Before I submit an example with screen shots and saves, of a possible problem with the retreat mechanics.
I was wondering what conditions need to exist for a unit to retreat, and how the Losses Orders options (Min, Norm, Max)effect this, especially when defending.
Is it directly related to the number of casualties, or does training, experience, morale, level of deployment, possibility of being surrounded, and other possible variables that I have not listed also play a part?
What would be the most likely reason for a unit to retreat out of a good defensive position, after taking NO casualties, from the variables listed above?
Retreat Threshold
Moderators: Arjuna, Panther Paul
RE: Retreat Threshold
Suppression, Morale, 'shock' effects (tank vs nill AT capability, tanks alone in close terrain in poor visibility etc.)
Lack of cohesion (from movement, fires and formation changes) can also make units rather more brittle.
Casualties are often fairly low in real life (currently they may be rather high, especially for the ammunition expended), and while bloodless combat is probably far from the normal, it shouldn't be assumed that casualties are always expected from brief clashes of small units.
Wargames generally have 'advancing over a field of corpses' as the only way of moving forwards, but this isn't supported by operational results where 5-10% casualties over a day are 'heavy', and some units receive many more, and many units far fewer than this.
Lack of cohesion (from movement, fires and formation changes) can also make units rather more brittle.
Casualties are often fairly low in real life (currently they may be rather high, especially for the ammunition expended), and while bloodless combat is probably far from the normal, it shouldn't be assumed that casualties are always expected from brief clashes of small units.
Wargames generally have 'advancing over a field of corpses' as the only way of moving forwards, but this isn't supported by operational results where 5-10% casualties over a day are 'heavy', and some units receive many more, and many units far fewer than this.
RE: Retreat Threshold
Thanks for the fast response Lieste.ORIGINAL: Lieste
Suppression, Morale, 'shock' effects (tank vs nill AT capability, tanks alone in close terrain in poor visibility etc.)
Lack of cohesion (from movement, fires and formation changes) can also make units rather more brittle.
Casualties are often fairly low in real life (currently they may be rather high, especially for the ammunition expended), and while bloodless combat is probably far from the normal, it shouldn't be assumed that casualties are always expected from brief clashes of small units.
Wargames generally have 'advancing over a field of corpses' as the only way of moving forwards, but this isn't supported by operational results where 5-10% casualties over a day are 'heavy', and some units receive many more, and many units far fewer than this.
I love it when you reply, as its clear you are extremely knowledgeable when it comes to military matters.
If we narrow it down to just units defending in dug in or entrenched positions for now, and expand on your reply.
1.) Suppression - If a dug in or entrenched unit is heavily suppressed due to bombardment, or heavy incoming direct fire, with no enemy unit actually advancing on their position, at what point do they decide it would be safer outside their cover than inside?
Is this based on casualties sustained so far in that position? Would a defend order set to max losses, see them leave the position after taking more casualties than a setting of minimum losses?
I assume that their stubbornness trait plays a part in the threshold of this?
2.) Morale - Morale seems to be directly proportionate to casualties sustained, so ill skip that for now.
3.) Shock effect - This is a very interesting one. Do you happen to know at what range the shock effect comes into play?
As almost all units have some close quarters AT capability I assume this starts at more than 200m?
Do Armoured Cars have the same kind of shock value as medium, and heavy armour, or does it scale with the armour value?
Does the combat value of a unit effect the shock value of an attacking armoured vehicle?
For instance could an armoured vehicle with a combat value of 1 cause a shock effect on a unit with a combat value of 7?
Does existing ammo stocks also effect the AT shock resistance of a unit? What I mean by this is does the game AI take in to account if the unit has rounds left for the AT guns of a unit, when considering retreat due to shock effect, or does it just detect that is has the guns?
RE: Retreat Threshold
Dunno really - I just use these as rules of thumb for planning assaults (or the defence of a vital position).
Strip off the enemy supports (infantry or armour) and then mob him with stuff he can't deal with in quantity...
Suppress the heck out of him and overrun with a mobile unit... (either in attack or defence). This will usually *at least* force a retreat, and can rout and 'disband' a unit which can't extricate itself. Destroying units means they won't need to be defeated again later.
I know the shock effects were present back in the RDOA days, but I've not done systematic testing to confirm their importance - they are good rules of thumb though (armour cannot hold ground, unsupported infantry can't hold against combined arms assaults), and planning should consider them (even though armour obviously *can* hold some ground, it cannot rest/recover properly in the front line and is best used as a 'sharp edge' infrequently, rather than burnt out in the line - IRL it has limited security due to low personnel strengths and additional maintenance requirements - which isn't explicitly modelled - beyond the refuelling action).
Strip off the enemy supports (infantry or armour) and then mob him with stuff he can't deal with in quantity...
Suppress the heck out of him and overrun with a mobile unit... (either in attack or defence). This will usually *at least* force a retreat, and can rout and 'disband' a unit which can't extricate itself. Destroying units means they won't need to be defeated again later.
I know the shock effects were present back in the RDOA days, but I've not done systematic testing to confirm their importance - they are good rules of thumb though (armour cannot hold ground, unsupported infantry can't hold against combined arms assaults), and planning should consider them (even though armour obviously *can* hold some ground, it cannot rest/recover properly in the front line and is best used as a 'sharp edge' infrequently, rather than burnt out in the line - IRL it has limited security due to low personnel strengths and additional maintenance requirements - which isn't explicitly modelled - beyond the refuelling action).
RE: Retreat Threshold
As for units reaction to fire - there is a point where this is a deliberate 'rational' (if misguided) action, and other times when this is more involuntary. Commander's judgement, aggressiveness, unit stubbornness, aggression, training, experience etc will all play a role - green, untrained units led by rash/timid commanders may be noticeably less firm than elite veterans led by a steady and aggressive one. Permitting higher aggression, losses and keeping better exchange of suppression (by manipulating ammunition usage, if supply state permits) may make a reasonably steady unit more so... but may have less effect on a brittle unit.
-
BletchleyGeek
- Posts: 4460
- Joined: Thu Nov 26, 2009 3:01 pm
- Location: Living in the fair city of Melbourne, Australia
RE: Retreat Threshold
David's exposition is good and right on the money, I'll only give a more concrete overview of what is being taken into account.
Units in the engine react to what happens to them in a variety of ways, such as retreating. These reactions are triggered if certain conditions are met. 'Bad' reactions basically involve that the unit stops carrying its mission, either by bunkering down, retreating or routing. We refer to these triggers as unit morale failures. These failures depend on the following factors (each factor listed makes units more likely to undergo a 'morale failure'):
[*] Low task aggressiveness threshold
[*] If the task involves movement, low unit aggressiveness, if it's static, low unit stubborness
[*] Low unit cohesion
[*] Low commander aggressiveness
[*] Low commander determination
[*] Unit being deployed in 'open order' (as opposed to being deployed in defensive positions)
[*] Terrain cover
[*] Receiving fire from unlocated enemies
[*] Distance from HQ
[*] Unit under direct fire (modified by equipment, i.e. tanks don't care much about small arms fire, etc.)
[*] Unit under bombardment
[*] Unit under air strike
[*] Low unit training
[*] Low unit experience
[*] Passive armour shock effect (this basically makes infantry outfits more sensitive towards enemy concentrations of armour)
[*] Enemy attack shock effect (amplified if the attacking enemy has a clear superiority of armour)
[*] Nearby friendly units are retreating (so if a friendly unit pulls back, neighboring units will tend to be dragged along)
[*] Casualty rate
Soft factors in Command Ops are important and modify units' behaviour in many ways, as you can see.
If the unit undergoes a morale failure, several behaviours can be triggered:
[*] Retreating in good order
[*] Routing (retreating in disorder)
[*] Surrender
[*] Bunker Down
[*] Continue carrying the assigned task
depending on how bad is the unit current morale level and whether it is under fire or not. The most likely result is to retreat in good order. Routing is more likely to happen if the unit is already retreating. Note that surrender in CommandOps is rarely an "all or nothing" affair, units disintegrate over time. Note that a morale failure pushes units down a slippery slope: a unit that bunkers down is more likely to attract more fires, and hence, to retreat. Retreating units are more vulnerable, and their casualty rates go up, triggering more important morale failures, so they eventually rout. Routing units are extremely vulnerable to fires, and are prone to disintegrate while on the run if under pursuit by enemy forces or exposed to important amounts of enemy firepower.
Note as well that 'surrendering' in the engine not only accounts for capture, but also for what would be temporary MIA or AWOL personnel.
Units in the engine react to what happens to them in a variety of ways, such as retreating. These reactions are triggered if certain conditions are met. 'Bad' reactions basically involve that the unit stops carrying its mission, either by bunkering down, retreating or routing. We refer to these triggers as unit morale failures. These failures depend on the following factors (each factor listed makes units more likely to undergo a 'morale failure'):
[*] Low task aggressiveness threshold
[*] If the task involves movement, low unit aggressiveness, if it's static, low unit stubborness
[*] Low unit cohesion
[*] Low commander aggressiveness
[*] Low commander determination
[*] Unit being deployed in 'open order' (as opposed to being deployed in defensive positions)
[*] Terrain cover
[*] Receiving fire from unlocated enemies
[*] Distance from HQ
[*] Unit under direct fire (modified by equipment, i.e. tanks don't care much about small arms fire, etc.)
[*] Unit under bombardment
[*] Unit under air strike
[*] Low unit training
[*] Low unit experience
[*] Passive armour shock effect (this basically makes infantry outfits more sensitive towards enemy concentrations of armour)
[*] Enemy attack shock effect (amplified if the attacking enemy has a clear superiority of armour)
[*] Nearby friendly units are retreating (so if a friendly unit pulls back, neighboring units will tend to be dragged along)
[*] Casualty rate
Soft factors in Command Ops are important and modify units' behaviour in many ways, as you can see.
If the unit undergoes a morale failure, several behaviours can be triggered:
[*] Retreating in good order
[*] Routing (retreating in disorder)
[*] Surrender
[*] Bunker Down
[*] Continue carrying the assigned task
depending on how bad is the unit current morale level and whether it is under fire or not. The most likely result is to retreat in good order. Routing is more likely to happen if the unit is already retreating. Note that surrender in CommandOps is rarely an "all or nothing" affair, units disintegrate over time. Note that a morale failure pushes units down a slippery slope: a unit that bunkers down is more likely to attract more fires, and hence, to retreat. Retreating units are more vulnerable, and their casualty rates go up, triggering more important morale failures, so they eventually rout. Routing units are extremely vulnerable to fires, and are prone to disintegrate while on the run if under pursuit by enemy forces or exposed to important amounts of enemy firepower.
Note as well that 'surrendering' in the engine not only accounts for capture, but also for what would be temporary MIA or AWOL personnel.
-
Deathtreader
- Posts: 1058
- Joined: Tue Apr 22, 2003 3:49 am
- Location: Vancouver, Canada.
RE: Retreat Threshold
Lieste and Bletchley Geek thanks to both of you for your comprehensive and useful responses! [&o]
Rob.
So we're at war with the Russkies eh?? I suppose we really ought to invade or something. (Lonnnng pause while studying the map)
Hmmmm... big place ain't it??
- Sir Harry Flashman (1854)
Hmmmm... big place ain't it??
- Sir Harry Flashman (1854)
RE: Retreat Threshold
Yes, thank you both for taking the time to answer this so thoroughly.
Here is what I have found from the manual about the acceptable losses order option.
Change Acceptable Losses Level
Ħ Select the Losses button for an acceptable level of casualties
The Acceptable Losses level determines how many casualties your force will take before it warns you. Your units will continue to attempt to carry out their orders even after reaching this threshold, so in effect it is a reminder to you to consider changing your plan and issuing new (less costly) orders.
One exception to this rule is the Probe order ¡V units engaged in a Probe will Bunker Down ( see Launch a Probe ) if they assess the opposition is too great and setting a low Acceptable Losses level will cause them to do this sooner.
There is no mention of the fact that the acceptable losses actually does anything to effect the resolve of the troops, when defending, or indeed attacking.
It only states that there will be a message that the unit has taken an excess of casualties, which I have never seen actually.
Has the use of this button been changed since the manual was written, or have I been using it wrongly, assuming that it effects the resolve of a unit when in fact is does nothing of the sort?
Could you please confirm what effect if any the Losses [Min, Norm, Max] buttons have on a defending unit?
Come to that what effect if any, does it have on an attacking unit?
What would be an example of the "warns you" message?
Here is what I have found from the manual about the acceptable losses order option.
Change Acceptable Losses Level
Ħ Select the Losses button for an acceptable level of casualties
The Acceptable Losses level determines how many casualties your force will take before it warns you. Your units will continue to attempt to carry out their orders even after reaching this threshold, so in effect it is a reminder to you to consider changing your plan and issuing new (less costly) orders.
One exception to this rule is the Probe order ¡V units engaged in a Probe will Bunker Down ( see Launch a Probe ) if they assess the opposition is too great and setting a low Acceptable Losses level will cause them to do this sooner.
There is no mention of the fact that the acceptable losses actually does anything to effect the resolve of the troops, when defending, or indeed attacking.
It only states that there will be a message that the unit has taken an excess of casualties, which I have never seen actually.
Has the use of this button been changed since the manual was written, or have I been using it wrongly, assuming that it effects the resolve of a unit when in fact is does nothing of the sort?
Could you please confirm what effect if any the Losses [Min, Norm, Max] buttons have on a defending unit?
Come to that what effect if any, does it have on an attacking unit?
What would be an example of the "warns you" message?
-
BletchleyGeek
- Posts: 4460
- Joined: Thu Nov 26, 2009 3:01 pm
- Location: Living in the fair city of Melbourne, Australia
RE: Retreat Threshold
ORIGINAL: dazkaz15
Yes, thank you both for taking the time to answer this so thoroughly.
Here is what I have found from the manual about the acceptable losses order option.
Change Acceptable Losses Level
Ħ Select the Losses button for an acceptable level of casualties
The Acceptable Losses level determines how many casualties your force will take before it warns you. Your units will continue to attempt to carry out their orders even after reaching this threshold, so in effect it is a reminder to you to consider changing your plan and issuing new (less costly) orders.
One exception to this rule is the Probe order ¡V units engaged in a Probe will Bunker Down ( see Launch a Probe ) if they assess the opposition is too great and setting a low Acceptable Losses level will cause them to do this sooner.
There is no mention of the fact that the acceptable losses actually does anything to effect the resolve of the troops, when defending, or indeed attacking.
It only states that there will be a message that the unit has taken an excess of casualties, which I have never seen actually.
Has the use of this button been changed since the manual was written, or have I been using it wrongly, assuming that it effects the resolve of a unit when in fact is does nothing of the sort?
Could you please confirm what effect if any the Losses [Min, Norm, Max] buttons have on a defending unit?
Come to that what effect if any, does it have on an attacking unit?
What would be an example of the "warns you" message?
That's a question for Dave [:)]
RE: Retreat Threshold
No the manual wrong. It does affect game play as the AI will take it itno account when determining things like TaskDelay:: IsBlockingPositionUntenable(), TaskDoctrine::StandardPullBackReassessment() and TaskDoctrine::StandardBunkerDownReassessment(),
RE: Retreat Threshold
Thanks Dave.
I take it this is the one that is called during a Delay task to determine when to pull back to the next blocking position?
This must be the one that is called for a standard defend task?
and this one when attacking?
assuming that all the other variables, like stubbornness, training, experience, fatigue etc. once added together make 100% resolve to stay put.
What kind of difference do the loss [min, norm , max] make to the decision to stay put?
For example a setting of [min] losses would change their determination to defend the position by -5%, and a setting of [max] losses would make them 5% more determined to stay put or is the effect higher or lower than this?
No need for an exact answer just an educated guess will do.
I'm just trying to understand how much affect these settings have on the overall performance of the unit.
If a powerful unit in good order, in a good defensive position, with a setting of [max] losses, is pushed out by a far weaker but lightly armoured one, without taking any casualties, should I submit it as a bug, or can it be attributed to armour shock effect, or one of the other variables like experience or something?
The message that the manual mentions, about when a unit has reached its casualty threshold for the loss setting. Is this a feature that never made it into the game, or has it been accidentally dropped/deleted from the code?
I think it would be useful to me as a player, if it was reintroduced at some point, and it probably should be a red flash message.
ORIGINAL: Arjuna
TaskDelay:: IsBlockingPositionUntenable()
I take it this is the one that is called during a Delay task to determine when to pull back to the next blocking position?
ORIGINAL: Arjuna
TaskDoctrine::StandardPullBackReassessment()
This must be the one that is called for a standard defend task?
ORIGINAL: Arjuna
TaskDoctrine::StandardBunkerDownReassessment()
and this one when attacking?
assuming that all the other variables, like stubbornness, training, experience, fatigue etc. once added together make 100% resolve to stay put.
What kind of difference do the loss [min, norm , max] make to the decision to stay put?
For example a setting of [min] losses would change their determination to defend the position by -5%, and a setting of [max] losses would make them 5% more determined to stay put or is the effect higher or lower than this?
No need for an exact answer just an educated guess will do.
I'm just trying to understand how much affect these settings have on the overall performance of the unit.
If a powerful unit in good order, in a good defensive position, with a setting of [max] losses, is pushed out by a far weaker but lightly armoured one, without taking any casualties, should I submit it as a bug, or can it be attributed to armour shock effect, or one of the other variables like experience or something?
The message that the manual mentions, about when a unit has reached its casualty threshold for the loss setting. Is this a feature that never made it into the game, or has it been accidentally dropped/deleted from the code?
I think it would be useful to me as a player, if it was reintroduced at some point, and it probably should be a red flash message.
RE: Retreat Threshold
What Bletchley_Geek lists in post #6 is very interesting.
Currently I have the feeling that units tend to hold out good as long as it looks ok but when the troubles start they seem to crumble very fast.
As he said: "Note that a morale failure pushes units down a slippery slope" and that seems very true and like a very steep slope.
Currently in Elsborn Ridge I have already destroyed 7 units of the US frontline and it's just past noon.
Höfen Ho-Down was a similar experience, I crushed some US units but I lost a company on the attack of Höfen North as it was again and again pounded by artillery and direct fire.
Overall it seems almost like the opposite of the way the game ran before this patch.
Currently I have the feeling that units tend to hold out good as long as it looks ok but when the troubles start they seem to crumble very fast.
As he said: "Note that a morale failure pushes units down a slippery slope" and that seems very true and like a very steep slope.
Currently in Elsborn Ridge I have already destroyed 7 units of the US frontline and it's just past noon.
Höfen Ho-Down was a similar experience, I crushed some US units but I lost a company on the attack of Höfen North as it was again and again pounded by artillery and direct fire.
Overall it seems almost like the opposite of the way the game ran before this patch.
RE: Retreat Threshold
Something happened between 4.6.269 and 4.6.270 that makes units reluctant to surrender even when reduced to almost nothing (as in, 1 man, 3 men, 6 men + 1 AV). Anyone else noticed this? E.g. I see three units with red zero's for power, adding up to 8 men in all, who have not surrendered or even fled anywhere for 20 minutes.
Edit: in fact units are not surrendering at all it seems; 5 man units are recovering from routs and fighting on.
Edit: in fact units are not surrendering at all it seems; 5 man units are recovering from routs and fighting on.
Have: Socks. Deodorant. £2 gloves. Mince pies.
Want: Line formation banned until I give a specific order to use line formation. Troops that don't take lie-ins until 0800 unless ordered to never rest.
Want: Line formation banned until I give a specific order to use line formation. Troops that don't take lie-ins until 0800 unless ordered to never rest.
RE: Retreat Threshold
Are they still in contact/engaged?
I have been finding that they will route off with just a few left as you say, but they go down quite quick, especially if pursued.
Unless you have the halting bug problem, where they get stuck on the flashing in and out of sight thing that Dave described.
My experience with regard to surrender so far, is that units will surrender a large portion of their men in just a few minutes.
I have hade a few Coy's with very light casualties surrender half their manpower in a second, when facing down a large powerful force.
This is desirable though is it not, especially when a faster force is attacking a slower one, as it simulates them being over run?
I have been finding that they will route off with just a few left as you say, but they go down quite quick, especially if pursued.
Unless you have the halting bug problem, where they get stuck on the flashing in and out of sight thing that Dave described.
My experience with regard to surrender so far, is that units will surrender a large portion of their men in just a few minutes.
I have hade a few Coy's with very light casualties surrender half their manpower in a second, when facing down a large powerful force.
This is desirable though is it not, especially when a faster force is attacking a slower one, as it simulates them being over run?
RE: Retreat Threshold
Engaged. I've had 125 of my infantry directly on top of a 7 man strong company for about an hour now. No surrender or retreat. I'll post a save if anyone is interested.
Have: Socks. Deodorant. £2 gloves. Mince pies.
Want: Line formation banned until I give a specific order to use line formation. Troops that don't take lie-ins until 0800 unless ordered to never rest.
Want: Line formation banned until I give a specific order to use line formation. Troops that don't take lie-ins until 0800 unless ordered to never rest.