"Run to mamma and get killed" bug still not fixed in v1.1 ?

[Guest tero]

I saw yesterday in a v1.1 PBEM a German FO get flushed out of a burning building on a hill. I suspected the house to be occupied by the said FO after receiving some arty on my troops and I did a bit of recce by fire.

That is cool except the FO ran out of the house into the open towards my Allied troops instead of retreating out of LOS behind the hill.

I though that "feature" was toned down. What gives ?

[von Lucke]

Well, if you were in a burning building, would you run out the closest exit (which in yr example I am assuming is toward the Ami's), or would you run back through the burning, smoke-filled house and maybe make it out the back?

[Guest tero]

Well, there WERE windows on all four walls of the buiding, including on the safe side, on all storeys (at least in the graphics).

Furthermore the FO also started running towards my troops not around the corner of the building to safety.

[PeterNZer]

I've read one person give two accounts of such action in ww2. In one case a shell-shocked soldier, literally, cried 'maama' and jumped up out of a foxhole and ran towards the machineguns..

In another, some new people to a platoon thought during some exchange of fire overnight that they were cut off and surrounded and so jumped out of their foxholes and ran off into a grain field. They were all (9 or so of them) found dead the next day (probably friendly fire). Platoon didn't even get to know their names.

Shell shock, or 'battle fatigue' can hit really quickly. This FOO reccounts survivors of Totalize (?) who were bombed by the RAF and went a bit potty for a couple of days being set off by one of his shells hitting trees in the area they were attacking, some half-dozen men in a company had to be re-evacuated to recover.

PeterNZ

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

"Patriotism is the virtue of the viscious" - Oscar Wilde

"Don't F*CK with Johnny Cash!" - Chupacabra

[jasoncawley@ameritech.net]

Yeah, guys that are panicked run towards the nearest cover besides where they are at. They seem to have very little sense of "safe" as a direction. I do notice, however, that when giving the "withdraw-run" command, the game enforces the "toward own home side" requirement (even more than anything about cover) pretty forcefully.

Perhaps the same rule should apply to most panicked units? I mean, sure, let 1/4 of them run around like headless chickens toward the nearest (submachinegun infested) cover. Variety is the spice - LOL. But let the other 3/4 or so run for "home", (even if, because the enemy got behind them, that isn't very safe either).

[Guest Germanboy]

Hmm, I just knew this behaviour would create more problems than it solves. The old rule again - don't put anyone in houses.

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

Andreas

Der Kessel

Home of „Die Sturmgruppe“; Scenario Design Group for Combat Mission.

[Guest tero]

Why does it seem it always assumed that when a German unit does something really stupid it can be explained away with natural causes, like panic, but when an Allied unit does the same it warrants further study to reveal it is indeed a bug ?

Panic is a real issue but:

1) It was recon by fire that set the house on fire. It was area fire, not aimed at any specific target. The FO was actually spotted only when it started moving to clear out of the house.

2) The incident is taking place during turns 3 onwards so battle fatigue setting in quickly seems far fetched as that recce by fire is my first active action.

3) I have never had any problems with manhandling units to enter and exit houses regardless of the openings on the walls.

4) Arty kept on falling on my troops indicating that the FO did not in fact panic. I have a FO under that barrage and its fire mission got cancelled because it was forced to take cover.

>They seem to have very little sense of "safe" as a direction.

They should. I would deem it prudent for units to know their mission, their frontage and responsibilities and the approximate location of friendly units when the battle starts. In case they are forced to widraw they should be able to know which way to go. Not just in the opposite direction the enemy is but also towards the friendlies. If they are cut off it is a different matter but in this particular instance my troops were nowhere near the FO's rear area. IRL defending units are assigned secondary and even tertiary (prepared) positions in case they have to widraw.

>Perhaps the same rule should apply to most panicked units?

Unless they are about to surrender I agree.

>I mean, sure, let 1/4 of them run around like headless chickens toward the nearest (submachinegun infested) cover. Variety is the spice - LOL. But let the other 3/4 or so run for "home", (even if, because the enemy got behind them, that isn't very safe either).

I agree.

[Guest tero]

>The old rule again - don't put anyone in houses.

Since it was not my FO that acted stupidly I think my PBEM opponent should be yelling bloody murder and not me...

But I would hate it if my troops did that.

[mensch]

yes yes yes.. we understand.. but why did not BTS program the most important rule after running out of a burning house

1. Stop

2. Drop

3. and Roll

no wonder they get shot up. jeez.

---------

www.derkessel.com Home of „Die Sturmgruppe“; Scenario Design Group for Combat Mission.

[David Aitken]

tero

> Why does it seem it always assumed that when a German unit does something really stupid it can be explained away with natural causes, like panic, but when an Allied unit does the same it warrants further study to reveal it is indeed a bug ?

How on Earth do you come to that conclusion?

Would you think it better if the FO had remained in the building, which subsequently collapsed and killed him? The FO is evacuating because the building is falling apart around his head. I argued against this behaviour, which was implemented in 1.1 (I think), because as you have found the AI does not intuitively know where safety lies. However, I also think you will find that half of the time the units in question exit a building on the safe side (if there is a safe side at all). It's a matter of chance, and you've witnessed an occasion where a unit made the wrong choice.

As always, though, why should strange behaviour always be regarded as a bug? In reality, this FO could have been forced to exit through the front because he thought it too dangerous to make his way to the back. This is not specifically simulated in the game just now, but it is accounted for. He could also have suspected that enemy forces were behind him, and felt that the front was his best bet. Another thought which comes to mind is that he was concerned to continue his fire mission. And above all, there is a degree of uncertainty programmed into the game, which accounts for simple bad decisions and mistakes.

David

[Guest Germanboy]

I think this shows the fundamental problem with programming more SOPs into the game. As you may remember, I also strenously argued against the change. If you look at the other debate on 'shoot n'scoot', you can easily imagine a number of situations in which the AI shoots n'scoots and that leaves you off worse than before. The same goes for an 'follow-to-assault' command. While these changes are very compelling on the surface, closer expectation (or in this case experience) shows them to create problematic situations and player complaints. Looks like a lose-lose situation for BTS to me.

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

Andreas

Der Kessel

Home of „Die Sturmgruppe“; Scenario Design Group for Combat Mission.

[Horncastle]

Another problem with infrantry when under fire I have found is repeat behaviour when a building is occupied. Im not sure if there is a maxium capacity of troops a building can have ( sure there is ), but I have found troops running into a house to get in cover to find its full to run back out, come under fire, run back in and so forth.

This not only exposes them to more fire, but they are unable to return fire effectively as they are too busy running back and forth.

[CMplayer]

I agree. I set up a beach assault scenario

where there was basically no cover anywhere.

The routed Americans tended to run towards

the German pillboxes in a kind of Kamikaze

charge.

regards,

--Rett

[Guest tero]

>How on Earth do you come to that conclusion?

When talking about the Allied units and their behaviour people SEEM to be more eager to look for solutions outside the "**** happens" cathegory.

I think it is too simplistic to dismiss this kind of abnormal combat situation behaviour as panic or bad judgement call, especially since the "run towards their death" was an acknowledged bug in previous versions.

>Would you think it better if the FO had remained in the building, which subsequently collapsed and killed him?

Nope.

>The FO is evacuating because the building is falling apart around his head.

That is reasonable. But why pick a direction that spells trouble when you can exit the house to an out-of-LOS location securely behind the steepish hill the house was on ?

The game is in turn 3 so I think it would be safe for the AI to assume that the rear of the house is secure since the over all LOS is 500+ meters and no other friendly unit is engaged.

>I argued against this behaviour, which was implemented in 1.1 (I think), because as you have found the AI does not intuitively know where safety lies.

That may be the underlying problem. I think that it could be worked around if the defender could assign secondary positions to his units. That way, if predetermined criteria (thresholds) are met, the defending unit would start widrawing to alternate positions without the move becoming a rout. And if there was a "widraw to secondary positions" order you could do it in a way that would be more controlled than having them use regular run/move orders.

>However, I also think you will find that half of the time the units in question exit a building on the safe side (if there is a safe side at all). It's a matter of chance, and you've witnessed an occasion where a unit made the wrong choice.

Coincidencies hardly ever are.

>As always, though, why should strange behaviour always be regarded as a bug?

Because in this instance it was said this particular feature was dealt with and it still resembles too closely the acknowledged bug. And since this is only a computer program we are discussing here pretty much every action follows certain predetermined paths and when abnormalities occur people compare them to other similar instances and make their own conclusions.

That is why MS talks about features and not bugs.

>In reality, this FO could have been forced to exit through the front because he thought it too dangerous to make his way to the back.

This is not specifically simulated in the game just now, but it is accounted for. He could also have suspected that enemy forces were behind him, and felt that the front was his best bet.

In this particular instance, no. Turn 3, LOS 400+ meters, house on a hill overlooking my approaches. Little woods with wide spaces in between. I started the recce by fire because I suspected that multistorey house on the hill was where the FO raining shells on my troops was.

>Another thought which comes to mind is that he was concerned to continue his fire mission.

As this was an AI controlled act I doubt it. As I said my Allied FO got his mission cancelled because he had to take cover eventhough I wished he would have remained on target.

>And above all, there is a degree of uncertainty programmed into the game, which accounts for simple bad decisions and mistakes.

If this is indeed the case then there is something wrong in the German FO algoritms. He is sitting there unspotted directing fire on my troops and when my recce by fire starts a fire he decides to reveal himself so as not to disrupt the fire mission. When, at the same time, my Allied FO takes cover and gets the fire mission cancelled. Sorry, that is more than a bit too thick.

[Tiger]

How would the FO know there was trouble in that direction? Were there troops directly in front of him he could see? Were the enemy troops in cover at the time or in the open field? I dunno. Troops don't have the same birds-eye view of the battlefield as the player does, whether buttoned up in a tank or busy calling in arty strikes while observing through binoculars/radioing in corrections. You did state that the arty was still falling on your troops.

jmtcw,

Tiger

[This message has been edited by Tiger (edited 01-29-2001).]

[PeterNZer]

I'm really against giving players MORE control over their troops. After reading a lot of accounts recently I can conclude we already, arguably, have more control than we should.

Accept that the FOO made a mistake (get that all the time in real life and in CM, damn broken squads running backwards across open fields etc..) and move on. You did well to start opening up on a house suspected of holding a FOO, that happened all the time in WW2.

I don't believe it's an issue of axis or allied stuff. I can't even recall which side this dying FOO is on.

The example of shellshock is just to show that yeah, crazy stuff happens and it's not an unrealistic event.

In the end my position comes down to this.

1) If nearly 100% of the time units leaving a building run towards the enemy then I have a problem with it, since it's clearly a bit funny.

2) If around 50% of the time units run forward, and 50% backwards, then I have no problem at all since it covers shelling and fire blocking exits.

It doesn't matter whether the enemy unit is firring directly at the unit or not, large shells exploding against and inside buildings are generally unhealthy if you're inside them, worse if they are set on fire.

The key question is then.

- Are units running towards closest cover all the time, whether or not that cover is infront or behind? (ie. will they go for the bushes 10m infront rather than the woods 15m behind)

- If the answer is yes, BTS might want to make it a bit more random or throw in some direction decision before they run for cover. If no, then don't change it.

PeterNZ

O, Horncastle, yes you can only fit so many units in a building. Generally a squad an no more than ta platoon HQ or team in small ones and a couple of squads per floor in big ones (and a HQ or other team)

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

"Patriotism is the virtue of the viscious" - Oscar Wilde

"Don't F*CK with Johnny Cash!" - Chupacabra

[Guest Germanboy]

Originally posted by Horncastle:

Another problem with infrantry when under fire I have found is repeat behaviour when a building is occupied. Im not sure if there is a maxium capacity of troops a building can have ( sure there is ), but I have found troops running into a house to get in cover to find its full to run back out, come under fire, run back in and so forth.

This not only exposes them to more fire, but they are unable to return fire effectively as they are too busy running back and forth.

Honestly - why do you put so many troops into houses that nobody else fits in? It is plain crazy to do that. I can see no reason to change code to prevent something occuring that is down to a tactical failure of the player.

Repeat after me: Houses are not cover, they are tombs in waiting.

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

Andreas

Der Kessel

Home of „Die Sturmgruppe“; Scenario Design Group for Combat Mission.

[Guest tero]

>How would the FO know there was trouble in that direction?

He WAS ordered to call fire to a locatin in that direction.

Fuzzy logic does imply some sort of intelligence.

>Were there troops directly in front of him he could see?

I assume my esteemed PBEM opponent saw, through the eyes of the FO or other yet-not-encountered assets, at least some infantry sound contacts. Direct visual contact is not out of the question as the woods I was moving my troops through is within good LOS of the building.

>Were the enemy troops in cover at the time or in the open field?

In a forest, hiding after sneaking through it, with good LOS to the building the FO was in.

>You did state that the arty was still falling on your troops.

When the FO came out to pose for target practice, yes. That would indicate that the FO was not paniced and it had not lost LOS to the target point.

[Teutonicc]

Originally posted by tero:

>You did state that the arty was still falling on your troops.

When the FO came out to pose for target practice, yes. That would indicate that the FO was not paniced and it had not lost LOS to the target point.

When the game is over you may want to ask your opponent if there is another Fo or mortar located out there.

Teutonicc

[Guest tero]

>I'm really against giving players MORE control over their troops.

I for one am not advocating that. I'd just like to see some intelligence in the AI actions in extreme situations.

>After reading a lot of accounts recently I can conclude we already, arguably, have more control than we should.

The perfect game would be a sort of "go do" game where the player issues big orders and the watches them go ashtray. 30 minute turns or something like that.

>Accept that the FOO made a mistake (get that all the time in real life and in CM, damn broken squads running backwards across open fields etc..) and move on.

I accept mistakes. But it is things like this that are really annoying. How many times IRL would an AT gun track a squad moving laterally across its LOS until LOS is broken when a tank enters LOS near the guns original facing ? You just can not order priority target classes.

>You did well to start opening up on a house suspected of holding a FOO, that happened all the time in WW2.

I like to think I play historical games

>I don't believe it's an issue of axis or allied stuff. I can't even recall which side this dying FOO is on.

This debate has shown that issue is immaterial.

I hope.

>The example of shellshock is just to show that yeah, crazy stuff happens and it's not an unrealistic event.

When we take into account that FO's are supposed to be above average IQ, it was sitting unspotted and it is only turn 3 into the game I think the shellshock issue is far fetched.

I think the FO is veteran so it should not stir too easily when I do recce by fire with a Vickers HMG, 2" mortar and some other assets. Had it suffered casualties it was a different matter. But the shells kept coming in so the FO was not shocked out of its wits.

>1) If nearly 100% of the time units leaving a building run towards the enemy then I have a problem with it, since it's clearly a bit funny.

Funny ? I think that qualifies as the understatement of the day.

>2) If around 50% of the time units run forward, and 50% backwards, then I have no problem at all since it covers shelling and fire blocking exits.

Even when the back of the house is out of LOS and there is no way for the AI to start thinking it is surrounded when it has been monitoring the approaches which are all within relatively unobstructed LOS ? This early in the game to boot.

The fire started on the wall facing my troops and the FO exited the house from that same side. Fire blocking all exits is feasible if they had to jump out of the second storey and the back wall was facing a slope. But why did they jump from the side facing my troops ? Why not through a window on the two walls that would have allowed some coverage from LOS once they were out.

>It doesn't matter whether the enemy unit is firring directly at the unit or not, large shells exploding against and inside buildings are generally unhealthy if you're inside them, worse if they are set on fire.

2" (50mm) HE mortar shell and Vickers HMG round are not THAT large. The big stuff got cancelled or was still half a turn away from opening up.

If the house catches fire you get out as fast as you can. The turn is 60 seconds so the fire can not possibly spead like an explosion. Not with these weapons. Once it starts the FO should have good time to assess options and decide it was healthier to go through the window at the back IF the door faces the enemy.

>- Are units running towards closest cover all the time, whether or not that cover is infront or behind? (ie. will they go for the bushes 10m infront rather than the woods 15m behind)

I think there is a row of hedges some way in front of the house. But in the back of the house there is a slope running down hill and out of LOS to my troops until I get that far.

I think I should say IF I get that far...

>- If the answer is yes, BTS might want to make it a bit more random or throw in some direction decision before they run for cover.

I for one would like to see something to that effect.

[Guest Germanboy]

Originally posted by tero:

I think the FO is veteran so it should not stir too easily when I do recce by fire with a Vickers HMG, 2" mortar and some other assets. Had it suffered casualties it was a different matter. But the shells kept coming in so the FO was not shocked out of its wits.

I think the point here is that the FO does not know that you do recon by fire. It ahs to assume someone has got a bead on it, and then it should bug out before the big stuff coming in confirms that suspicion. 'Had it suffered casualties' it would be dead. FO teams are only two guys and can not afford to take any casualties. If they get under fire, it proves they were put into the wrong place by the player and should get out.

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

Andreas

Der Kessel

Home of „Die Sturmgruppe“; Scenario Design Group for Combat Mission.

[Horncastle]

Originally posted by Germanboy:

Honestly - why do you put so many troops into houses that nobody else fits in? It is plain crazy to do that. I can see no reason to change code to prevent something occuring that is down to a tactical failure of the player.

Repeat after me: Houses are not cover, they are tombs in waiting.

Well Im not actually saying Im cramming the place to the rafters for a big knees up.

Imagine this situation I have a platoon that I want to keep with in command of their HQ unit, the only decent cover is a building so I place the HQ and a squad in the building. I leave the other squad outside, it doesnt make tatical sense in such a situation to send them 200m to some woods really. When they then come under fire rather than retreat to some dead ground they run back and forth into the building and then outside.

Of course in the next turn I shall order them into dead ground, but there is a small problem with the AI in that it will allow repeated behviour such as this, and has little to do with my tatical incompetence( a major hurdle I must admit!).

As to the maxim "houses are tombs in waiting" I would very much disagree. Of all the most deadly and dangerous places to attack in a real war is the built up environment, due to the cover from weapons and sight that it affords those inside ( not to mention those who have time to dig in a fortify such places). As such it should be reflected in CMBO, in both tactics and game modelling IMHO.

Laters

[Guest tero]

>I think the point here is that the FO does not know that you do recon by fire.

Agreed.

>It ahs to assume someone has got a bead on it, and then it should bug out before the big stuff coming in confirms that suspicion.

Yes. But at what point ? Is the up to the player or the AI to decide ?

>'Had it suffered casualties' it would be dead. FO teams are only two guys and can not afford to take any casualties. If they get under fire, it proves they were put into the wrong place by the player and should get out.

I agree. But this entire incident occured during the 60 sec movie on turn 3 so it was an entirely AI run operation on my opponents part.

[PeterNZer]

Fair points.

I thought at first you had the building on fire from shelling buy a tank or something. When at work i tend to read posts a little less closely than I should

I agree it would be nice for them to have a little 'think' before moving. I believe at present CM doesn't really model how 'much' fire and stuff is going on. The building is either brand new or the towering inferno. I'm happy with that abstraction but I am interested to know how troop movement from the buildings has been implemented in game.

It's something I think would be interesting to try and test, and probably the direction they run tends to be entirely random.

I agree it could be argued that they should, 7 or 8 or 9 times out of 10 run towards your own lines, and perhaps, if paniced, run in any random direction, I guess it comes down to whether it's worth implementing at this late point. I'd be interested as to what BTS think about all this

Horncastle:

You're missing the trees for the wood, so to speak. While true that buildings in cities provide excellent cover and so on, it's not true if the enemy can stand 200m+ off and shell you.

If you are in a city, as you mention, then you wouldn't have the problem of finding somewhere to sit your men. If you're out of a city and relying on the cover of just one house when you're advancing then I'd suggest you'd be better off advancing over different ground!

I think for many players scattered houses in enemy territory are prime targets for a bit of pummeling and I'll often buy, as Allies, a HMC to do just that job while tanks go off and play (damn i love that HMC, cheap and effective).

PeterNZ

[This message has been edited by PeterNZer (edited 01-29-2001).]

[Guest tero]

>When the game is over you may want to ask your opponent if there is another Fo or mortar located out there.

I think I am about to find out...

I am however going up hill so an on-board mortar is highly unlikely as it would have to have LOS on my positions which in turn would mean it would be already exposed. Unless it is located inside a house...

An other FO is another matter.