Retreat in SC2?

[bloodstar]

Short and sweet. Will defenders retreat in SC2? or will they fight to the death as in SC1?

(Personally I think retreats add more verisimilitude to a war game.)

thanks!

Mark

[pzgndr]

It's a definite maybe at this point. Hubert has some ideas, but we'll have to see what works and what doesn't before announcing anything.

[Exel]

I definitely support the idea of making units retreat. That'd add a lot to the realism of amphibious landings, as you could no longer prevent them simply by blocking the beaches with guard units.

[ April 29, 2004, 01:11 PM: Message edited by: Exel ]

[Night]

It has been said before about how units being totally destroyed is unrealistic. A retreat opition would be nice.

I think the player should be able to set certain unit's to no retreat, like a check box on the side or bottom when you have that unit selected. Since in some cases retreat would serve to hurt the unit more. For ex. If you are on a front line and only 2 units can hit you and they do but make you retreat, then tiles open up for more units to come in and finish you off as well as breach the line and hit other units, whereas they would just hold with 2 or 3 str until next turn.

[Edwin P.]

I think the player should be able to set certain unit's to no retreat

Excellent idea.

[SeaMonkey]

What happens when you have to retreat to an occupied tile and there is no stacking? What effect(s) are enemy ZoCs going to have on a retreating unit? Like I said before, think abstractly, the unit is not destroyed (how many corps/army sized units were completely destroyed in WW2), it has been made combatively ineffective. You must reorganize, refit......ie rebuild the unit from its cadre.

[bloodstar]

you could simply have it lose additional readiness and strength an be pushed back to the nearest non occupied non enemy zoc hex, skipping over allied units...

of course if it has to do it too much, the additional strength losses could eliminate the unit.

and we all would agree you would never be allowed to retreat into an enemy ZOC...

I hope...

Mark

[Bullwinkle]

I think retreat could be an option for highly trained units - the SS for example. But general cadres would be eliminated.

Naturally the combat ratings of the unit would be seriously diminished when retreating.

If there were no areas to retreat to because of friendly units, I would suggest that the unit behind it be augmented in combat factors by a point or two to reflect the addition of trained personnel from the defeated unit.

[KDG]

What happens when you have to retreat to an occupied tile and there is no stacking?

The unit is destroyed instead if it can't retreat.

[Night]

I still think the idea of setting some to no retreat so you can customize is the best choice if you are going to put in a retreat opition

[ev]

Originally posted by Bullwinkle:

I think retreat could be an option for highly trained units - the SS for example. But general cadres would be eliminated.

Naturally the combat ratings of the unit would be seriously diminished when retreating.

If there were no areas to retreat to because of friendly units, I would suggest that the unit behind it be augmented in combat factors by a point or two to reflect the addition of trained personnel from the defeated unit.

In general, mechanized units should be more likely to succesfully retreat. From my readings of WWII, infantry had a very hard time retreating when attacked by fast tank units, while mechanized units were most often able to get out of harms way. There is a publication of Rommel's Diary with great notes on his battle at El Alamein and his concern that foot infantry would not be able to retreat after a British breakthrough.

Foot infantry only had a fair chance to retreat from attacking mechanized units when defending closed terrain (hedge rows, woods, hills or mountains, and the like) or when defending river crossings. Otherwise, infantry retreating from tanks usually suffered huge losses.

...by the way, I would not allow air units and naval units in port to retreat.

[ev]

...and, by the way on Bullwinkle note, I agree with the spirit of Bullwinkle's suggestion. Troops with more experience, troops that are better supplied, troops that are attached to a better HQ, and, specially troops with mechanized transport should have a better chance of retreating.

I understand retreating in front of a superior enemy is a most difficult and riky maneuver. I am for allowing retreat, but, there should be a good risk model associated with it.

[Shaka of Carthage]

ev and Bullwinkle

Your examples and your ideas are at the wrong scale. We are talking about multi division actions performed by Corps and Armies. And don't mention the ability in SC2 to edit the scale. Right now we are discussing what the basic SC2 system will deliver.

I'm not saying that a unit shouldn't retreat, I am saying we should understand the scale of the unit we are talking about and offer suggestions in relation to that.

So, in my mind, I would favor a solution that made the losser of a combat, when he lost by a certain ratio, be forced to abandon his position (ie retreat).

[Desert Dave]

An awful lot of good points here, and that's not even the half of it.

There is the difficulty of how and when the AI selects a route for retreat?

How much more TIME consuming for the ACTIVE player to set up his retreat orders for next turn? How long will you want to wait while the opponent "fusses" over his retreat orders?

And what should the "standard criterium" for the AI be?

Retreat from cities and forts, yes or no? Ever? Into mountains? Across rivers?

What of air superiority? Or, armor advantage? How about HQ in range, or not?

What of blocked retreats, and which is the alternative route that is fairest and most easily managed?

The list is long. No doubt everyone can think of 20 others for each one that I have stated.

Well,as Bill has said, it's... a definite maybe.

Which means, of ALL the various components that must, and WILL fit together seamlessly, this one is one of the most problematic.

Personally, I very much like the "chess-like" simplicity of SC the Original.

You don't have "retreats" in a game that has lasted since Middle Ages or forever, as Chess, true?

Why here?

Just because other games have done it that way? And sure, I agree, if you have ANY sort of battle or skirmish, regardless of the kinds of implements of War, the losing side will often retreat.

But there is something to be admired in REQUIRING the "master strategist" to set up his forces beforehand, with each game-piece fulfilling a particular function.

For instance: there are units that are static, and assault the defensive position... usually those with a lot of "firepower" so that the defending unit is heavily damaged (... after, a lesser unit or 2, or the Strat Bomber has weakened the fortification or entrenchment) and THEN... the "blitzing units" will take advantage of the ensuing break-through.

(... how frustrating when you have done all you can and yet... that defending unit stands there and laughs and defies you with a strength of...one. But that is also part of the "charm.")

IF a unit has "geared up" and mustered a cache of supplies and prepared for an all-out assault, why should they then get to move also?

And why, necessarily, should a defensive unit HAVE TO retreat?

This is VERY large scale gaming.

It might be assumed that the "rear guard" is established by a competent HQ (... at this scale, nothing prevents the imagination from "moving" that rear guard into place) by a near-by battalion or even a company, whose sole function is to... delay the advance.

Anyway, I really like the different and unique challenge of trying to "set-up" your offensive units so to maximize a possible break-through. :cool:

That is what we have now.

Beta testing will determine if any sort of retreat option will work... "seamlessly," and really, either will be fine by me, but

As said... there are many, MANY potential problems with putting in retreats, and they just may not finally be worth the toil & trouble.

Yes?

No?

[Edwin P.]

I agree, at this scale I don't see retreat as a required feature. I would much rather see time spent on fine tuning features with more strategic impact, such as operational movement, or improving the AI. (in a fast game tonight vs the AI I launched a sea lion and the UK Air Units bombed the empty hex of Brest instead of my transports off the coast of London)

[ May 04, 2004, 01:26 AM: Message edited by: Edwin P. ]

[bloodstar]

I guess the way I look at retreat is an automatic realignment of forces. perhaps on this scale, consider it to be a moderate breakthrough forcing the unit in the hex to reposition themselves further back to prevent dissolution. I do agree that only low strength units should have a chance to retreat. A unit that hasn't taken losses won't be retreating unless they get the order.

I suppose the way I look at it, think of retreat as an alternative to destroying the corps/army/army group. The rules can be as simple as the following:

a) when strength is less than 4 and

readiness is below a certain level

a unit has a chance of retreating if it would otherwise take enough damage to be destroyed.

if a units readiness drops below say, 0, then the unit is destroyed anyways, and make retreat take a big hit out of readiness. so a unit can only retreat once (or maybe twice) in a turn.

not knowing any real details about the combat model, I'm presuming that each attack takes a certain amount of readiness from a unit. so one big massive attack can do a lot of damage and yet not drop readiness too far. Hence, to simulate a massive strike, if readiness is high, the unit won't retreat right away.

I would also say a unit with an entrenchment above say... 2 won't retreat, instead preferring to hunker down. this should solve the question of holding cities and mountains, etc.

heck, if you want to be really fancy, you could have a 'no retreat' order (which under this concept would be counterproductive, as a unit won't ever try to retreat unless it would be destroyed anyways...but might be useful in a nicer retreat system). So that a unit is much less likely to try to retreat when attacked. But such an order should only be a global order. (don't want to have the players spending time clicking on units and deciding to retreat or not).

ok, with that in mind, which direction would a unit retreat?

Make it a simple sequence, if it's surrounded by enemy ZOC, it can't retreat, even with other units adjacent. it simply dies.

if it can retreat, it will retreat in the most opposite direction that the attack came from allowable. if it can retreat directly back, it will. otherwise if it still has a direction to retreat, it will go in the most opposite direction permitted. In a case where 2 directions are equally viable, simply go 50/50.

if you want to go a step further, subract the units mps so they have less the next turn. (and if you want to be really nasty, let a unit that has been retreated only be able to move and not attack.)

The image of the formula I have in my mind is:

unit has taken enough damage to be destroyed, check retreat

percent chance to retreat =

1 + UXP + URDY * 10 + LMECH + 5(if ARMOR TRUE)

UXP = Unit Experience

URDY = Percent unit readiness

LMECH = Level of Mechanization tech on unit

and a generic 5% bonus to retreat if it's an armor unit.

(you could also have other tech levels affect a units chance to retreat. and perhaps even have a negative chance factored in if the units involved are high mech, high experienced armor units. Creating a formula to determine when a unit retreats isn't difficult... the difficult part is making the formula reasonable.)

so a inf with 4levels of experience and a readiness of 40% with 5 levels of Mechanization would have a:

1 + 4 + 4 + 5 = 14% chance to retreat.

Not a great chance, but that's a chance that an otherwise destroyed unit lives to fight one more battle.

Retreat is not a game breaking issue to me, but I do feel it adds a feel of verisimilitude to any wargame.

Thoughts? Comments?

thanks!

Mark

[ev]

Yes bloodstar, I like your proposal very much.

[SeaMonkey]

I understand that retreats are realistic, but I agree with Edwin and DD, its not appropriate at this scale. I'm kind of hesistant on this issue of being for or against. But I'm thinking I'll see a cluttered battlefield and these retreating units could hinder breakthroughs, unless they lose their ZoCs. Now I'm as guilty as the rest of us, but I must state that we need be mindful of the AI competence when providing for many decisions.

[ev]

No one ever speaks about advance afer combat. Most of the time advance after combat would not be interesting in SC. But, there are a few cases where advance after combat would be important:

1. where a unit is surrounded and trying to reunite with the main body.

2. where the attacker completely anhiliated the defender of a city, but there is no other unit to ocupy.

3. where a paratrooper or amph "landed" on an enemy occupied hex.

4. when a unit is trying to break out of the beachhead where it landed the previous turn.

Also, I have not heard about the possibility of multiple attacks by highly mechanized units.

Any ideas or comments on any of these.

[Exel]

Advance after combat has been suggested (by me among others) for tanks as a new "blitz breakthrough" option. That'd increase the significance of tanks as breakthrough exploitation units they historically were. But I wouldn't allow other units to do the same, only tanks.

EDIT: Of course the above would be possible only if the unit hadn't already moved, so it couldn't move-attack-move, only move-attack or attack-move.

[ May 05, 2004, 03:56 PM: Message edited by: Exel ]