Slitherine

I am playing the grand campaign and am finding that the enemy, be it Russian or English, has loads of fighters. I have 4 ME 109’s, still early 1941, to try and gain air superiority. My tactic has been to focus on one enemy plane until it’s destroyed which usually uses up the turns of all my fighters. Should I be doing that or just weakening most of their fighters instead of destroying one? Any advice would be helpful.

The Polish and the French are the only countries that are weak enough to fall before your Luftwaffe. By Minsk, most players have 6+ fighters. My strategy (and this works! - at least in '44W) is to have 0 fighters, and just have around 8 Anti-Air. Keep 1 AA with every 3-4 units, and then the enemy won't be able to bomb anything, without taking significant damage. The Russian Airforce is to numerous to give your Stukas much chance, so why buy 3 of them every scenario? If I were you, I would disband your fighters and bombers at the start of Smolensk41, and then buy ~6 AA.

AA and Tigers = unbeatable core! (and 50k pp)

- BNC

Of course you have to adjust to the situation. If there are lots of bombers that are going to severely damage your troops, you should deplete them as much as possible. If it is a matter of Russian fighters coming over, it is best to take them out, even if one at a time. I generally use at least 6 fighters in Russia to quickly gain air superiority as I quite dislike the soviets annoying my troops.

Concentrating on a single aircraft can work against fighters. For dangerous opponents, I surround with several good fighters first, to get some nice mass attack bonus.

If I can, I park all my fighters next to a cluster of enemy aircraft so that most of them can target multiple opponents. Then I try to maximise the mass attack bonus against a single opponent, although not all fighters will attack that target.

So, for example an enemy fighter that is escorting a bomber comes in view. I place four fighters so that they all target the fighter, but two (with less experience/initiative) can also attack the bomber. As soon as the escorting fighter is destroyed (say it needs three attacks, but you profit from the extra mass attack bonus of your fighter) the remaining fighters attack the bomber.

This way, you can concentrate against dangerous opponents without wasting attacks. I even use Bf 110 fighter-bombers for this tactic, as their air attack will contribute to the mass attack and they can engage vulnerable bombers without too much trouble.

AAA (anti-aircraft guns) can also do a lot of damage, later in the war I usually have two, sometimes more, and they can be useful in taking a few points off before your fighters attack.

One of the most succesful tactics is setting traps for the AI, either a 'FlaK trap' or a fighter trap, to get cheap kills. But this requires good planning and knowledge about the AI's units. If you park an inviting target just within spotting range the AI will often go for it. Position covering AAA and/or escorting fighters just out of the AI's spotting range, and sometimes a fighter or bomber will blindly attack the bait, getting shot up badly.

Generally, like shawkhan, I try to attack bombers that aren't escorted and at least try weaken them, so they won't inflict too many casualties on my ground forces. This also often forces them to return to base to reinforce, instead of flying around hurting things. Hunting enemy fighters while his bombers are destroying your artillery can be a bad tactic. Attacking ground targets with your fighters can be dangerous, you often lose one or two points and it cannot provide escort/support. I prefer not to search for targets with my fighters, but make sure they stay just out of spotting range of the AI, covering as many ground units as possible, hoping the AI will give me a free attack when it doesn't see my fighter. This has given me many cheap kills on bombers.

One final 'air strategy' tip: try to manage your airfields. Sometimes it's very important to seize and hold an airfield that is close to the action. A lot of planes have short range and being able to refuel quickly means your planes will be more effective. You don't want to be flying far away from the frontline for two full turns to refuel, return, and already have half your fuel expended when it finally gets back.

Denying the AI airfields will force it to fly damaged planes back and forth for many turns, while you can keep your fighters close to your units. And try to 'rotate' them: which means that some of your fighters are busy refueling, while others are available for attack. It can be problematic if you have to refuel every plane simultaneously and the enemy shows up.

My tactic is a little different.
Stick with two fighters until France 40,then add one more.
I use Bf-110's through the whole game. In France I use one or two.

3 fighters are sufficient for me, but sometimes I will go with 4 late game.

I like a mobile AA one at least one 88 AA if not two.

One of the beauties of this game is that there are multiple paths to success.
Lots of different ways to get the job done.

Until 1.21 came in I used to have a larger air force, typically with 6 active fighters in most scenarios, and a couple of tac bombers. If I had a spotting hero I would do deep reconnaissance missions once the 109F gave them enough range.

Now I seem to be managing OK with three ME109, a 110, plus a Ju88 and a pair of 88s (which also provide excellent AT support until about the end of 1942). I will occasionally add an extra fighter - either to ensure I have a couple of experienced ones in reserve or to provide cover for a larger bomber force. While I like the attack stats of the Ju87, the extra spotting of the 110, the ability to contribute to mass attack, finish off damaged aircraft and bombers in combination with a reasonable ability to survive unescorted when needed make them a better choice.

Generally the fighters act as bomber escorts or troop cover - most importantly sitting directly over the 88s when they are in the half tracks to avoid stray tac bombers coming to play. I don't deliberately set traps for the AA very often - I'm not cunning enough to do this reliably - but a bit of planning can give good protection and keep your air cover and AA out of spotting range, which improves the odds of unexpected encounter, AI attacking protected units (because it didn't see the protection) or just being able to hit attacking aircraft with at least one AA and a fighter on my turn. As a general rule I like to finish off damaged units but when heavily outnumbered in the air, it can be more effective to damage as many as possible so that they return to base, repair then come back - with the odds more in your favour you can either kill off or weaken the survivors in the next two turns according to preference.

I haven't yet gone as far as no aircraft and a ton of 88's but I could see myself trying this in AK.

In the GC's I use Me-109's and Me-110's and the Rudel Stuka. I love the AA 88 and have at least one. If you go to GC 42-43 West you only can take eight units so I upgrade the Rudel Stuka to an Me-110 and take the best Me-110 and Me-109 that I have and the AA 88. In the later years you really can't get air superiority but you can defend and localize superiority during counter attacks and defense which are more prevalent in the GC 44 & 45

It greatly depends on scenario, campaign or mod that you are playing. Against numerically superior opponents (which is basically the only way AI can jeopardize good player) you have to have very experienced fighters so that AI doesn't try to attack them. That is why you have to have 3-4 fighters from early years in order to buff their experience. In my case 5 is ideal if GC is the case. Alongside fighters in later years no matter who you play against it is very wise to have several AA units. 88 early on (or Sdkfz 7/2), Ostwind later on (say 2-3), depends on the core size. Also in later years forget about escorted bombers. Conversion to Fw190G is the only feasible option, as they can both attack and defend themselves.

Know that it is impossible to cover all ground units against air attacks, so some losses are inevitable (as in real war). But try to lure enemy aircraft into AA fire first, and finish them off with your fighters. This way you will reduce huge losses on expensive fighters. It is never wise to attack La7 head on or Meteor. It takes patience not to push hard in every move, but to sometimes take more defensive stance and wait and lure enemy aircraft. Also try to make interceptions as they can significantly damage AI aircraft (put your fighters in the path of expected enemy aircraft movement but out of the spotting zones). Try to cover artillery and transports at all times, be it with your fighters, or AA or both.

The last advice is to be on the watch at all times because AI depends heavily on scripts and there is no knowing until you replay scenario when the new spawn of AI aircraft will come.

i was wondering how to setup air ambush (not the AA trap but to catch the enemy plane in transit). i tried to do that several times when i knew the location of enemy plane and the location of airfield where it should retreat to and i tried to block the assumed path, but it never worked intentionally so far. similarly, me myself i never got into what would look like intended air ambush, but crushed couple times very stupidly into planes over the enemy airfield.
is there any known algorithms for plane path?

LevV wrote: ↑Tue Feb 08, 2022 6:34 am i was wondering how to setup air ambush (not the AA trap but to catch the enemy plane in transit). i tried to do that several times when i knew the location of enemy plane and the location of airfield where it should retreat to and i tried to block the assumed path, but it never worked intentionally so far. similarly, me myself i never got into what would look like intended air ambush, but crushed couple times very stupidly into planes over the enemy airfield.
is there any known algorithms for plane path?

It's a bit tricky, because if there is more than one airfield in enemy hands, you need to observe carefully which airfield is currently used by the enemy, and even that is sometimes hard to predict. The only time when making traps for aircraft in flight is effective is when the enemy has only one airfield left. From there it gets simpler. Because the pathfinding system is pretty straightforward: there is a horizontal alignment, after which comes a vertical alignment. Because seeing some of the program code people make when cloning games like Panzer General you learn a lot, actually. A program algorithm always has an order in which it resolves an issue. So for instance in path finding, it looks for the location of the plane, then the location of the airfield (or more accurately, the current location of your cursor once you select a unit - which ultimately IS an airfield once you commit to that choice). So the algorithm will count the number of hexes in the x-axis, then the y-axis. This information is useful because if you have a diagonal path (the path from the airplane to the airfield is not a straight vertical line or a straight horizontal line), you can see how the computer calculates which path it will use by observing how the arrow is changing. You can see that it first aligns with the vertical (so the difference in x-axis hexes are counted), then the remaining hexes are all vertical down to the airfield.

For example: your plane is 3 hexes away from the airfield on x-axis, and say 10 hexes away from the airfield on the y-axis. Once you click the plane to move to the airfield, the plane will first move the 3 hexes to the right, then 10 hexes down. I then use this knowledge to find a path which the enemy plane will take to get to its airfield, and you can position your fighter in its way - but it must be out of any enemy vision range.

A clarification: The x-axis count uses a diagonal as well because of the way hexes align when forming a grid (there is no straight horizontal line - it's a zig-zag), and the way the algorithm takes the first hex that it will align it with the rest of the path. So if I use the above example x = 3, y = 10 hexes, the 3 hexes of the x-axis will also use 3 hexes of the vertical. So when the pathfinder reaches the y-axis, 3 hexes are deducted from the 10 hexes of the y-axis as well, which results in using the remainder of 7 hexes of the vertical. However, the fields being hexes and not squares, some combination can get a bit tricky.

Funny thing is, the animation of the airplane moving appears to be wholly diagonal, but that is really not the case. The model actually moves as I described above, but the animation does not stick to the hexes. I hope my explanation is somewhat clear.

Oh, and another thing: if there are visible obstacles in the path, say, when there is a visible enemy plane in the path, the pathfinder will logically find a way to bypass it. In ground units, this is where Zone of Control check is used (as a component of the pathfinder algorithm), but since the planes have no Zone of Control it only bypasses the hex where the enemy plane is located.

auda wrote: ↑Mon Feb 14, 2022 9:43 pm Oh, and another thing: if there are visible obstacles in the path, say, when there is a visible enemy plane in the path, the pathfinder will logically find a way to bypass it. In ground units, this is where Zone of Control check is used (as a component of the pathfinder algorithm), but since the planes have no Zone of Control it only bypasses the hex where the enemy plane is located.

thanks for your input. well, i finally just spent some time in scenario editor, modelling various situations and taking notes on results.
i would try to put it here in terms, which look more understandable to me.
since the game is hex-based, there are obviously 3 axis or 6 directions for each starting hex. i will name directions from 1 being vertical from plane to north and 2 being from plane to north east and so on clock-wise. the sectors between these directions lets name 1-2, 2-3, 3-4 and so on.
generally, as described above before me, the plane moves in broken 2-leg line (unless the destination doesnt need deviation from the original directional line or if there is an obstacle), in most cases first leg is as far as possible by one of the directions (1-6) from initial hex (hereinafter - IH), and second leg would be the shortest way by one of the directions from the turning hex (TH) to the final hex (FH). the change of angle between 1 leg and 2nd leg is always only 60 degree.
However not all situations are the same and oddly enough in some cases the pattern of movement isnt uniform.
i have considered three type of situations:
1. no visible obstacle.
for this situation, as said, the plane would first go in one of the IH directions as far as possible, to be able to go the direct way from TH to FH. for FH in sector 1-2 the plane goes first direction 1 to TH, then direction 2 to FH. for FH in sector 2-3 first direction 2, second direction 3 from TH to FH. same pattern 60 degree left hand side swing for sectors 3-4, 4-5 and 6-1. however for FH in sector 5-6 the swing direction is opposite - first direction for IH - TH is 6 , second direction for TH - FH is direction 5. that makes me think that moving in direction 5 (for example approach target in direction 3, return to airfield in sectors 5-6 or 6-1; or attacking target in sectors 5-6 or 6-1) is more predictable, because for two sectors it is known that plane will first go direction 5. same time direction 4 from IH seems never being used in situation 1 conditions.
2. visible obstacle on one of the directions from IH.
for FH in sector 1-2 the plane approaches the obstacle in initial direction , passes in on right hand side, continues flight the initial direction till TH and FH same as no obstacle situation. same patterns for FH in sectors 2-3, 3-4, 4-5 and 6-1. for TH on initial direction behind the obstacle the plane would go 1 hex left from initial direction, pass the obstacle in direct line on left hand side and then turn right to the FH when its adjacent. same pattern for TH on initial directions 1, 2, 3, 4.
for FH in sector 5-6 the plane approaches the obstacle and passes it on left hand side.
for FH on direction 5 behind the obstacle the plane turns right from IH, passes the obstacle on right hand side, turns to ini direction just after the obstacle and continues straight till the FH. for FH on direction 6 behind the obstacle the plane approaches it, passes on right hand side, continues straight parallelly to ini direction, and turns left whenever FH is adjacent.
3. visible obstacle one hex to the left from ini direction line.
FH's in all sectors approached same way as for situation 1. FH's behind the obstacle (from the original situation 1 pattern point of view) are reached via passing the obstacle on right hand side, i.e. in that case first direction for FH in sector 1-2 will be 2 (and not 1 as original), then turn left after obstacle, then continue straight in direction 2 till the FH. same pattern for sectors 1-2, 2-3, 3-4, 4-5, 6-1. for sector 5-6 again it gets weird, and obstacle is being passed on left hand side.

if you have read to this point i would suggest you to make your own experiments and make the simple schemes for above 3 situations on paper.
so far i can only make out that sector 5-6 and its adjacent directions are not uniform with other and must be treated with more care. i wonder what programming issue forced such a weird pattern for these directions.
the air ambushes are really more predictable to me now. I can even imagine that knowing the patterns you can trick the enemy in a very complex trap. but that would have to be state of the art trap, next level stuff for me.
the other issue is that in some cases, when the known FH for enemy is more then 1 turn flight away, the intermediate hex can be decided by the human in a special way (that is also a topic for discussion - how it must be decided to get to FH faster and to avoid possible traps). for AI 1+ turn path i also havent experimented yet.
also, seems the same patterns apply to warships' movement. my idea is to use the default swing pattern with fast long range detecting warships (SB in original game) to scout and get back to safety (for example in and out of bunker/radar FOW within one turn)
lastly it all depends on FOW you and the enemy have. over well guarded/patrolled enemy territory i would not suggest try playing such games and instead try to get back to home field asap and safely.

Unfortunately, I cannot confirm or deny if what you wrote is correct. I only wrote what I deducted from my experience playing PanzerCorps, and since the PC code is not open, I cannot say if you and I are correct. Perhaps some of the more knowledgeable forum users here can help you out.

However, I believe things should be much simpler. The AI uses basic functions to get a move range according to unit stats and terrain cost (that's just for ground units), a function to find a shortest path (which also takes into account all visible enemy units in order to avoid them), and a function to choose directions. The last function, I believe, is also used to determine which hexes will be used to bypass an obstacle. It uses cardinal directions in a clockwise fashion. At least, the code I've been reviewing does so. So in order to bypass an obstacle, the first available hex is used, the rest are discarded.