Combat

[Binary]

In smac(x) when two units are fighting,how does the computer determine which unit fires first or how many times it will fire in a row?It seems to be partially random to me.

I ask this because sometimes when one of my units is fighting one of the computer's units,sometimes it will unfairly fire several times in a row,killing my unit,even if the calculated odds are in my favor and it's clear that my unit should win.

This seems to happen a lot when i'm attacking one of the computer's bases.For example: my 6-3-1 missile infantry unit with full health attacks a 1-3-1 plasma garrison with a perimeter defense,the odds are slightly in my favor,my unit fires once or twice,and then the computer's unit will keep firing until my unit is dead.

Am I the only one who's noticed this?

[WhiteElephants]

I can't say for sure, and maybe you should check the data links for a formula, but I'm pretty positive combat doesn't work in terms of who fires first and how many shots each unit has. The way I interpret combat is that each "round" the odds are calculated and the loser of each round takes damage. So you could lose several rounds in a row without ever "firing a shot". In the example you gave I don't really think that your 6-3-1 unit has very good odds against the 1-3-1 unit in a base with a perimeter defense. I'm thinking that the odds would be somewhere around 3:2 which means out of three attacks you are likely to only win once so the odds really weren't in your favor. Of course I don't know the morale rating of the troops invovled so I can't give a very accurate estimate.

[Sikander]

I believe that combat is conducted in the following way:

Strengths of both units are compared, and each unit is assigned a range = to their percentage of (strength X + Strength Y = 1).


A random number is generated from 0 to 1, and whichever side's unit was assigned that part of the range inflicts a hit upon the other side's unit. Thus if an unmodified Laser (2) attacked an unmodified Scout (1), then the Laser would hit on a random number between 0 and .6666, while the Scout would hit on any higher number generated.


At this point an animation is run showing the hit, and the hit points of the target are reduced. If the target has no more hit points, then an explosion is shown, if it still has hit points then another round is resolved.

[Binary]

Thanks for replying to my question,the datalinks should have had some information on this.

[RedFred]

A word of warning: something still doesn't seem right when the odds against native units are calculated.

The other thing that I have suspicions about is that when I have 2:1 odds I rarely seem to lose and when I have 1:2 odds I rarely win.

[WhiteElephants]

RedFred -- I know that when you start dealing with higher reactors against native life forms the percentages are skewed in the favor of the unit with the reactor. In this instance combat is still worked out correctly, but the odds given before combat are incorrect.

Sikander -- I don't know for sure, but I think in the example you gave the odds would be 50/50. I imagine the formula for the odds would be something like:

(strength of defenders armor/strength of attackers weapon) = percentage of chance for attackers to "hit"

or something like that... Where's that Marione character when you need him?

[Ogie Oglethorpe]

Unca Red,

I agree psi combat does appear to off the mark with respect to odds. Does anyone ever know if they fixed the odds calculation for fusion and higher reactors? I know that conventional units suffer damage so that reactor level is immaterial but the odds didn't show it. As a result say for example you are attacking with a fusion based unit and results were indicated as 4:1. then realistically you needed to divde your attacker rating by reactor (2) to get 'true' odds (actual 2:1).

Also has anyone noticed that the game gives you a break in the early game in that maurauding worms attack at much lower strength for the first few attacks then what they should? A hatchling should be attacking at a base of 3 to 2. So 'gainst a green unit a hatchling should be 3*.76 or 2.28 vs. a green unit at 2 *.88 or 1.76. In a base with +25% it should be 2.28:2.20. I've seen the first attacks as low as .45 in the first few turns. Low enough for a noncombat colony pod to survive in fact.

[WhiteElephants]

I agree with your observation about early native life. I've had formers survive early mindworm attacks on occassion.

In other news, I think I've confused myself about how to interpret the odds so I'll pose this question:

Would 2:1 odds indicate that the attacker is twice as likely to win and the defender or would it indicate that the both attacker and defender have a 50% chance of winning the battle? I'm thinking the later is correct and that would mean my posts above are incorrect.

[RedFred]

WE, your first suggestion about 2:1 is correct. I guess I didn't explain myself very well. What I meant was that instead of winning twice as often as losing it seems more like ten times as likely to win.

OO, you are right on that first hatchling. It seems too easy to defeat it. The other thing is that I seem to have poorer luck than you would expect going up against a fungal tower at even odds. Bad luck on my part, or a flaw in the odds?

[big_canuk]

2:1 odds means twice as likely to win with each hit point resolution. It seems it usually takes 4 resolutions to destroy a reactor 1 unit, so the final odds are much higher. Somebody help me with the math but it may be something like at 2:1 chance of winning is 1-(.33x.33x.33x.33)x100%. Vel put it as at 2:1 odds, you are very likely to win with about 50% damage to your unit.

Mind worms are more variable, cause it only takes 2 or three resolutions to destroy the unit. I think maybe some "extra" variability is added with native units.

When reactor levels go up, odds are still calculated on basic strengths, so be careful. If you have a 2 reactor fighting a 1, and your "odds" as the computer calculates them are 4:3, you often will not win, as your odds in each resolution are actually 2:3, but you have twice as many hit points as your opponent. Again, someone help me with the maths, but your odds are 2:1 cause of reactors, and then .6x.6x.6x.6 cause of strengths, so say double 15% is only 30% real chance of winning.

I think the formulas could be calculated, but again, there may be something "extra" included especially with native life. The moral of the story, is just as in real combat. If you want to win, go in with higher odds, but if you NEED to win, else disaster, then plan on having either overwhelming odds, or overwhelming numbers or better yet both.

[WhiteElephants]

Hmmm... Now that I think of it I'm not sure I like the way combat is resolved concerning the value of armor vs. the value of weapon. It generally plays out that the attacker has twice the value of the defender which would always favor the attacker. How often do you have 6 defense when you have 6 attack? I always thought the defender has the advantage in battle? I realize the defender usually has the advantage of modifiers (Children's Creche for a morale boost, Perimeter Defense, Rocky or Forested Squares, Sensors), but it isn't often that you get a combination of those factors to equal 100% which would merely even the battle not put the defender at the advantage. Anyway, just an observation, it probably has more to do with the layout of the tech tree than the way combat is resolved. No wonder the late techs are rarely of any real value.

[Googlie]

RF:

Fungal towers have a defensive bonus (see the readme.txt) not mentioned in the manual - I quote:

*Fungal Towers receive a 50% bonus when attacked, due to their ability to
co-ordinate and direct native life forms in the vicinity.


Is this always shown in the displayed odds?

G.

[big_canuk]

Googlie: The fungal tower +50% is in the odds

WE: The modifiers are multiplicative, not additive. If a 3+ defender in forest, with a sensor is attacked by a 6 attack rover, then the odds would be 6:3x1.5(for the +)x1.5(for the forest)x1.25(for the sensor)= about 6:3x2.83 = 6:8.5

The defender in this case does have the advantage.

Early game attackers often have the advantage, but later on, special units, and base defenses often give the defender the advantage. Mind you, those defenses can eventually be bypassed first with probes to disable them and eventually by blink troops, so the very late game may favour the attacker again. I don't know, cause I've never really had a battle there, but we are trying to engineer one in Nato Unravelled.

[WhiteElephants]

BigCanuk -- If that's the case then it does tend to even things up a bit.

[RedFred]

Thanks - especially to big_canuk for clearing up some stuff that has been bothering me for some time.