air intercepters question

[Roland Johansen]

In Conquests mods I set it so units had a higher chance of retreat. Also set bombardment, by air, land and sea, to non-lethal until the very last level of units were reached. This was to represent that air attack and naval bombardment did not usually completely destroy a force. The only exceptions for the earlier units is I allowed air to destroy ships. I gave the later units hit point bonuses and higher rates of fire to minimize the spearman vs tank syndrome. I enjoy trying to make the game more in tune with history. To me that is the fun part of this game. Just leaving it at a attacks b and one of them is destroyed is boring. And not very historical. Most of the time, one of the opposing was not annihlated, but we usually hear more about the battles where one side did get annihlated and so think that was the norm in war.

It's boils down to personal preference in how one would like to see this represented in a game. Panzer General did a pretty good job of representing how military units usually were ground down incrementally rather than outright destroyed in most circumstances. One usually had to surround them on most sides to get a surrender or annihlation of the unit. I prefer to model the combat in civ along those lines. Well as much as I can.

I agree that rebuilding damaged units should cost. This is one of those things that should be in civ, but wasn't. Disagree about having units damaged rather than destroyed being a balance issue. For what I've seen in mods I've made, it adds more depth and enjoyment to the game by offering more strategy.

In civilization 3, the retreat value from the editor wasn't the real retreat chance. There has been done some testing of the real retreat formula a few years ago and finally it was discovered that the formula was as follows:

(100*Attacker Retreat Bonus)/(50+Defense Retreat Bonus)

This actually means that if you give two units a retreat chance of 100 in the editor and they fight one another, then the loser has a 2/3 chance of retreating. In general, the retreat chances are lower then the ones you set in the editor if you would use high values in the editor. I modded civ3 too and I understand that it can be frustrating to learn that the formula isn't what you expected it to be.

The article that I mention can be found here. The last post mentions the formula.

As long as civ has no repairment cost (building a new knight costs as much as repairing two half strength knights to full strength or something like that), I don't think it is a good idea to implement high retreat chances for units. The human player will be far more able to use this then the AI. Especially because the AI was not programmed for that kind of combat. It's also more difficult to program an AI that can perform a war of attrition and that is good at avoiding escape of the damaged units of the human player.

You should also know that retreats in civ4 have been changed from civ3. Now only an attacker can retreat, making it a lot less powerful. It's not that well documented, so maybe you missed that and that could be frustrating when you start modding civ4.

 

[Roland Johansen]

Thanks for the retreat info.

I don't think I've ever seen the ai retreat from one of my attacks, so that they can in Civ3 is new news to me. I modded Conquests to give the units more movement and have noticed the ai move forward and attack a unit, then use the remaining movement points to retreat to a city. Could have been to reinforce the city, I've no idea how many units were in that city or it could have been to repair damage or both. In another mod, the ai moved a stack of damaged units into a city to relieve the undamaged stack residing in that city who then headed over to where my units had just took another city (as the city was closer to my units this was probably an example of a relay). I've been experimenting with these faster unit speeds and have found the ai is a bit smarter because of it. Could be the lack of movement in the original game is affecting ai behaviour. Since the ai apparently forgets what it did, or what happened in the game, the turn before. If an ai got damaged that turn, but had no more movement left that turn, it doesn't remember the unit is damaged during movement the next turn.

There are lots of instances where a unit cannot retreat in civ3 while it is a fast unit.
1) Both units involved in the fight are fast units.
2) A defending fast unit will not retreat when it is fortified on a city or a colony.
3) The fast unit starts the battle with 1 hp.
4) The defending unit has only 1 hp left. The fast attacker will try to kill it, even if it reaches 1 hp, it will continue the fight.
5) When there's no room to retreat to.

In civ3, the AI would indeed retreat units after a battle if they were wounded and had hitpoints left. However, if the units were blitz units, then they would often/always attack to the death.

I believe that we have totally changed the topic of this thread. Maybe, we should stop this discussion about civ3 in a civ4 forum.

 

[Snerk]

Hi,
I'm new in this forum and a litte bit concerned that I'm too late in this thread, as the last post is over a week old already.
I read almost all of the posts in this thread, more and more amazed that nobody complains about the issue I want to address:
Why isn't it possible in Civ 4 to gain air superiority and why does nobody care?

Yes I totally agree, it seems like air superiority isn't a consept in this game.
In my little experience it seems like the only way of taking control of the skies of another civ is to capture their cities. An alternative is trying to constantly keep their aircrafts cripled/damaged. But for that you need a superior air force wich is difficult on noble or higher difficulties. I miss the feeling of fighting battles in the skies. I allmost just use my aircafts for intercepting and destroying improvments. I thinks there should be some way of attaking planes. As far as i can see its impossible for any unit to try to directly seak out and attack an enemy aircraft. Fighters should defininitly be able to do this.

 

[kernok]

IMO the problem is not about fighters that are not able to attack others fighters with a special mission button (these kind of mission doesn't not exist in real life) but it is all about:
a fighter going through an air territory protected by another fighter should engage a "dog fight". in this kind of air battle where both fighters have a chance to damage the enemy plane odds to win have to be different than 10% for attacking plane versus 90% for defending one

 

[Bushface]

If indeed one unit of, say, bombers represents a squadron, then it is scarcely reasonable for a successful interception to prevent any surviving raiders from dropping their bombs and doing some damage, though less than would have been caused by an unintercepted raid. As things stand in Civ4, any damage to the attacker causes the raid to be a total failure. This is quite unlike real life: consider the famous Dam Buster raid in which several aircraft were lost but two of the three targets were destroyed, or the damage done in London during the 'blitz' despite the many Nazi planes that were shot down, to mention but two examples.
I accept that this does not help in resolving the original question in this thread, that of the chances of interception, but it is relevant to calculating the results of such interceptions.

PS - why is 'interception' such a long word ? Tires both my index fingers.

 

[Arkturus]

IMO the problem is not about fighters that are not able to attack others fighters with a special mission button (these kind of mission doesn't not exist in real life) but it is all about:
a fighter going through an air territory protected by another fighter should engage a "dog fight". in this kind of air battle where both fighters have a chance to damage the enemy plane odds to win have to be different than 10% for attacking plane versus 90% for defending one

You are right with the special mission "Shoot down all enemy fighters". In real life, no interceptors would circle over friendly territory without any threat. They would rather be sent up when an attack is underway.
Thus, we don't need a new mission for shooting down enemy fighters. What we need is to modify the current attack mission in a way that the attacking fighter has a better chance of damaging the defender.
Despite of that, there is another thing that should be changed: When sending a bomber on an attack run, own fighters on intercept mission in range should engage the defending fighters when they try to intercept the bomber - that would result in a kind of "bomber escort mission", which was also requested by some users.
I think I will report a bug for the above mentioned issue. So far, I found nobody in this forum who had an argument why the defender should have such a great bonus. And nobody disagreed that this should be changed, either.
Unfortunately, however, it seems to be impossible to do this change by the current modding possibilities. I hope they will fix it in a future patch.

If indeed one unit of, say, bombers represents a squadron, then it is scarcely reasonable for a successful interception to prevent any surviving raiders from dropping their bombs and doing some damage, though less than would have been caused by an unintercepted raid.

That's again the big question that is discussed in several threads in this forum: Does one unit in Civ represent a single vessel, say tank, or is it a squadron ore something like that? In my opinion, this is not possible to decide: In case of a warrior unit, for example, it's for sure more than one man (it is even shown as three men, actually) - whereas in case of a tank it could be just one vehicle. I don't think this is already clear - in my opinion the game behaves more as if one unit represents just one tank/bomber/fighter... especially the navy units are of this kind. But I'm far from sure about that.