ORIGINAL: Alfred
Combat passes and related issues such as how long an aircraft may stay in the fight before leaving is one of the more opaque aspects of AE. Very little has been disclosed by the devs.
When shipped, the maximum number of air combat passes was 200. This was essentially an "arbitrary" number chosen by michaelm. Subsequently, circa 2011, this maximum number was increased to 300, another "arbitrary" number. Although I say "arbitrary" one must understand that michaelm had to take into account non direct combat considerations. As so much else in the game, it is an abstraction which needs to accommodate things like "loiter time", "fuel reserve", CPU processing demands, the design change from classical WITP of witnessing huge slugfests to the AE paradigm of raids, to mention only some relevant considerations.
The number of passes is the theoretical maximum number possible. Most air combat stops well short of the maximum. In part this is because each pass involves a flight and up to 8 aircraft can constitute a flight, so a theoretical maximum number of 2400 aircraft could participate in combat. Which means with absolute certainty, that sending more than 2400 aircraft guarantees the excess will not participate in combat.
However the number of 2400 is somewhat rubbery as I don't recall the devs ever clarifying whether the flight was cut out of the total number of aircraft present or out of the number of aircraft in each unit. IOW if 4 air units, each of 12 aircraft, are in combat is the number of flights 6 [(4x12)/8] or is it 4 [(12/8)(4)] if rounded down or 8 if rounded up. I suspect it is unit based and rounded down but I don't have access to the code.
Then of course there is the ubiquitous die rolls which can certainly reduce combat to much less than 300 passes.
A variable which I did not specifically mention in my previous post is pilot fatigue which can have a detrimental effect on pilot skill levels. Of particular relevance here is that flying aircraft at or close to their maximum altitude level will, in of itself, further increase pilot fatigue with a consequent impairment of combat performance in that combat. Where this additional pilot fatigue clicks in exactly is somewhat unclear as michaelm has in two separate posts stated that 75% and 80% of the maximum aircraft model altitude is the threshold for this additional malus.
The additional pilot fatigue brought about by flying too high complicates the organising of flight operations. Send two different fighter models, model "A" having a maximum altitude of 30k and model "B" being 40k, into combat at 24k will see the pilots flying model "A" accruing this additional fatigue whereas the model "B" pilots will not be so afflicted. Ceteris paribus, the "A" unit will perform worse (which can include RTB earlier than otherwise, missing their opponents, more prone to being shot down etc) in that specific combat.
The altitude fatigue is a deliberate design feature intended to minimise the unhistorical and unrealistic practice of many players flying their fighter sweeps at maximum altitude. When combined with distance flown from airfield to target, the greater the distance flown the more pilot fatigue is accrued, a substantial increase in pilot fatigue (whose new quantum is not disclosed to the player) can result in quite a poor combat performance which comes in well under player expectations. These are all design decisions intended to reduce the slugfests of classical WITP.
Another design decision is that even when sufficient escorting fighters are present, flying close to base with no fatigue problems and at appropriate altitudes, with excellent leaders etc, the algorithms are written to provide a chance for the CAP to eventually bypass the escorts and tangle with the bombers. The classical WITP feature that CAP had to defeat all of the escorting fighters before it could tangle with the bombers, does not exist in AE.
As to one air unit impacting on the performance of another friendly air unit, that is not how air combat is generally handled in AE. Individual units perform on the basis of their individual characteristics derived from their own leaders, pilot ratings, aircraft model stats. Of course if a Top Gun unit is present at the start of combat and it eliminates the best enemy aircraft, later arriving non Top Gun friendly units will probably find it easier to rack up kills against lesser quality opponents in that combat.
Alfred
Your contributions and explanations are greatly appreciated.
The only thing is you seem not to have mentioned anything to do with the cooperating sweep issue. This is the crux of the problem in this and other sweeps/strikes I've experienced with short combat duration.
This has nothing to do with loiter, etc, as the longer distance strikes did not behave this way and in previous situations the groups arriving as single groups in the replay do not suffer the short combat, while those (from the same base, same fatigue, same settings, same airframe type, and same pilot skill level and experience) have a short combat while cooperating on a sweep with multiple groups present.
It also has nothing to do with altitude, fatigue, etc, since as stated, (and shown in previous combat reports) only the groups with cooperating sweeps suffer short combats, and all groups are up high (but with low fatigue, high morale since they haven't been flown for days ahead of those sweeps.
The A6M8 escorts "sweeping" ahead is a real problem in this kind of combat, and although some of the strike got through, the CAP and A6M8 (and other dedicated escorts that did not sweep) simply didn't interact for more than a few passes. This is easily read by noticing the much lower relative losses of escorts to strike planes. I can also confirm that the CAP was not in the combat at full numbers since it was so short, so it would be rare (never seen it) for them to bypass the escorts so early to tackle a decently large strike of nearly 100 planes.
I know it's not going to be fixed in the code, but it is a problem. It's not a player control problem, not a settings problem, not a fatigue/morale or altitude problem. I've played enough to know it's a code/metrics/under the hood problem.
I am wondering if we have some differences in game setup that cause this. I did a complete new install between strikes though, and the nearly identical result occurred, so probably not.
If there is anything else anyone would like to see, I can probably dig that replay and settings out of archived dropbox saves and run it again. Maybe even put it up on youtube. I'll see if I can retrieve it.
