four hexes work any any search after that isn't going to succeed no matter what?
Reread what I wrote, you apparently misunderstood my response.

If that's the case then why would I want to run a mission past four hexes? That is not the case. If that's not actually the case then is there any benefit to NOT maxing out the range past 300 miles? Yes, there is a benefit.

[Chris21wen]

I understand the basics of search, in that each aircraft will do a ten degree arc of the defined search radius and fly both in the morning and afternoon. If I have more aircraft assigned than ten degree segments it will double up aircraft per ten degree segment (So 6 aircraft assigned to a 30 degree arc will have two aircraft fly 270-280, two flying 280-290 and two flying 290-300).

My question is, if I set my search radius to be pretty short, say only three hexes while the aircraft is capable of flying 9 hexes of search, will the same aircraft fly the pattern multiple times in each phase? IE, 3 searches along that 3 hex arc since it could do 9 hexes?

Not sure where 4 hexes game from but I've always used 10 hex as the max viable range, this is through others tests. The shorter the range the more chance you have.

I'll stick my neck out here. I believe Endurance also plays a part with patrol boats with their large endurance flying at short ranges have a better chance of finding somting than a FP set to the same range. Exactly how the actual coverage works I don't care, it does, or appears too.

[Yaab]

Manual p.151 talks about 5 hexes.

" Searches within 5 hexes of the base are more likely to detect TFs as there will be a number of planes transiting to
and from the search areas."

[BBfanboy]

I remember the statement about automatic (i.e. no player set patrol) search within 4 hexes during any air operation by any aircraft type! Can't remember if that was from Alfred or another scattered gem in the manual. It is meant to abstract that aircraft crews flying to and from the base will have their eyes open for whatever is out there. Don't worry about it and and don't count on it - the chances of actually detecting something firmly seem to be small.

The effectiveness of search depends partly on the range because the search arc gets wider as you go out further from the base. We were advised by Alfred or a developer that the math was that a range of 12 hexes at 6000 feet was the max effective search because the aircraft can still see the entire width of the arc at that point. After that the chances of sighting decline. Add to that the greater chance of ops losses if your patrol aircraft is going closer to its max range and it is not worthwhile unless you have intel that something juicy is out there (like a sighting by one of your subs). Also note that flying higher might miss small targets like subs and single xAKLs or even single DDs.

Airborne radar or ESM equipment can extend the range of detection and, IIRC, it will give a +1 to the DL.

IME, a given aircraft rarely flies twice in the same day - only if the range is much shorter than its max, the weather is good, the air support is plentiful, and the crew is experienced. It will take Ops points to refuel the aircraft and feed/dewater/rest) the crew after their first mission so there must be plenty of time for that to happen.

Best chances of detection seems to be overlapping search arcs from multiple bases or carrier TFs.

[Arkham]

Thanks. I'm going to start splitting my NS units in half or maybe into thirds and have overlapping arcs. So 000 - 020, 020 - 040, and 040 - 060 of 4 planes each flight for example. That way each 10 degree arc gets two planes flying the same mission and the overlapping helps increase the odds of a spot.

[BBfanboy]

That is not what I meant by overlapping, but I have no idea if you plan will do anything better than the normal search orders. You can cover more vectors that way but you decrease the number of planes searching the full squadron vector and then increased them by having more parts of squadrons covering the same vectors.

Overlapping usually means coming from various directions so the arcs crisscross. That isn't always possible, but if you have that in mind you will plan to take more than one island in an area. The other benefit of multiple islands is mutual support from CAP, bombing enemy ships or airfields, and a place to land if runways one base get shut down.

[Arkham]

That is not what I meant by overlapping, but I have no idea if you plan will do anything better than the normal search orders. You can cover more vectors that way but you decrease the number of planes searching the full squadron vector and then increased them by having more parts of squadrons covering the same vectors.

Overlapping usually means coming from various directions so the arcs crisscross. That isn't always possible, but if you have that in mind you will plan to take more than one island in an area. The other benefit of multiple islands is mutual support from CAP, bombing enemy ships or airfields, and a place to land if runways one base get shut down.

Well, part of it is the mathmatics of repeating probabilities. If 1 plane has a 50 percent chance of spotting something, two planes working together have a combined chance of 87.5% of at least one plane spotting. I heard somewhere that planes can 'spot' hexes adjacent to them on their search path and not just the one they are flying over, so perhaps thats where I got confused by your overlapping comment. If they don't actually spot in adjacent hexes during a search pattern then having an overlapping pattern like I described doesn't give anything mathematically, so i'm better off doing a 00-20, 30-40,50-60 and so on of two flights each.

[RangerJoe]

That is not what I meant by overlapping, but I have no idea if you plan will do anything better than the normal search orders. You can cover more vectors that way but you decrease the number of planes searching the full squadron vector and then increased them by having more parts of squadrons covering the same vectors.

Overlapping usually means coming from various directions so the arcs crisscross. That isn't always possible, but if you have that in mind you will plan to take more than one island in an area. The other benefit of multiple islands is mutual support from CAP, bombing enemy ships or airfields, and a place to land if runways one base get shut down.

Well, part of it is the mathmatics of repeating probabilities. If 1 plane has a 50 percent chance of spotting something, two planes working together have a combined chance of 87.5% of at least one plane spotting. I heard somewhere that planes can 'spot' hexes adjacent to them on their search path and not just the one they are flying over, so perhaps thats where I got confused by your overlapping comment. If they don't actually spot in adjacent hexes during a search pattern then having an overlapping pattern like I described doesn't give anything mathematically, so i'm better off doing a 00-20, 30-40,50-60 and so on of two flights each.

I have a question. Why make more work for yourself?

I suggest that you do a search for "Naval Search" and read some threads on the subject.

[Arkham]

Because splitting search units into sub-units gives me more flexibility on which areas I am searching and which areas I can ignore, or have a more granular search.


Say I think the enemy fleet is coming from somewhere to the west, but it could be from the southwest but is more likely coming from the west-northwest. Lets also say that I have a 12 unit squadron of Cats with aggregate NavS skill of 63.

Now I could set up one search pattern and be done with it, searching from 250 - 000 to cover all the radials. My squadron will have 110 degrees of arc to search in the morning, with one excess plane going back to the 250-260 arc. I can't put that excess plane in the more likely area, which is somewhere along the 290 - 320 radial. I _could_ just narrow my search area to get more planes flying in the more likely approach, but that leaves me with arcs uncovered that the enemy might actually approach from. That is going to give me, at best, a 63% chance of making a detection along any radial.

Now, lets say instead that I break my 12 squads into 4 subunits of 3 planes. Two of those subunits are assigned the less likely avenues of approach, 250-280 and 330-000. That is the same search percentage as my first example. However, my other two units are assigned the critical 290 - 330 range. That's six aircraft going where the enemy is more likely coming from. Each ten degree arc will have two planes searching there, giving an aggregate success rate of 86.3%.

Basically I'm rearranging my forces to get the most probability in the expected areas of approach while not leaving my flanks unsecured. Considering how much clicking and work this game is, doing a few extra mouse clicks for this sort of search pattern isn't that much more work.

[Kull]

Many of the most experienced AE veterans do not use search arcs at all, and claim they get equal or better results without them. I still use them primarily so that I can hit the Z button and get a quick visual indicator as to which bases are running searches, and whether they are Naval or ASW.

Don't be too quick to assume you understand exactly what's happening under the hood.

[Arkham]

Many of the most experienced AE veterans do not use search arcs at all, and claim they get equal or better results without them. I still use them primarily so that I can hit the Z button and get a quick visual indicator as to which bases are running searches, and whether they are Naval or ASW.

Don't be too quick to assume you understand exactly what's happening under the hood.

How are they searching if they are not using the arcs? I'm basing my logic based on what I've read here and what I've read in the manual, that each 10 degree arc gets flown twice a day by 1 search plane, that multiple defined arcs are given an allocation of planes and if there are more planes than arcs then the planes will be put back to the first arc and allocated again until there are zero remainders.

Is that not correct on how search works?

[Platoonist]

I can only go by anecdotal information I see but when I don’t use arcs I seem to light up more search objects. And they're usually in the threat vector too.

[Arkham]

What do you mean by 'don't use arcs'? Do you just set it to be 000 - 000 for a full circle and search quantity to 100? Whats the range for the search you are using, the recommended 12 hexes?

[Kull]

Treespider was one of the guys working on the AE dev team, so he should know. I've highlighted the relevant points. Including the "4 hex" comment at the bottom.

If you set your search arc 0 to 0 you search all 360 degrees but with PENALTIES.

If you set your search arc smaller than 360 degrees then each individual plane will only search a 10 degree arc in each phase - morning and afternoon.

So If I have 6 planes searching... and i want to search 0 to 180.... I can't....as I can't cover everything unless I set the unit to search 360.

However IF I do set the unit to search the arc 0 to 180 this is what happens:

In the morning -

Plane 1 searches 0-10
Plane 2 searches 10-20
etc etc
Plane 6 searchs 50-60.

Then in the afternoon -

Plane 1 searches 60-70
Plane 2 searches 70-80
etc etc
Plane 6 searches 110-120


So with six planes from a unit searching I can search 120 degrees in a given day but only 60 degrees in each phase....or I can have the planes search 360 degrees with penalties.


EDIT : IF you have more planes than arcs - each arc is given one plane, then the first arc is given two, and the second two etc...until all of the planes have been assigned.

EDIT EDIT : This only applies to searches of greater than 4 hexes.

[Kull]

That post was linked in davbaker's "FAQ/Info for Newb's" thread which is stickied at the top of the "War Room" forum. Still just as relevant today as when he first posted it.

https://www.matrixgames.com/forums/view ... 6&t=167065

[Arkham]

Yeah, I read those posts but thanks for pasting them again.

So you are finding that not setting arcs and taking the penalty for a 360 search is getting you better results than setting arcs and focusing your search patterns? That's interesting. I wonder if that's because its doing multiple checks per hex for X number of planes. You might be getting a 25 percent hit per check, but if you offset that by multiple checks...


off to the probability calculator!

[edit] ok, ran some off the cuff numbers with some pretty big assumptions on how its coded. I'm assuming detection chance is based on the tonnage of the ship with bonuses and maluses applied based on skill, fatigue, leadership and so on. But for the sake of argument I'm going with this:
Detection chance of .5, basically a coin flip.
If doing a 10 degree arc one plane has a .5 chance of detection. If two planes then its a .75 chance. So each hex out to 12 has a .75 chance of getting a detection.

Now, lets say I'm not doing a focused search and instead doing the full 360. I'm going to apply a 50 percent penalty to the search, so .25 chance of detection. However because its 12 planes in the squadron each getting a roll, that comes out to a whopping .96 percent chance of making a detection.

https://www.statology.org/probability-o ... alculator/

[edit2] I tried testing this using the Coral Sea mission and was getting a number of detections. However I think the results were partially skewed due to coastwatchers raising the DL of the enemy TFs. Let me see if I can find another one to test with.