Triplea: During NCM TripleA allows releasing planes (land them) in territories that have not been friendly at the start of the turn - with incorrect consequences

[panther2]

How can the problem be recreated?

During NCM land planes in just captured territories. This is not allowed, the engine should prevent that at that time, as it did in former releases.
An error message is issued, the planes are crashed when neglecting that. This is an incorrect consequence.

Do you have any ideas for an expected fix?

During NCM planes may only land in territories that have been friendly at the beginning of the turn.
So the engine should not allow the players to release the planes (land them) in previously hostile territories at the time of moving them during Noncombat Move Phase.

Attach a Save Game

Example from v3:
IncorrectNCM.zip

Both planes land in territories that have been just captured.

If playing a prerelease, which version are you using?

2.1.20738

Game Version: tested with current versions of v3 and v5

If playing a prerelease, does this happen on the latest release?

Yes, it is reproducible in the current stable.
Also retested with 20786.

Is there anything else we should know?

During NCM planes may only land in territories that have been friendly at the beginning of the turn.
Should be fixed with regards to #7246, IMHO.

[DanVanAtta]

I wonder what I might be missing. I loaded the attached save game (thank you!) and clicked done for the planes to land. The game engine did inform me that the planes could not land and they both crashed when NCM was completed:

[Cernelius]

Off topic, as probably virtually everyone knows, you should not have the option to let air units crash if you can land them all. Just pointing out.

[panther2]

@DanVanAtta I agree there is an error message following and the planes can be incorrectly crashed.

But: In prior versions the engine even prevented the player from moving and placing the air units in(to) those territories already during Noncombat Move when selecting and moving them - the same way as other units that try to move incorrectly are handled. That is the problem here.
I see I did not explain that deep enough, sorry.

I have adjusted the initial description accordingly.

[Cernelius]

But: In prior versions the engine even prevented the player from moving and placing the air units in(to) those territories already during Noncombat Move when selecting and moving them - the same way as other units that try to move incorrectly are handled. That is the problem here.

@ron-murhammer @DanVanAtta @RoiEXLab A regression is a regression with respect to what? I assume it should be with respect to the current release, which is 2.2, that has the exact behaviour described at this issue. So, how can this be a regression? For example, if something worked correctly in 1.8.0.9 and doesn't in 2.2, would this be a regression? How about with respect to a TripleA version released over 10 years ago?

@panther2 I tested your save-game with the current TripleA 2.2 release and a new game using TripleA 1.8.0.9, in both cases during Non Combat Move only.

What TripleA 2.2 appears doing is not allowing you moving any air unit into a zone where and whence it can't land, while allowing you moving the same into a zone where or whence it could land: You can move an air unit into a zone where it cannot land only as long as the unit has enough mobility left to reach a zone where it can land. Then you can decide not to move the unit to a zone where it can land, making it crash (which is against the rules).

What TripleA 1.8.0.9 appears doing is not allowing you moving any air unit into a sea zone where and whence it can't land and not allowing you moving any air unit into a land zone where it can't land, while allowing you moving the same into a sea zone where or whence it could land and allowing you moving the same into a land zone where it can land: You can move an air unit into a zone where it cannot land only as long as the zone is sea and the unit has enough mobility left to reach a zone where it can land. Then you can decide not to move the unit to a zone where it can land, making it crash (which is against the rules).

Therefore, I would say that TripleA 2.2 appears to have extended the sea-only behaviour of TripleA 1.8.0.9 to land too. If regressions can be traced back beyond the current (2.2) release, I'm not sure this can be counted as a regression, since both behaviours are not fully rules compliant (TripleA 1.8.0.9 allows you moving into a zone where you cannot land only as long as the zone is a sea zone and the units has enough mobility left still to reach a zone where it can land, while TripleA 2.2 extends this to land zones).

TripleA 1.9.0.0.13066 appears to have the same behaviour as 1.8.0.9, on this matter.

[panther2]

The movement behaviour of the situation I pointed out with the savegame above and its consequence have become worse than in 1.9 (and prior). That is what I call regression.
During 2.x development some code change for whatever reason must have led to this current behaviour - which should be reverted, IMHO.

[Cernelius]

The movement behavour of the situation I pointed out with the savegame above and its consequence have become worse than in 1.9 (and prior). That is what I call regression.

The "regression" label literally says: "A problem introduced in pre-release and not present in latest release".

That is an unnecessary wordy way of saying "A problem not present in the latest release".

The current latest release is 2.2.20790:
https://github.com/triplea-game/triplea/releases/tag/2.2.20790

Therefore I believe "regression" can only mean "A problem not present in TripleA 2.2.20790".

I understand we all agree this problem is present in TripleA 2.2.20790.

[tvleavitt]

I think the presumption of that description is that these problems would be
identified prior to a new release. The label probably should say "not
present in prior releases", as obviously they can be identified after a
release is out.

Thomas Leavitt

On Fri, Aug 21, 2020, 8:38 AM Cernelius [e-mail] wrote:

The movement behavour of the situation I pointed out with the savegame
above and its consequence have become worse than in 1.9 (and prior). That
is what I call regression.

The "regression" label literally says: "A problem introduced in
pre-release and not present in latest release".

That is an unnecessary wordy way of saying "A problem not present in the
latest release".

The current latest release is 2.2.20790:
https://github.com/triplea-game/triplea/releases/tag/2.2.20790

Therefore I believe "regression" can only mean "A problem not present in
TripleA 2.2.20790".

I understand we all agree this problem is present in TripleA 2.2.20790.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
https://github.com/triplea-game/triplea/issues/7350#issuecomment-678355719,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/ABB5HAJAEJK4WPOL5LDJGZTSB2IGBANCNFSM4P3C6RMA
.

[panther2]

Also let us not forget that the difference between "latest release" and "pre-release" is just a tag in Github with some additional communication on the forum(s)/website.
It looks like we are still moving from 1.9.0.0.13066 heading for a "really stable" version 2.X.

[panther2]

I erroneously hit "Comment and Close Issue", so @DanVanAtta please add that back to the "Release 2.3" Todo-list. Thank you.

[Cernelius]

I think the presumption of that description is that these problems would be identified prior to a new release. The label probably should say "not present in prior releases", as obviously they can be identified after a release is out. Thomas Leavitt

So, again, if something worked correctly in 1.8.0.9 and doesn't in 2.2, would this be a regression? How about with respect to a TripleA version released over 10 years ago?

If, maybe, you intended to refer only to a release immediately preceeding the current release (I also assume that, in this case, regressions would be current problems that didn't exist in the latest release or in the release immediately preceeding the latest release or both), we would need to test if this problem already existed in TripleA 2.1. Moreover, this formulation would have the challenge that you need to find out what was the release immediately before the current one. @tvleavitt, did you intend to mean this or not?

[Cernelius]

The regression label is a useful marker because it has a very clear and obvious utility in that (as long as its meaning is understood by everyone, which doesn't seem to be the case at the moment) it answers the question "are we adding any problem by releasing the current latest TripleA version". As long as this issue is concerned, it seems that the answer to this question is "no", since this problem is already present in the current release. That is why I believe the "regression" label should be removed from this issue.

[DanVanAtta]

I would tend to agree with @Cernelius here ; if we have a pre-release that introduces a problem, it is a regression. If that pre-release is released, then the regression then turns into just a problem.

[DanVanAtta]

I'm not quite clear on the expected behavior. Was it not the case before that this was basically how it worked? Would the game engine not be a bit too restritive to say "no, you can't move those units there". Perhaps someone has a fleet of planes and needs to move them so they can move other planes.

Is the rule truly, you must land planes? I know the rule is you can only make movements such that planes can land, but wouldn't a player be entitled to say "meh, they call crash". I can't think of a scenario where that would ever benefit someone, perhaps if there were a capturable plane of some sort, but then we are getting outside of the core wWII rules.

I think what would be best is a description of when this used to work different and the expectation of what should be done instead to get there.

[Cernelius]

I'm not quite clear on the expected behavior. Was it not the case before that this was basically how it worked? Would the game engine not be a bit too restritive to say "no, you can't move those units there". Perhaps someone has a fleet of planes and needs to move them so they can move other planes.

The program used to have the weird and inconsistent behaviour that it would always forbid non-combat moving into enemy land territories (that are land zones where you cannot land), while moving into sea zones where you cannot land was allowed as long as you had enough mobility left to go somewhere you can land. Now the program behaviour is consistent for land and sea, as what used to apply to sea only applies now to land all the same, as far as I know. That is why I'm not even sure the current behaviour would be a regression, anyway (just two wrong different behaviours).

Is the rule truly, you _must_ land planes? I know the rule is you can only make movements such that planes _can_ land, but wouldn't a player be entitled to say "meh, they call crash". I can't think of a scenario where that would _ever_ benefit someone, perhaps if there were a capturable plane of some sort, but then we are getting outside of the core wWII rules.

You cannot let planes crash if you can land them. This rule goes as far as obliging you moving carriers to allow your planes to land (another thing that TripleA fails to oblige you to do). Maybe you should be allowed to at least if the "Kamikaze Airplanes" property is true, however? I'm not sure about this.

I think what would be best is a description of when this used to work different and the expectation of what should be done instead to get there.

I'm not sure if going back to the previous behaviour would be much of an improvement. It was certainly more rules consistent for land zones, but I don't like the inconsintency of enforcing the same matter differently for land and sea, also since this can be more confusing. I suggest that the behaviour of disallowing going into zones where and whence you cannot land is enforced for both land and sea all the same, not for land only, as it used to be the case (currently it is for neither).

[panther2]

This issue is about landing in territories and the current implementation is definitely a step back compared to 1.9 and prior.
I am really wondering when and why it has been introduced...

@DanVanAtta always reminds me to keep the issues small and as "step by step" as possible so I suggest to open a separate issue for landing in sea zones. The rules systematics for territories and sea zones are different anyway.

[DanVanAtta]

@panther2 I was unable to get 1.8.0.9 to launch on my system : (
My understanding is the previous behavior is the move was disallowed. If we do change that, could we be adding a bigger hindrance in some cases? Perhaps slightly a devils advocate argument, I wonder for example, let's say someone is trying to set waypoints to cross territories on their way back to friendly territories and the map is too large to see the final territory. So specifically a case where you have to land in a territory that is not on-screen.

[Cernelius]

Actually, I just tested that this problem appears to be the same in 1.9.0.0.13066.

Reproduction steps:

• Select the "World War II v3 1941" game.
• Enable "Low Luck" in "Map Options".
• Send everything in Poland to East Poland.
• Send everything in Bulgaria Romania to Ukraine.
• Verify that, during the subsequent Non Combat Move, you are allowed to move the fighter from East Poland to Ukraine, then letting it crash.

So, @panther2, what's the difference? It seems to me that 1.9.0.0.13066 and 2.2 work exactly the same for everything detailed at or related to this issue.

As far as a fix goes, I reiterate the program should just not allow you ending the Non Combat Move phase if you have one or more air units that will crash and, at the start of the phase, you could have avoided them crashing without causing an equal or higher number of other air units to crash (this is referring to the case you haven't enough carriers to land all your fighters, so you can choose which ones to land, but only amongst moves that allow you to land the highest possible number).

---

During NCM land planes in just captured territories. This is not allowed, the engine should prevent that at that time, as it did in former releases.

I disagree with this wording. Moving and landing are two different actions, that happen in two different phases. The wording should be something like "During NCM end movement of planes into newly captured land territories. This is not allowed, the engine should prevent that".

[Cernelius]

Also I see no differences between 1.8.0.9, 1.9.0.0.13066 and 2.2 on everything detailed at or related to this issue. Sorry for the fact that I said differently previosuly. I'm thinking I just got somehow convinced that anything changed, maybe because of all the "regression" argument. Let me know if I'm missing something.

[panther2]

I have to apologize.

What stunned me was that when moving/landing the plane from East Poland it was not blocked like this:

... but the engine allowed to move and release here (despite the territory has not been friendly at the beginning of the turn):

For any reason it was in my head that this is a new unwanted behaviour - and as I have a new computer with only the prerelease installed I missed to test against older versions. I was so damn convinced that this was new - as usually you would not even think about moving a plane and releasing it like this. I just stumbled accidentally over this when playing and testing around.

That leaves the issue as is - but - as @Cernelius has tested and pointed out - not as a changed behaviour from prior releases.

My apologies again. Maybe it is about time for me to retire.

[Cernelius]

Well, on @panther2 defence, the behaviour itself makes almost no sense at all, since there is almost no reason to allow you moving into newly conquered/liberated land territories but not into enemy owned ones, as they are just both zones where you cannot land. Thus it's confusing that the engine handles them differently.

I said almost because I believe the only case in which you may want to non-combat move air units into newly conquered/liberated land territories is when having v1 (III edition only) paratroopers (the only other game that has paratroopers is v3, but v3 paratroopers are combat only). However, v1 paratroopers is a feature that I believe is completely missing from TripleA. This would restrict the application only to custom games with v3 paratroopers that can non-combat move (that is one of the few properties that only exists to support custom maps, as v3 type non-combat moving paratroopers is something not even optionally allowed in any basic game, as far as I know).

[DanVanAtta]

If we change this I'd want to avoid any unintended consequences. I think disallowing the final movement of an airplane to a territory where it cannot land makes sense. If we disallow movement to any territory where a plane cannot land, I'd be concerned that would be annoyingly too strict.

[panther2]

@DanVanAtta Definitely, as flying over enemy territories (in general) is totally legal.
It my stated example it would be "enough" if releasing (landing) the plane in the territory that has not been friendly at the start of the turn (above image 2) was prevented the same way as releasing (landing) the plane in a hostile territory (above image 1).

[DanVanAtta]

Summarizing, in essence we'd want newly captured territories to be treated the same way as enemy territories. The only caveat is to make sure kamikaze rule still works.

I suspect we'll need to still allow planes to crash for flexibility in case there is a pathing bug and/or so we don't have to deal with carrier options just yet. In such cases I think the expectation is the player will edit to correct. AFAIK the rule book says that a player must state their intention of where to land planes, particularly with carriers. Fundamentally that is just something that is difficult to enforce as a video game.

[Cernelius]

Summarizing, in essence we'd want newly captured territories to be treated the same way as enemy territories.

Or, instead, treating enemy territories the same way as captured/liberated territories, but not allowing ending the phase if you have planes that are going to crash while having enough remaining mobility to go in a zone where they may not.

As I said, I'm uncomfortable with the fact that this thing is partially enforced. Now it is partially enforced for land and not enforced for sea. If it would become fully enforced for land and not enforced for sea, it would still remain partially enforced. I would prefer it either being enforced for both land and sea or for neither.

The only caveat is to make sure kamikaze rule still works.

This needs to be investigated first (these "kamikazes" are a general option for the 3rd Editions of the Classic map (so one must be careful not to confuse this matter with the "Kamikaze Suicide Attacks" of Pacific 1940, that are a completely different thing)). All that pos2 says is:

For the isKamikaze unit option:

allows an air unit to use up all of its movement to go to a battle

For the Kamikaze Airplanes property (having it true equals giving all air units the above option true):

Kamikaze Airplanes will allow all air units on the map to use all their movement to get to a battle

If we take only these definitions, then nothing at all changes for everything at this issue, as these special rules are not touching the Non Combat Move phase at all. According to the information given (and quoted above), then you can Combat move in a way to put the air unit in a situation where it cannot land, yet, once the Non Combat move phase starts, the unit being "kamikaze" doesn't allow you to disregard the rule that you must assure the unit to land if possible. However, it may be that pos2 is giving incomplete (or even wrong) information. What does @panther2 think?

As I said, let's not forget also that you may want to send an air unit in a friendly territory where you cannot land if you have paratroopers that can transport during Non Combat Move (you would move such unit twice during the same Non Combat Move phase, first to get where you want to drop the land unit, then to get where you want to land).

[panther2]

What does @panther2 think?

@Cernelius
Talking about that special Kamikaze Air Unit I am not familiar with this rule (I don't have and know 3rd edition rules at all).
What happens to a plane that reaches the battle using all movement points? Is it automatically destroyed in case it survives the battle?

Talking about NCM-paratroopers-transport it would be fine to place the Infantry unit in a just-captured territory (as land units can always reinforce those) - while a plane must move on to a territory friendly at the beginning of the turn to land.

So to me it seems that those cases might be somehow unrelated to the current issue, in case I have not missed or misunderstood something.

On a side note, @Cernelius , I understand what you feel concerning landing at sea. But to me these are different pairs of shoes.
Moving to sea zones and landing there is different from moving to territories and landing there.
You are not allowed to land in territories that have not been friendly at the beginning of the turn. You are not allowed to land in a territory that was just captured.
But you are allowed to land in a sea zone that has been hostile at the beginning of the turn and is not any longer. You are allowed to land in the just cleared sea zone even if your plane was part of the battle there.
What you are not allowed here is to move illegally during CM-phase (in case there can't be a carrier later or another landing space). It is (possible) carrier movement that determines legal landing during NCM phase, too.

But this issue is about landing in specific territories during NCM-phase, not about moving during Combat Move Phase.
I agree however that the situation at sea deserves a better solution.

[Cernelius]

What does @panther2 think?

@Cernelius
Talking about that special Kamikaze Air Unit I am not familiar with this rule (I don't have and know 3rd edition rules at all).
What happens to a plane that reaches the battle using all movement points? Is it automatically destroyed in case it survives the battle?

I don't know. That's why I said it should be investigated. Looks like we need to know those rules (Classic 3rd Edition) or find someone reliable that knows them. Otherwise, we can just go with what documented in pos2 (assuming TripleA has no other documentation on the matter) and, in that case (since that documentation refers only to movements done before combat), as I said, the air unit being kamikaze or not changes nothing during the Non Combat Move phase (it must land if it can and cannot be sent where it cannot land, like all other air units).

I'm just saying that I don't think we can surely assume that the special rule allowing kamikaze to move disregarding landing, during Combat Move, automatically implies they can do the same during Non Combat Move.

As far as the example is concerned, I assume that a kamikaze air unit that used all its movement points to reach a land zone can only be destroyed at whatever point the air units must land (this is supposed to be different amongst rules sets) if it cannot land in that zone (however, think about limited combat round games, in which you can be the attacker in your own land territory, or more-than-two sides games, in which you can fight in a territory of a faction friendly to both sides). But, for example, it may be that if that unit is on sea and you can move a carrier to make it land, you must do it (I don't know).

[Cernelius]

By the way, the title still says "Regression".

[panther2]

@Cernelius This is from v2 Optional Rules:

1. Kamikaze Attacks
   A terrifying development was the willingness of Japanese pilots to Fly their planes directly into U.S. ships.
   They even developed “Flying bombs” piloted by the soldiers inside.
   Your air units may make a combat move without having to land in a friendly space afterward. You must
   declare during the combat move phase that an air unit is making a kamikaze attack. Each such air unit
   attacks during the opening fire step of combat, and only an enemy sea unit can be chosen as a casualty.
   That air unit automatically becomes an opening fire casualty in addition to any other casualties inflicted in
   this cycle of combat.

(bolded by me)

TripleA handles this totally incorrect by assigning this ability not as "National Advantage" to gain for Japan only but to every power's aircrafts from the beginning of the game when this option is set.
Then the planes are not destroyed as opening fire casualty - but incorrectly later during NCM.

I tested the TripleA implementation of "Kamikaze" with v1.3rd ed. and it works the same way as in v2.

[Cernelius]

@Cernelius This is from v2 Optional Rules:

1. Kamikaze Attacks
   A terrifying development was the willingness of Japanese pilots to Fly their planes directly into U.S. ships.
   They even developed “Flying bombs” piloted by the soldiers inside.
   Your air units may make a combat move without having to land in a friendly space afterward. You must
   declare during the combat move phase that an air unit is making a kamikaze attack. Each such air unit
   attacks during the opening fire step of combat, and only an enemy sea unit can be chosen as a casualty.
   That air unit automatically becomes an opening fire casualty in addition to any other casualties inflicted in
   this cycle of combat.

(bolded by me)

TripleA handles this totally incorrect by assigning this ability not as "National Advantage" to gain for Japan only but to every power's aircrafts from the beginning of the game when this option is set.
Then the planes are not destroyed as opening fire casualty - but incorrectly later during NCM.

I tested the TripleA implementation of "Kamikaze" with v1.3rd ed. and it works the same way as in v2.

Actually, just like in the case of Global, that would be merely another unrelated rule, that just so happens to be called as "kamikaze" too.

[Cernelius]

There are the "kamikaze" of Classic 3rd Edition, there are the "kamikaze" of Revised, there are the "kamikaze" of Pacific (the non-1940 one) and there are the "kamikaze" of Pacific 1940 and Global 1940.

I think we can safely assume that the "kamikaze" we are talking about here are those of Classic 3rd Edition. So, as I said, let's not confuse this matter with other unrelated rules that just so happen to have "kamikaze" in their names too.

[DanVanAtta]

Kamikaze just means planes are allowed to crash. Somewhat more precisely,
plane (and carrier) movement is not restricted by landing considerations.

On Mon, Aug 24, 2020, 2:57 AM Cernelius [e-mail] wrote:

By the way, the title still says "Regression".

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/triplea-game/triplea/issues/7350#issuecomment-679031652,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/AC6SZOP3GXGKIGJSKNP7PEDSCI2Q3ANCNFSM4P3C6RMA
.

[panther2]

A question is when exactly the planes crash. Chances are that they are removed already during Conduct Combat Phase.

I have asked publicly in the forum:
https://www.axisandallies.org/forums/topic/35670/kamikaze-in-classic-3rd-edition
and contacted Krieghund asking him to look into it.

[DanVanAtta]

While kami might be defined in a rulebook, there has been a working
definition as commonly played. Arguably a kami rule consistent with
rulebook is a new feature and we'd need to rename current kami rule to
something like "allow planes to crash"

On Mon, Aug 24, 2020, 7:13 AM P@nther [e-mail] wrote:

I have asked publicly in the forum:

https://www.axisandallies.org/forums/topic/35670/kamikaze-in-classic-3rd-edition
and contacted Krieghund asking him to look into it.

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/triplea-game/triplea/issues/7350#issuecomment-679152343,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/AC6SZOLAQGKQA5UENQDGTXLSCJYR3ANCNFSM4P3C6RMA
.

[panther2]

@DanVanAtta Please note that I did not ask for anything here.

I have been asked about my opinion regarding "Kamikaze" - and it turned out that answering this aspect is difficult as long as no one - including me - knows the actual rule behind.

In case it turns out that TripleA follows the Classic 3rd ed. rules - fine. If not - someone might want to create a feature request.

The issue I raised above is about landing planes during NCM - not about Kamikaze attacks or crashing planes at whatever time.
On the contrary: Resolving the issue requires that planes do not crash.

[Cernelius]

Kamikaze just means planes are allowed to crash. Somewhat more precisely, plane (and carrier) movement is not restricted by landing considerations.
…
On Mon, Aug 24, 2020, 2:57 AM Cernelius @.*> wrote: By the way, the title still says "Regression". — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#7350 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AC6SZOP3GXGKIGJSKNP7PEDSCI2Q3ANCNFSM4P3C6RMA .

As I said, from what it is explained within "pos2", that is the case for Combat movement, but not for Non Combat movement.

Kamikaze Airplanes will allow all air units on the map to use all their movement to get to a battle

It literally says that you can do it only if you are moving to take part in a battle (thus never during Non Combat Move).

[panther2]

... what maybe implies that Kamikaze planes never "crash" during NCM phase. But we will see.

[DanVanAtta]

The Kamikaze option would be better renamed "require planes to land". Turning off the 'allow planes to crash' prompt is too much of a change for this issue. Though, having the captured territories behave the same as enemy territories seems extremely reasonable.

I state the above despite whatever incomplete spec is detailed in PoS or an obscure V2 rule that is what it is. The current behavior of kamikaze is what players have come to expect, that you can move planes however and let them crash if desired - this we should not change.

That rule comes into play because when we code up the validation logic, we must be sure that the current 'kamikaze' option is still an override. In other words, if kamikaze option is on, then yes, planes can end non-combat in newly captured friendly territories and they would crash.

[panther2]

@DanVanAtta

In other words, if kamikaze option is on, then yes, planes can end non-combat in newly captured friendly territories and they would crash.

Yes, why not. It is one of many options that might be useful on custom maps.

@Cernelius
Krieghund answered that he only has the Rulebook for that edition. It only mentions that the Kamikaze option exists, but says nothing about how it works. I have contacted another forum member who seems to own the computer game. From the screenshots he once posted (https://www.axisandallies.org/forums/topic/32562/3rd-edition-rules-compendio-v3-5-final) one can conclude that "Kamikaze" is an option offered next to others, even unrelated to the four rules (changes) that characterize the 3rd edition (in Spanish but easily to figure out).

[Cernelius]

Ok, then it looks like it is open to personal opinions.

In my opinion, "Kamikaze Airplanes" and "isKamikaze" should only allow you Combat moving the air units without the need of being able to land them by the turn's end (meaning that, during the Non Combat Move phase, you would be still obliged to land the biggest number of aircrafts you can, regardless whether or not the aircraft is a "kamikaze" one).

This is based mainly on my understanding that only a chance of inflicting damage to the enemy, during the same turn, can be used as reason for surely sacrificing the "kamikaze" unit. Therefore, you can never "kamikaze" during Non Combat Move, instead (if your Combat movements put the unit in a position where it cannot possibly land, then it will crash, otherwise, once the Conduct Combat phase is over, you are obliged to do whatever you can to land it (comprising carriers movements, if needed), just like any regular air unit).

Meaning that the "Kamikaze Airplanes" property and the "isKamikaze" option should influence the Combat Move phase, but be irrelevant during the Non Combat Move phase. I guess this would also simplify the program's coding, as you would not have to distinguish between "kamikaze" and "non-kamikaze" air units during the Non Combat Move phase.

[panther2]

Talking about personal opinions, here is my argumentation for a possible feature request (that I am not going to open):

It is the nature of Kamikaze to perform a combat action and die. So in TripleA I would not allow a Kamikaze-plane to survive the Conduct Combat Phase. The Kamikaze unit would be removed during Conduct Combat Phase in the context of its action. NCM would be irrelevant for those units. That would be comparable to those rulesets where we have written rules for whatever Kamikaze solution, too.

[DanVanAtta]

In triplea games, notably cold war, kami is often used when for example you
expect 2 out of 4 fighters to die. In such case if they don't have enough
movement it does not matter so much and you get the extra attack power.

How kami works is reasonably well established.

On Tue, Aug 25, 2020, 5:54 AM P@nther [e-mail] wrote:

Talking about personal opinions, here is my argumentation for a possible
feature request (that I am not going to open):

It is the nature of Kamikaze to perform a combat action and die. So in
TripleA I would not allow a Kamikaze-plane to survive the Conduct Combat
Phase. The Kamikaze unit would be removed during Conduct Combat Phase in
the context of its action. NCM would be irrelevant for those units. That
would be comparable to those rulesets where we have written rules for
whatever Kamikaze solution, too.

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/triplea-game/triplea/issues/7350#issuecomment-680006607,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/AC6SZOJDPMMYOZYPNDNSFILSCOYALANCNFSM4P3C6RMA
.

[DanVanAtta]

I just looked into this some more:

• if a plane has additional move points, you can release it in a newly captured territory
• if a plane uses its final movement points to move into a newly capture territory, the move is not allowed

I think this is okay behavior. We do not require for a unit to be fully moved when landing. In other words, the game handles moving one territory at a time and using 'shift' to create waypoints as equivalent. I don't think therefore we can fix this issue without breaking that aspect of the game.

There is certainly a convenience aspect as well if for some reason it's difficult to move planes directly to where they will be landing. Some examples of that include when the final territory is not visible and needs map scrolling, or if there needs to be coordination to move an AC, or if it's purely convenient to move a group of planes all at once to a territory that is not the final landing zone.

[panther2]

@DanVanAtta

I just looked into this some more:

• if a plane has additional move points, you can release it in a newly captured territory

This is far from being okay behaviour.

A plane is not allowed to land in a newly captured territory. For landing it doesn't matter if there are movement points left or not. Of course the plane may fly over that territory to land somewhere else. But not there.
The problem here is that planes are crashed, what should not happen at all. The player who tries to incorrectly land should at least be prevented from finishing his move!

So in case you feel that the movement code should not be touched please at minimum address this incorrect consequence resulting from incorrect landing.

[DanVanAtta]

I think what it comes down to is movement via shift clicking and moving one territory at a time should be equivalent.

The player who tries to incorrectly land should at least be prevented from finishing his move!

I agree, but I can see scenarios where this would not have a good outcome. For example carrier movement, sometimes the engine offloads the wrong air unit from carriers (in the case of allied cargo) and while there is actually capacity, the engine does not make it so. I can also think of scenarios where reverse connections were missing in maps so a plane could adequately arrive at a target but on the return route, due to defect, could not actually land. There is no way to edit force a turn to be over, these units have moved and/or would have been in combat so you probably can't remove them via edit. This would leave the players stuck.

It's hard to think of a scenario where it is to the players advantage to have planes crash. Maybe on some maps, perhaps, but 99% of the time players are not going to do so willingly. I think the focus should then be on ensuring that the outbound flight path is valid so there is a place to land and then trust that players would serve their interests and would land all planes.

Bottom line, if we disallow end of turn in these situations, we could cause games to become stuck without any remedy. My thinking is that is a more severe consequence and cannot be fixed until we guarantee there would be some way to advance the game in case there is some odd landing bug.

[panther2]

@DanVanAtta

All right then. This is a very unsatisfying result of this discussion/issue. But I see the side effects you mention and tend to agree that it might be tolerable to leave it as is.

[DanVanAtta]

I would tend to agree @panther2 . We could try to improve the messaging to make it clear a play must try to land and if forced to crash planes then they should be allowed an edit (or their move was not legal to begin with). If you have any suggestions there we can certainly follow-up.