State of the Air War in AE

[obvert]

I have nothing real to add to this thread as a new player. I just want to say thanks to those are put so much time and energy into clarifying these issues for the players who have not been around here for years and who are just touching on some of the complexities being mentioned and discussed.

We all start this game thinking in succession, amazing! Then wait a minute, why are these things happening? Then, how do I even find out what is really happening? And at some point hopefully realize it's only through communication here on the forum, from sharing experiences and ideas, through learning the game by playing it MANY TIMES, that we will get close to understanding how it works. Could there still be changes? Sure. Does it all need fixing to enjoy playing the game? No.

I'm thoroughly enjoying my PBEM, and I'm consistently learning on my own and through reading what you all have to say. This thread has enlightened me to some things that are happening in the air war in a way that I think my opponent and I can use, can make work, and can accept that part of the combat system far into the future of our game.

So, just a thank you.

[PaxMondo]

ORIGINAL: Canoerebel

.... (as noted, I'm very happy with AE and getting involved in this discussion is above my pay grade).

+1

(Thanks Elf for the insights. And Alfred for your words as always. Brilliant writing.)

[Commander Stormwolf]

Basing gun acc on ammo and muzzle velocity rather than ROF (one of your wrong theories)

I am not a programmer, but a historian, aeronautical genius, bad amateur pilot, and progeny of SSBN designer (see if you can guess which one [;)])

My knowledge of the pacwar is based on the history that was written long ago. I have had many encounters with the veterans from the RAF in the Bob,
the Malaya campaign, and Bmr command

What is attempted is to separate what is history as it was written when the war ended - how it is rembembered by those who participated, apart from
how it is re-written today by artists who are more on the side of creativity than fact (however this is to be expected, as time moves forward, the knowledge of history will become more confabulated)


Since few of the posters have first hand experience with WW2 air combat, it is suggested to base
the concepts of the air combat on the words of those who did.

Jimmy Thatch --> complains about F4F-4 and the small ammo capacity "if you miss with 4, you miss with 8"
but the RAF staff thought it's a good idea to have 6 guns with only a little bit of ammo [8|]

Saburo Sakai --> says he shot down 70% of the enemy with 7.7mm MG because of the low muzzle velocity and small
ammo (60 rounds) of the 20mm... but IJNAF thought ammo supply doesn't matter [8|]

[Commander Stormwolf]

Posted this a few times, this is my rendition of what the A6M2-21 should look like in the game



Guns:

-7.7mm range incrased, effect decreased to 1 - it hits often but no serious damage
- 20mm accuracy massively decreased, effect increased from 4 to 5
- 10% penalty for gun accuracy due to heterogenous armament with large balistic incompatibility

MVR:

1-stage supercharger has good low altitude performance, but poor between 15-20 (50% penalty)
and cannot fight at above 20,000 feet - it was a historical fact.. that has been forgotten
A6M2 was unable to intercept B-17 flying at high alt and P-39 was unable to intercept G4M at high alt
due to 1-stage supercharger

Remember this is the year 2012, and 50 years from now history will be even less accurate
There is only one game I am conviced when these things were modeled accurately,
the flight sim Aces of the Pacific that was made 20 years ago with help from WW2 pilots

[Commander Stormwolf]

Here is the F6F for comparison.. remember Marianas Turkey shoot? How did McCambell shoot down 9 planes in one day.. without ammo?

F6F has 2-stage supercharger with medium alt performance -

6x12.7mm with 400 rounds each, homogenous armament (+10% acc bonus)

range and acccuracy increased to 500 yards, effect decreased from 3 to 2. M2 Browning hits often but not hard
(kills Zeroes and Judys)

[Commander Stormwolf]

As it stands in AE right now, the A6M2 and A6M3a are identical
when really they were not --> once the A6M3a was in service
the A6M2 was sent to training units... you just couldn't hit a corsair
with 60 rounds of low velocity type-99-1..

.. and even 120 rounds of high velocity type-99-2 wasn't enough
for the rookie pilots at Marianas (USN lost 27 hellcats only?)

Germans, Soviets, RAF, and finally Japan with the N1K and J2M
realized you need 20mm with at least 200 rounds so you can
use your tracers to adjust fire and do some deflection shooting

.. USN pilots had 40 seconds of fire and could blast away until they hit something

[Commander Stormwolf]

but this this the most important component of the Air Model - keeping Japanese quantity
under control by keeping the correct ratio of Steel versus Aluminum (about 50:1)

..to stop people from halting ships to use HI to build AC when this was not possible

[castor troy]

Stormwolf, can't you just bring all this into scenario design forum. You act like a bot, really.

[LoBaron]

ORIGINAL: Commander Stormwolf

Basing gun acc on ammo and muzzle velocity rather than ROF (one of your wrong theories)

I am not a programmer, but a historian, aeronautical genius, bad amateur pilot, and progeny of SSBN designer (see if you can guess which one [;)])

My knowledge of the pacwar is based on the history that was written long ago. I have had many encounters with the veterans from the RAF in the Bob,
the Malaya campaign, and Bmr command

What is attempted is to separate what is history as it was written when the war ended - how it is rembembered by those who participated, apart from
how it is re-written today by artists who are more on the side of creativity than fact (however this is to be expected, as time moves forward, the knowledge of history will become more confabulated)


Since few of the posters have first hand experience with WW2 air combat, it is suggested to base
the concepts of the air combat on the words of those who did.

Jimmy Thatch --> complains about F4F-4 and the small ammo capacity "if you miss with 4, you miss with 8"
but the RAF staff thought it's a good idea to have 6 guns with only a little bit of ammo [8|]

Saburo Sakai --> says he shot down 70% of the enemy with 7.7mm MG because of the low muzzle velocity and small
ammo (60 rounds) of the 20mm... but IJNAF thought ammo supply doesn't matter [8|]

I am not a certified historian, but I read books on history since I began to read. Many of them well founded works, some of
them are reccommended in several threads around the board.
I am not an aeronautical genius, simply because this is not how I would describe myself. I don´t believe I am a "genius" at
anything. But I think I know a lot about aircraft and air warfare.

Also I am quite sure of where my area of expertise begins and where it ends, and there I usually ask or let others talk.

I happen to know a lot about programming, because I work together with programmers on a daily basis.
And I am pretty sure I am able to combine my knowledge on history with my knowledge on programming.

Your expertise ends with programming, and you make obvious mistakes when trying to implement your (partly incomple) knowledge on
WWII into the game mechanics, and also obvious mistakes in the results you expect through this.

And this is what you do not seem to understand.

[Commander Stormwolf]

http://www.youtube.com/watch?v=Jnq6ak-I ... re=related

watch at 3:00 minutes ~ emperor nero fixing the artwork on the ground [;)]

PacWar is my passion,
.. the curse of a man who has never toiled, never voted, blessed by his majesty king george V to keep
the empire's savages in line.. collect interest in banks.. and like nero... produce art [;)]

my thanks to the AE team for making this world possible - i stay on the floor with my scalpel
changing the art on the streets of rome

[Hoplosternum]

LoBaron, TheElf, Alfred et al while I agree with a lot of what you say Greyjoy came across a clear bug in the code. You cannot seriously argue that the Escorts over a certain level making the bombers immune to CAP was either intended to work like that or has any basis in reality?

Lets start with what we probably agree on.

WitP AE has a much better air model than WitP and I appreciate the work done both before and after the release of AE.

The model works best with historical numbers being used historically. You don't have to use them in the historical areas, just avoid packing everything in to one area and then expecting the game engine to work pefectly, and things will go much smoother. For those that grumble many things can be house ruled. If you find multiple PT TFs in a port hex make the port immune to bombardment TF and you don't like this then House Rule it. No need for code changes to 'fix' it.

We can all see the Raedar/Greyjoy game pushed everything to the extremes and the fact that the game engine struggled was only to be expected. But they came across a bug that affects both sides and is all but impossible to avoid late war. And could not realistically be avoided by house rules. Even if Greyjoy was not battering on a very narrow front just off Japan it would be hard to avoid this in any hard fought over bit of real estate late war. There will be plenty of strikes with over 200 escorts. Even if you HR limit the number of escorts from one base/hex, coordination can occur from multiple bases/TFs.

The problem is that giving the bombers absolute immunity from any size of CAP be it 20 or 2000 (and regardless of the bomber strike size) just by having 200 escorts (and they don't have to be in anyway capable of fighting the CAP - just be there as targets for the CAP) will have big affects. And whether people divide up their CVs, attack in multiple areas or on a narrow axis they will come across this as plane densities increase in the late war. There are enough other people reporting it in the late game now for us to rule out it being a quirk of Greyjoy's experience and testing I think.

You may dispute that this is a real issue and have plenty of data or experience that large escorts don't block CAP (of any size) from getting at the bombers, thats fine. And if you want to quibble about the exact numbers thats fine too. My guess is that there are enough random elements that the number for bomber safety is not fixed. Or if you want to say that it is just a late war issue and so not that crucial I can accept that. And there may be no solution if code changes are needed and there is no one available or willing to make them. We will all have to live with that if thats the case. But to claim it is working as intended seems bizarre. It is neither realistic or a good game mechanic. It rewards one sides concentration while punishing the other for doing the same thing.

If you would accept there was a problem then there are potential fixes. Yes they would require code changes but they need not be ones that will affect other areas and so need not cause lots of unintended consequences.

Here are my two possible solutions. They both are based on an assumption about how the code works which the game suggests but may not actually be the case. That is that there comes a point in the air strike routine where the numbers are known on both sides BEFORE the combat is resolved. I believe it flashes the numbers up on screen before the combat replay is played out. At that point you insert some extra code which resolves the problem.

So you have all the code which determines who is coordinated and finds the target and the CAP appearing and gives you the numbers - say 240 Escorts, 480 bombers and 360 CAP. The new code then applies a limit to the strike size and simpley reduces the strike to that size. Lets say the limit we think the engine can handle is 300 we simply cut everything down to that in the ratios we have. So we would end up in this case with 100 escorts, 200 bombers and 150 CAP. The rest simply don't take part. We know the engine can handle that size of combat without uber CAP or the greyjoy bug issues.

Some might consider such a reduction in strike size too drastic, although I like it as it would discourage the fantasy mass air battles the current game model encourages. So as an alternative you put in code that breaks up the strike in to multiple waves and fight each one seperately. This would be harder to code as you would have to store the different waves (and results) and change what you display in the combat replay. Basically you would run multiple air battles in a series. Using the figures from my first example the total strike size is 720 (240+480) vs a maximum strike size of 300. So you split it in to 3 and run three air battles on the existing model. Each with 80 escorts, 160 bombers and 120 CAP. Even if CVs or air bases are put out of action on the first raid this should not affect the CAP of later waves. They are already up in the air and committed. They just take part in a different wave of combat.

Either of the above systems has the advantage of making sure the air combats are of a size the engine can handle. They are both relatively straight forward (assuming the code currently reaches apoint where it has determined the strike/CAP sizes but has not yet begun the combat). And because they don't interfere with the other mechanics - coordination, fatigue, altitude changes, number of passes etc. - there should be little chance of breaking other things. And we already know combat works for these numbers.

[vettim89]

JOMHO, but I think this thread is diverging from the OP's original intent.

First, I think the AE Team did a bang up job on the game. While there certainly things that are abstracted and quirky, the game does what it is supposed to do remarkably well. Is it perfect? No, it is not. That said, there is no other game like it that I have ever encountered.

With that out of the way, lets look at the initial question: have some things been introduced into the code that have created problems that were unanticipated? For all the talk about "legacy code", the truth is that the original WITP was altered for AE as far as the air combat routines are concerned. The introduction of pilot skills as well as experience, the altitude based maneuverability ratings, and the new CAP routines all indicate that the code was indeed altered. Now whether, the actualy guts of the system was changed vice new subroutines were added is a question for those that know. I suspect it was the latter. My point being that these facts disprove the premise that we are stuck with the legacy code because any change to the core program would require a complete rewrite of the code. I do concede that some of the things we see now may have always been there but hidden from us. I would specifically be referring to things like raid detection range, number of fighters in the air versus scrambling, etc.

From the day AE was released, there were concerns about how CAP subroutines were working. If you recall, many of us fired up the Coral Sea scenario immediately after installing the game. There were numerous posts about the ineffectiveness of CAP from those initial few days. Now some of that may have been the process of learning the new system, but I think the point could be made that there was some concern from the beginning. I personally remember posting something along the lines of "we have gone from Uber CAP to Useless CAP." Now after playing the game for a few years, I am not sure if things are or were nearly as bad as we first thought. One thing I can definitely say is setting an effective CAP in AE is much more fine grained with every little setting coming into play. I think some of the "faults" in the system are really just poor understand of the game sometimes combined with poor play.

The only thing that has been brought up that I would agree with is the "ablative armor" effect of Escorts. I have seen this in my own games and I think it is worth further investigation. I won't repeat myself completely, but I still think this has to do with the code "piece mealing" the CAP but allowing the escorts to act as a combined, very cohesive unit. Again repeating myself, if the goal of the AE Air Combat routine was to create leakers, then it should go both ways. If bombers can leak past even numerically superior CAP, then some CAP should be able to leak past the escorts. This should certainly be true if the CAP has a vast numerical superiority. Putting some sort of balance check into the CAP routine would not be that hard. Is it really too much to ask of the system for there to be a point where the escorting fighters would be overwhelmed by the CAP? I think this may actaully be in the system now going by some combats I have watched. Perhaps it just needs to be tweaked.

[LoBaron]

Vettim, Hoplosternum:

Noone denies that we have an ablative armor effect of cheap escorts as soon as planes imvolved surpass a high but reachable threshold.
I also think that there are already some ideas around here which justify further investigation, keep in mind that ressources here are an issue.
Personally I am unable to invest much time into such a topic, and this probably applies to everybody else who needs to work for a living, including
devs like Iain.


But what astounds me a bit is, that only few notice this ablative escort effect comes at a pretty high cost.

The cost is:

1) High airframe usage of the player who implements this kind of method
2) Similary high usage of the pilot training programme
3) With the above the occupation of a large part of the Japanese industy, this is why there are more complaints in scen#2 than in scen#1
4) A decline in fighter pilot quality, either draining the pool of trained pilots or draining the recruit base for trained pilots, both effects are negative
5) A low sustainability against stubborn fighter defense as CAP is usually in advantage to escorts
6) A dependency on high-value or easy-to-hit targets in order to hold some balance between benefit and drawback of this strategy

If you look at the list above there are a lot of easy solutions to make such a strategy more costly and less damaging. This would be:

1) Defend from interlocking mutually supportive bases (use dispersal)
2) Stay flexible and engage the enemy fighter concentrations where they are based and not exclusively when they attck, this way you add another dimension
to What Alfred designates as "degradation of enemy force structure and pinning of the enemy forces "
3) Minimize the exposure of high value targets in the contested area to remove the benefit from ablative armor tactics
4) Cut the ressource flow where possible to drain the basis of a ressource intense strategy
5) Add as many new theat axes as possible, to force dispersal of enemy forces
5a) If reaction is not like expected in 5), engage the opponent where he chooses not to locate his force concentration

There are many more suggestions how to mount effective counters, but these would venture into tactical details, and there are lots of other threads on this topic.

An opponent playing such a strategy against me might soon have second thoughts on the cost/benefit balance.


I think that many of those experiencing issues are trying to mount a higher operational tempo than it is feasible in their situation.
This has always been a game which teaches patience.

A war must be won by moving forward, true, but it is true at the same time that the prerequisites to create a situation from where
winning is possible does not neccesarily include moving forward. And it does under no curcumstances include any imperative to move
forward at this specific moment in that specific area.

Time and location is the freedom of the player to chose. Both are often wrongly chosen.



I find this an extremely interesting an productive discussion by the way.

[Rainer79]

While others have already pointed out that fiddling with the whole CAP vs. sweeps vs. escorts balance may introduce unwanted side effects, on the top of my personal wish list regarding air combat in general would be something else.

I would like to see bombers attacking land targets to have a chance to miss their targets completely. In history bombing the wrong target was quite a common occurrence. There is already such a functionality regarding naval strikes so hopefully it wouldn't even be that hard to code.

In the game though long drawn out battles like the reduction of Rabaul can't happen since the airbase will be closed after the first massed assault. Sure, allied losses may or not be pretty bad during that day but once the damage is done the base will never be able to reopen if even some modicum of pressure is kept up. And daily milk run to do just that can be flown pretty easily.

Just my € 0,02.

[janh]

ORIGINAL: LoBaron
I think that many of those experiencing issues are trying to mount a higher operational tempo than it is feasible in their situation.
This has always been a game which teaches patience.

A lot of strange outcomes simply come from "strange play". Certainly one could close all of these "loopholes", and a huge lot were already closed in the transition from WitP to AE. There are evidently some remaining, but most recurring ones were in my opinion successfully addressed by Dababes mod (SuperEs, AAA etc.). If resources were unlimited, it would be a worthy task to port them over to stock and fix the also other loopholes. But in the end, even if you did that, people would still play the game "unhistorically, irrationally, unprofessional" .-- you name it -- manner. Well, people will always play the way they like it, there is nothing to stop them from it other than learning how to perform operations etc. "the proper way" -- which, given the overall accuracy that this game achieves, is probably close to the "book". But I suppose most of us are just not professional staff officers of any sort.

Even then, this game -- like any -- is missing one key factor, which in my opinion leads to what LoBaron mentioned: high op-tempo. This game has no consequences, and allows to take risks and be bolder than a RL counterpart would be. This is actually one of the benefits why people play games, to change history, and see what risks are "manageable", or how "elegantly" we can beat the opponents. Yet also this high op-tempo leads to similarly "strange" situations, even for the "best" players who otherwise "play by the book" in very reasonable style. In the end, it is "only" a very realistic, detailed game, or simulation, depending on what definition you prefer.

Regarding the ablative armor effect: it usually takes a while to accumulate the airframes, Nates are too short legged later, but enough spare Oscars will only be available by mid/late 43 with no usage other than kamikaze missions. It is a one-shot thing, but sometimes loosing rookie pilots is what you want as IJ players since the pools can be so full with no button to release them, that they are just a drain on HI at the later stages (there are AARs indeed...).
Done with initial key attacks, like on a 4EB base, or an LBA on a amphibious attack and its carrier support, it might give you the lucky hit you need before the following attack from KB. Even small strike packages would be enough? As far as I toyed with it (against AI though), it works well, and since the IJ used kamikaze attacks, why not allow kamikaze escorts? If it is not a (serious enough) bug, it is a feature. If it were a bug, it ought to be addressed, right?

Yet this 200 pass limit for attacking CAP flights, where does it come from, and what exactly is the rationale behind it? On first look it seems arbitrary. Was it just to make air combat tractable? Probably not.
Is it a representation of duration and the spread out character of large scale battles? I will pick Europe for a guesstimate, since I can find info quicker for that theater for large scale situations: a "1000 bomber raid" in Europe would take some ~60-90 minutes over the target, a little more if you consider a larger radius for intial interception or pursuit. The typical time a fighter, say a Bf109, engaged was about some ~30 mins. So it "never" could have made passes at all of the bombers, which came in at ~300 per 30 mins. If this pass limit ought to mimic that, then I suppose the "200" should be the either a maximum or average rate (under adverse weather or coordination issues etc) at which planes pass a CAP sector? If it were, should that be a constant, or depend on weather, plane speed etc?
That brings up a subsequent question: Is the AE air model so detailed that it would be able to mimic European type raids of realistic scale, or would it fail just as AE is not meant to recreate European tank warfare?

A very interesting an productive discussion this really is. Probably no war ever fit a purely attritional scheme, nor was there probably one purely aimed at the enemy ability to wage war or enforce his "plans" by means of military force, though? Sounds like the usual academic question?

[Miller]

Coming very late to this one, but my view is the "Bombers always get through with enough escorts" problem can be negated to an acceptable level using the various work arounds already mentioned (splitting TFs into multiple hexes, attacking on a wide front etc...)

My big concern is the uselessness of the escorts in any capacity other than bullet sponges. 70 exp pilots in Franks are just as worthless as 30 exp pilots in Oscars.......well maybe the kill ratio will be 1:20 instead of 1:100 lol. As has been said though, this seems to be a very difficult problem to address without making potentially harmful changes elsewhere in the air model. We will probably have to live with it, AE is still a great game regardless.

[btbw]

ORIGINAL: Commander Stormwolf

Posted this a few times, this is my rendition of what the A6M2-21 should look like in the game


Guns:

-7.7mm range incrased, effect decreased to 1 - it hits often but no serious damage
- 20mm accuracy massively decreased, effect increased from 4 to 5
- 10% penalty for gun accuracy due to heterogenous armament with large balistic incompatibility

MVR:

1-stage supercharger has good low altitude performance, but poor between 15-20 (50% penalty)
and cannot fight at above 20,000 feet - it was a historical fact.. that has been forgotten
A6M2 was unable to intercept B-17 flying at high alt and P-39 was unable to intercept G4M at high alt
due to 1-stage supercharger

Remember this is the year 2012, and 50 years from now history will be even less accurate
There is only one game I am conviced when these things were modeled accurately,
the flight sim Aces of the Pacific that was made 20 years ago with help from WW2 pilots

After your fix i wish GOOD LUCK to find japanese opponent.
7.7mm type 98 MG used for tracer. Game never can simulate real air combat when pilot can use his cannons only when enemy fill out gunsight. Game strafe out any weapon which have enough range.
I dont know who said you about 20k feet. Because it was 30k feet. And Zero can intercept B-17. Of course it fortress but still can be downed.
Please learn more about war history and forgot your alternative no-ideas-what-i-talking.

[Commander Stormwolf]

7.7mm type 98 MG used for tracer.

7.7mm was the primary weapon (ask the Japanese pilots), and it was found to be pretty useless against armor.

Game never can simulate real air combat when pilot can use his cannons only when enemy fill out gunsight. Game strafe out any weapon which have enough range.

90 exp kido butai pilots did that. 30 exp pilots at marianas fired at long range and hit nothing.

I dont know who said you about 20k feet. Because it was 30k feet. And Zero can intercept B-17.

A6M2 could not intercept B-17 at midway. Only A6M3 could intercept B-17 at 30,000 feet at Rabaul,
with 2-speed supercharger.

Remember it is supposed to be speed that changes at altittude, not MVR. Just because a plane can stay in the air at 30,000 feet without stalling, doesn't mean it can fight at this altitude.

---------------

Remember, if you could put 500 rounds per gun for 20mm ... you could shoot down 9 planes in one day.. just like the Hellcat did [;)]

[Commander Stormwolf]

Now here is a good fighter.. when you raise the muzzle velocity from 1970 ft/s to 2560 ft/s
and increase ammo from 60 to 250 rounds.. and put those 20mm guns on the cowling.. now you have some real accuracy

[PaxMondo]

According to the docs I have read, and the basis of its design the MaDeuce, the Ho-5 did not sync well with the prop.  So, in a CL configuration, your ROF drop by 75%.  This is in contrast to the German MG 151/20 which did lend itself quite well to prop sync.
 
If you're going to put the Ho-5 firing through the prop, you need to account for the ROF.