MTSG Retreat Bug still Present

[vicberg]

This is an engine bug I reported a while ago. MTSG involved. Look at the results. 4 retreat attempts and 0 failures. That shouldn't be happening. 1 retreat attempt and 0 failures. Battle lasted 6 turns because units tried to retreat and for whatever reason and because MTSG was involved, units aren't retreating.

[ATTACH]37749[/ATTACH]

[Captain_Orso]

How do you know that MTSG is involved?

I'm also a bit surprised that 4 retreats were successful, but the battle just doesn't seem to want to end. I thought that if once side successfully retreats, if units arrive through MTSG or normal movement another battle ensues, and not the same one. But maybe that is part of the streamlining of the game, so that multiple battles don't take 5 minutes to handle.

[vicberg]

Testing. I've done multiple combats now. For whatever reason if MTSG is involved, units "successfully" retreat (either attacker or defender, I've seen both attempt to retreat) and there's no actual retreat. If MTSG isn't involved, the retreats occur fine.

In the three cases I've seen this now, a large city was involved in the combat, so it may have something to do with entrenched units involved in the combat

[Captain_Orso]

First off, if other corps or army stacks are MTSG'ing into an already occurring battle, you should see them added to that side's forces when comparing one round of battle to the next; they will be added to the list of Friendly or Enemy Commanders, depending on which side is being reinforced.

Battle in general and retreat in specific is rather complex.

The very first rule you encounter is the 5% MC rule. If your side has <5% MC in a region you enter, regardless of enemy presence, your moving stack changes to OP (Offensive Posture) automatically, unless it is in PP (Passive Posture). This generally forces a moving stack to attack an enemy force in the region into which it is moving. As soon as your stack has changed to OP, it will gain 5% MC in the region, which it will maintain until it leaves the region, even it is forced to retreat.

Once at least one faction has >=1 stack in OP in the region, the game checks which of the stacks in the region find and engage each other. From this grouping of engaged stacks, if one side is greatly outnumbered, it may already be put into retreat before any fighting actually takes place.

Once the fighting has started, if there is a corps or army stack(s) of the same army in a neighboring region(s), it may be called to MTSG. A time factor is calculated on how long it would take for the MTSG'ing stack to arrive in the region of the battle, and if it can arrive on time, it will be added to the battle some time after the first round of battle.

It is possible that if there are multiple stacks on one side, that one or more stacks on that side break and go into retreat, while one or more other stacks stand their ground, but because if one stack breaks, other stacks on the same side have to check their morale, they may also break and go into retreat.

Retreating stacks may be pursued by cavalry, depending on the difference between the good-order cavalry of the retreating and the pursued side; the more cavalry the better. Pursuit can cause very heavy casualties even compared to the battle itself.

In older games where the battles took much longer to calculate, generally if one side retreated and escaped pursuit, the battle was over. But in WoN the battles have been streamlined to not take several minutes to run, so it may be that although one side goes into retreat, forces arriving through MTSG may yet be giving an opportunity to engage in fighting.

Also, although one side retreats from battle, it will not always retreat out of the region, but to within the region, or may not have already exited the region, when other enemy stacks enter, which may cause an additional battle, especially if those moving stacks have targeted any of the retreating stacks for interception.

The battle engine has obviously changed some since CW2 days, so there are probably some things I don't know about in the details.

[vicberg]

All I know is that this is consistently happening.

I can recreate each time by MTSG into a large city (though it may be happening elsewhere). Attacks always last 6 rounds. Combat logs always show successful retreats, but no actual retreat as the MTSG units are always re-engaging. In some cases, the attacker is attempting to retreat and the attacking MTSG units re-engage. In the above case, the defender is trying to retreat.

It's a bug related to what you just said about MTSG and engagement. The MTSG units are engaging even after a successful retreat.

[loki100]

...

It's a bug related to what you just said about MTSG and engagement. The MTSG units are engaging even after a successful retreat.

I don't know why you are so keen to call 'bug'.

Think about the dynamics. 2 Corps engage in a battle and send a message to a neighbouring force to help. They lose and start to fall back but the neighbouring force is now fully committed. It will now probably lose but that maybe enough to allow the others to rally etc.

Its all about how you set up and manage your forces.

also don't forget the totals shown are all the units in the province(s) that could have fought, so its often misleading as to the numbers actually engaged.

[Captain_Orso]

IIRC the units you see when you click on the numbers (the battle rounds) are the units actually fighting during that round. What you see under (S) and (E) are Start and End units in the region, whereby under (E) units which were destroyed and/or captured are crossed out in the lower section.

[vicberg]

Here's why I believe this is a bug

Nappy was adjacent to Wien. Soult and Lannes were on Wien. I looked at battle log.

Soult and Lannes attempted to retreat on round 2 and were succesful, but Nappy kept attacking, keeping Soult and Lannes fighting for 4 more rounds. I have to believe something is wrong there. Soult and Lannes should have retreated and the combat over, but Nappy being adjacent kept engaging each round and kept them fighting at enormous loss.

Just like the sample I posted above. 4 retreats, 0 failures. Doesn't make sense. Why didn't the defender retreat? He could have retreated to the city or an adjacent region, but 4 times? I didn't look at battle log, but I'm guessing that the same thing occurred. Nappy kept engaging.

[Captain_Orso]

It would depend very much on what is in Nappy's stack.

You will only retreat into the city if you have the Enter Structure SO selected, and even then this is a small chance that you still won't retreat into the city.

There is one thing you are overlooking, and that is that the 'bug' might be that a 'retreat' is show when none takes place. It could be that everything else is WAP.

[lycortas2]

Why I call this a bug is Murat dying to the last man and horse while set to retreat, avoid combat, and successfully retreating but Ney came in to support so the battle continued.... but Ney did no fighting just Murat who had already retreated. And the ALL of the cavalry died.
You can defend this all day until you are blue in the face but I am just about done with AGEOD.

[fred zeppelin]

Why I call this a bug is Murat dying to the last man and horse while set to retreat, avoid combat, and successfully retreating but Ney came in to support so the battle continued.... but Ney did no fighting just Murat who had already retreated. And the ALL of the cavalry died.
You can defend this all day until you are blue in the face but I am just about done with AGEOD.

Sounds more like March to the Sound of Suicide.

[vicberg]

In the case of Nappy, Soult and Lannes, Soult and Lannes attempted to retreat and were successful and went into PASSIVE MODE. Then the logs showed Nappy engaging forcing Soult and Lannes to continue combat at minuses for being in Passive Mode.

No way you can convince me this is WAD. Something with MTSG isn't working.

[loki100]

In the case of Nappy, Soult and Lannes, Soult and Lannes attempted to retreat and were successful and went into PASSIVE MODE. Then the logs showed Nappy engaging forcing Soult and Lannes to continue combat at minuses for being in Passive Mode.

No way you can convince me this is WAD. Something with MTSG isn't working.

sorry can't see the 'bug' in this one. If they were passive their chance to be selected in the combat engine is much lower, so you are trying to break off from a much larger enemy if they are still being targetted? Sounds like good modelling of a monumental mess where Napoleon's intervention has worsened the situation.

one of the delights of the various parts to the AGE system is the way it models and allows for the really undesirable (by the player) to happen. Which is why I suggested higher up that this is a situation where better army organisation may well have helped

[fred zeppelin]

sorry can't see the 'bug' in this one. If they were passive their chance to be selected in the combat engine is much lower, so you are trying to break off from a much larger enemy if they are still being targetted? Sounds like good modelling of a monumental mess where Napoleon's intervention has worsened the situation.

one of the delights of the various parts to the AGE system is the way it models and allows for the really undesirable (by the player) to happen. Which is why I suggested higher up that this is a situation where better army organisation may well have helped

Respectfully, this sounds to me more like post hac rationalization than a "delight" of design.

[TJD]

Respectfully, this sounds to me more like post hac rationalization than a "delight" of design.

+1

[MarshalJean]

Respectfully, this sounds to me more like post hac rationalization than a "delight" of design.

Respectfully, this opinion would mean a lot more to me if it was coming from someone who knew as much about the inner workings of the Athena system as Loki does...

I guess we all have our respectful opinions.

[fred zeppelin]

Respectfully, this opinion would mean a lot more to me if it was coming from someone who knew as much about the inner workings of the Athena system as Loki does...

I guess we all have our respectful opinions.

I don't need to know the intricacies of a clock's design to know whether the time it displays is correct.

[MarshalJean]

I don't need to know the intricacies of a clock's design to know whether the time it displays is correct.

If the Athena system was more analagous to a clock than it actually is, your point would be valid. But it isn't.

A clock is designed to have such precision for the purpose of providing utter predictability...it can therefore be simply checked by other clocks or even by looking outside, in some cases. Our ability to evaluate its accuracy is completely predicated on its predictability.

But Athena is quite different. The entire system is designed to provide a significant array of unpredictability while still managing to provide unpredictable results within a field of likely possibilities which are communicated to the player and affected by the players many decisions. In other words, to evaluate whether it's working properly means having a thorough knowledge of how the system is weighing likelihood with possibilities and still providing predictable, and sometimes unpredictable, results.

So, no. I don't accept the analogy. To borrow language you used before, evaluation of the system cannot be simply made by superficial rationality. The variables are too great for that. And for someone who doesn't know much about the variables to tell someone else who does that they are dealing in whimsical rationalization is hypocritical at the very least.

[fred zeppelin]

If the Athena system was more analagous to a clock than it actually is, your point would be valid. But it isn't.

Fair point. Bad analogy.

But it's also fair that sometimes folks are so close to something that they can't see its flaws. Athena may be a wonderfully intricate engine, but it's still only as good as its results. And, respectfully, these results make no sense no matter how one rationalizes them in the context of the game system.

[loki100]

Fair point. Bad analogy.

But it's also fair that sometimes folks are so close to something that they can't see its flaws. Athena may be a wonderfully intricate engine, but it's still only as good as its results. And, respectfully, these results make no sense no matter how one rationalizes them in the context of the game system.

the point I am trying to make is that in the context of how the engne works this is not automatically a bug. Its rare for stacks in a green stance to be attacked unless the enemy has a lot more stacks (in offensive stance). The units arriving clearly restart the battle, being outnumbered (at least in stacks) means the units falling back get hit again.

I realise its much more fun to shout bug and suggest that anyone who doesn't join in is unable to see the flaws.

The big issue here is that for some reason some posters believe that war and military campaigns goes to plan. One of the delights of the AGE system is that every now and then it will go very wrong before your eyes. Now if you want chess like predictability thats cool ... but just because AGEOD's system doesn't deliver it doesn't make it a bug.

You are happily joining in with calling this a bug as that is clearly what a small group do. All I can see is a single line of a battle report and the comments of the player. Not enough, especially when I can quite easily think of an explanation.

[Drake001]

But it's also fair that sometimes folks are so close to something that they can't see its flaws. Athena may be a wonderfully intricate engine, but it's still only as good as its results. And, respectfully, these results make no sense no matter how one rationalizes them in the context of the game system.

Agreed. Proof is in the pudding's results and they show something wrong, as it has been for awhile.

[vicberg]

the point I am trying to make is that in the context of how the engne works this is not automatically a bug. Its rare for stacks in a green stance to be attacked unless the enemy has a lot more stacks (in offensive stance). The units arriving clearly restart the battle, being outnumbered (at least in stacks) means the units falling back get hit again.

I realise its much more fun to shout bug and suggest that anyone who doesn't join in is unable to see the flaws.

The big issue here is that for some reason some posters believe that war and military campaigns goes to plan. One of the delights of the AGE system is that every now and then it will go very wrong before your eyes. Now if you want chess like predictability thats cool ... but just because AGEOD's system doesn't deliver it doesn't make it a bug.

You are happily joining in with calling this a bug as that is clearly what a small group do. All I can see is a single line of a battle report and the comments of the player. Not enough, especially when I can quite easily think of an explanation.

There's another thread where I posted battle logs when I first reported this. I have much more detail.

Captain Orso's response above may be the real issue

"Once at least one faction has >=1 stack in OP in the region, the game checks which of the stacks in the region find and engage each other. From this grouping of engaged stacks, if one side is greatly outnumbered, it may already be put into retreat before any fighting actually takes place.
.
.
.
In older games where the battles took much longer to calculate, generally if one side retreated and escaped pursuit, the battle was over. But in WoN the battles have been streamlined to not take several minutes to run, so it may be that although one side goes into retreat, forces arriving through MTSG may yet be giving an opportunity to engage in fighting."

If both corp in Wien were passive and Austrians set to defensive posture (and they were), there should not have been combat, but there was. So Captain's second statement may be where the issue is, understanding you don't believe there is one. Is the check for MTSG engaging in combat looking at what's going on within the attack region or is it happening REGARDLESS of the attack region.

[fred zeppelin]

If both corp in Wien were passive and Austrians set to defensive posture (and they were), there shouldn't not have been combat, but there was. So Captain's second statement may be where the issue is, understanding you don't believe there is one. Is the check for MTSG engaging in combat looking at what's going on within the attack region or is it happening REGARDLESS of the attack region.

Good observation and question. Something certainly seems not quite right.

[Captain_Orso]

Some things have changed in the battle engine since CW2. If I understand correctly, from the descriptions above, the two French Corps in the Wiener region both broke and when into retreat--they were in PP and possibly, but not necessarily heading out of the region, but Napoleon's stack, which had MTSG'ed, was still in good order. At this point, somebody--I would guess--was still in OP so the battle continued.

As was stated, target stacks were selected, but it's not all simply random chance. The only thing completely random is when the very first non-moving stack in OP is selected. That can be from either side. It picks a target from groupings of enemy stacks with the following characteristics and in this order:
non-moving stacks in OP
moving stacks in OP
non-moving stacks in DP
moving stacks in DP
non-moving stacks in PP
moving stacks in PP

Then the next non-moving stack in OP is selected, but if on the enemy side there is already a targeted stack which is not saturated (50% more opposing power than the target stack(s)) the currently targeting stack will also target that targeted stack.

Eventually all stacks are processed to determine whether they are targeting or targeted--although there is some chance that some stacks might not target nor be targeted, although I haven't been able to determine how exactly that works--.

So if Napoleon's stack, yet in good order, and is considered in the region through MTSG, that the battle continues is not terribly surprising, as long as there is a stack in OP, and from the battle report, Napoleon is in AP, which is the same as OP for field battles.

[vicberg]

That explains. Nappy was in OP. So he kept attacking.

I'm going to have to try Nappy in DP and see if that makes a difference with MTSG.

[RickInVA]

the point I am trying to make is that in the context of how the engne works this is not automatically a bug. Its rare for stacks in a green stance to be attacked unless the enemy has a lot more stacks (in offensive stance). The units arriving clearly restart the battle, being outnumbered (at least in stacks) means the units falling back get hit again.

I realise its much more fun to shout bug and suggest that anyone who doesn't join in is unable to see the flaws.

The big issue here is that for some reason some posters believe that war and military campaigns goes to plan. One of the delights of the AGE system is that every now and then it will go very wrong before your eyes. Now if you want chess like predictability thats cool ... but just because AGEOD's system doesn't deliver it doesn't make it a bug.

You are happily joining in with calling this a bug as that is clearly what a small group do. All I can see is a single line of a battle report and the comments of the player. Not enough, especially when I can quite easily think of an explanation.

Given all that it seems the real challenge is to be able to distinguish between a WAD unusual result and a bug. I do feel it is a burden properly placed on those that would say it is a WAD unusual result to fully explain why that is the case. I love AGEOD games, but I also desire to know what levers I need to pull to increase my chances of getting the results I desire. Full explanations of why strange results are WAD assists greatly in this and is always greatly appreciated.

[fred zeppelin]

Given all that it seems the real challenge is to be able to distinguish between a WAD unusual result and a bug. I do feel it is a burden properly placed on those that would say it is a WAD unusual result to fully explain why that is the case. I love AGEOD games, but I also desire to know what levers I need to pull to increase my chances of getting the results I desire. Full explanations of why strange results are WAD assists greatly in this and is always greatly appreciated.

No one expects, or even wants, predictability. (Dullards that we are, loki, we have managed to stumble across a history book or two.) The fortunes of war should always be a factor.

It's not the unpredictability of the results so much as their extremity that make this look like a bug.

[loki100]

No one expects, or even wants, predictability. (Dullards that we are, loki, we have managed to stumble across a history book or two.) The fortunes of war should always be a factor.

It's not the unpredictability of the results so much as their extremity that make this look like a bug.

There's another thread where I posted battle logs when I first reported this. I have much more detail.

Captain Orso's response above may be the real issue

"Once at least one faction has >=1 stack in OP in the region, the game checks which of the stacks in the region find and engage each other. From this grouping of engaged stacks, if one side is greatly outnumbered, it may already be put into retreat before any fighting actually takes place.
.
.
.
In older games where the battles took much longer to calculate, generally if one side retreated and escaped pursuit, the battle was over. But in WoN the battles have been streamlined to not take several minutes to run, so it may be that although one side goes into retreat, forces arriving through MTSG may yet be giving an opportunity to engage in fighting."

If both corp in Wien were passive and Austrians set to defensive posture (and they were), there should not have been combat, but there was. So Captain's second statement may be where the issue is, understanding you don't believe there is one. Is the check for MTSG engaging in combat looking at what's going on within the attack region or is it happening REGARDLESS of the attack region.

aye, forgive me. I do keep on forgetting that anyone who disagrees with your consensus really has no right asking questions ... please ignore me and I'll revert to my earlier view that avoiding this sub-forum is the easiest solution (for me)

[lodilefty]

OK, everybody take a deep breath and stay civil....

[vicberg]

Xanax time again

Actually, I'm fine with the response. i was under the impression that MTSG is there to support the attack region. Looks like it's considered a full part of the attack, even if adjacent. I was under a different impression on how MTSG works.

So, the way to work MTSG such that I don't get the same results I got in the past is to put the MTSG units on defensive posture. This will bring them into the combat and not force the combat to continue if units in the attack region wish to retreat.

Works for me.