Combat Formula question for .exe modders.

[Sanctus Ferri]

How difficult would it be to change the combat formula so that rather than comparing the attacker's attack value to the defender's defence, it would compare each unit's attack to the opposing unit's defence instead?   

[Buster's Uncle]

...Moving to Bug/Patch and bumping...

[scient]

Could you describe a bit more what you are trying to achieve? I don't quite follow.

[Sanctus Ferri]

The only thing I disliked about this game was that combat was resolved by comparing the attacker's weapon to the defender's protection, and ignoring the weapons of the defender or protection measures of the attacker.  The attacker could be destroyed by the defender’s armour, essentially, which is conceptually troubling.  I wanted to do away with that, and instead have the attacker and defender simultaneously attack each other every round, essentially, attacker's attack vs. defender's defence and defender's attack vs. attacker's defence.  The defender's defence would still benefit from defensive bonuses, but not their attack vs. the attacker's defence. 

[Yitzi]

How difficult would it be to change the combat formula so that rather than comparing the attacker's attack value to the defender's defence, it would compare each unit's attack to the opposing unit's defence instead?

Fairly easy, though if it doesn't come with other (somewhat harder, but also not all that difficult) changes, it would make there be not much difference between attack and defense values.

[Sanctus Ferri]

Thank you, Yitzi.  Yes, I see what you mean about that.  I didn't want to start thinking too grandly until I had an idea of what was reasonable.  Ideally, I would like to use some kind of s-curve probability function like the attached graph of probability of kill for armoured fighting vehicles, given effective armour thickness vs. weapon penetration, but I need to derive the function first. 

[Nevill]

How difficult it would be to make HP values for various reactors in alphax.txt correspond to actual values, i.e. make them soft-coded, rather than hard-coded? How would that affect Psi combat that currently has units, say, with Fusion reactor lose HP in combat at double the rate?

[Yitzi]

Thank you, Yitzi.  Yes, I see what you mean about that.  I didn't want to start thinking too grandly until I had an idea of what was reasonable.  Ideally, I would like to use some kind of s-curve probability function like the attached graph of probability of kill for armoured fighting vehicles, given effective armour thickness vs. weapon penetration, but I need to derive the function first.

Actually, what I'm considering is to have round-based combat, and damage done each round equals weapon - armor + a random variable (normally distributed with mean of 0 and standard deviation of 2).  (Plus an extra bonus based on weapon/armor types, because that system makes things more interesting and IMO would allow for a more interesting alternative to the current resonance/pulse weapons and armors).

One consequence of this system would be that armor significantly ahead of the weapon would be nearly unbeatable (though a stalemate might be achieved, giving the technologically inferior party time to catch up...plus, nerve gas could be reworked to give a small damage bonus regardless of armor), while the costs would be such that weapon slightly ahead of the armor would have an advantage, but too far and it wouldn't be worth it.

How difficult it would be to make HP values for various reactors in alphax.txt correspond to actual values, i.e. make them soft-coded, rather than hard-coded?

Should be fairly feasible.

How would that affect Psi combat that currently has units, say, with Fusion reactor lose HP in combat at double the rate?

However we want it to (within reason).