If you follow the link in my signature you will also find a link to the CCV forum. In the DL section you will find a list of mods that are in use and links to descriptions. But I must confess that not all of them are up to date. But you will get an idea of the current version.
Solving the SoD issue
- Thread starter QuantumProjects
- Start date
duckstab
Child of Noble Family
Well, I didn't claim it was going to be easy, and I actually anticipated a couple of these potential problems.
To overcome any shortage of units in the early game I was thinking of having a floor value, again scaled to difficulty. So the limits wouldn't even likely be a factor early in the game. As for unit composition, I was already thinking of having separate limits for land, naval, and air units. But you're absolutely right, the AI needs to be able to deal with its limits intelligently and either upgrade or delete units when at or close to the limit.
I think a more serious problem is that the current AI almost needs SoDs to be in any way effective. Humans can typically get away with smaller numbers of units, but my main problem with that is how it messes with diplomacy.
I hated stacks of doom in Civ4. The simplest means to control military is to completely eliminate the amount of "free" support a military gets. A Civ has to pay upkeep for each and EVERY military unit. And this upkeep cost should increase for later units to account for techs enabling a more powerful economy. This is not a hard limit on unit count, but certainly does mean a civilization cannot simply have as many units as they can produce. It 's also possible to increase the amount of support a unit requires when in enemy territory, thus making waging war more difficult.
kiwitt
Road to War Modder
I have done this as well in my mod.
An Army Groups costs 3 times an Army. An Army costs 3 times a Corps. A Corps costs 3 times a Division
And then I simply doubled the costs for Tank units , i.e. a Tank Army would cost 6 times a corps, and 3 times a tank corps, etc.
flitz
Warlord
- Joined
- Sep 16, 2009
- Messages
- 245
I don't agree.
The advantages of a SoD over smaller groups are:
1. That a SoD allows many units to arrive in one place at once. If you limit units/tile then you might get problems maneuvering through bottleneck spots. So some xUPT-scheme is bottleneck-friendly one could say.
2. If you attack a SoD with another SoD, your attackers will most likely meet fresh defenders from the opposing stack for quite a while because the hurt defenders will just sit back and not defend anymore.
But if a SoD is surrounded by many smaller groups, it can weaken the members of a group with first attacks and then finish them off because they have noone to guard them from the second wave of attack. Then in the opponent's turn, his small groups cannot do the same to the SoD because of the described issue. That is my main point why amassing troups in one spot gives an advantage - and thats why mechanics like surround&destroy or colleteral damage try to weaken it.
I tend to like the xUPT idea for some unit number around 20. It just feels more appropriate somehow.
And if the AI already takes gold production into account when deciding about unit production, then I'm also for raising maintenance.
Acronym2
'Cedo Alteram'
I used to feel that way, but Antmanbrooks does make a good argument. Expert players still seem to win on very high difficulty settings in spite of large stacks of enemy units. Maybe he has, in part, a valid point. I would simply like to see the scale reduced a little bit. One thing I have noticed is that if one doesn't mobilize for war but just thinks they can continue "business as usual" and let their standing army fight off the invasion is going to lose cities......possibly even get destroyed if the AI attacker estimates you are too weak to resist being conquered.
Wait, are you sure about this? Because in a personal xml mod of mine, I've been getting periodic CTDs, and I've given a lot of units (most archery, tanks, anti-tank inf, sam inf, mobile sams, and gunships) collateral damage. Would there be any way of getting around this problem? If I were to make these other units unitclass siege, would that fix the problem?
Anyways, I think SoDs are appropriate for pre-modern warfare. However, by the time of the advent of machine guns and artillery, there need to be incentives to engage in warfare more along fronts. I've done this by making the number of units given collateral damage equal to the strength of the unit. So artillery does collateral damage to 18 units (well, 17 actually, since I decreased their strength value). Mobile arty does collateral damage to 24 units. Now there is an incentive to spread out one's units in the modern era into little packets of 4 or 5 units with complementary counters. (Of course, there's the little problem of the AI still not being programmed to do this). I've also, though, decreased the damage cap to 20% for cats/trebs, 40% for cannons, and 50% for artillery. First-gen aircraft only do 20% collateral damage, but they will inflict it to 16+ units. I find that this provides plenty of incentive for me, at least, to not rely on massive SoDs during the modern era.
Dancing Hoskuld
Deity
Yes, I am sure. It was in Rise of Mankind. I was just an observer of the discussion at the time. I think it may have been put in the to hard basket. Since there were only a few units with that capability those units were removed or changed so that they don't cause collateral damage.
I'll check with some others to see if a fix was found.
kiwitt
Road to War Modder
I have not being getting CTDs for my units that are non-siege doing Collateral Damage. But I have had random CTDs
flitz
Warlord
- Joined
- Sep 16, 2009
- Messages
- 245
I agree with you. But that is not a problem with the SoD, but rather with military focus in this game altogether.
Maybe what leads to war so often in this game is the underlying simplistic system that still, more cities automatically mean more research, more culture, more influence, more parallel production and (when caring a little about timing and placement) usually more money. Sure, city-spam has been penalized in CivIV compared to CivIII, but the system still works that way. And the question now is not only 'How do I get as many cities as possible?' but more like 'How do I get as many cities as possible which I can still support?'.
Another thing might be that the AI is either not focusing enough on its 'home economy' or that there just isn't enough economic depth to deal with in the first place. That leads to the question 'Every thing is in order at home, what do I do now ?' and nations tend to go for conquest in this game.
But all that is leading away from the question of this thread
Dancing Hoskuld
Deity
Looks like I was wrong about the collateral damage from non-siege units causing CTDs - Sorry
I agree with you. But that is not a problem with the SoD, but rather with military focus in this game altogether.
Maybe what leads to war so often in this game is the underlying simplistic system that still, more cities automatically mean more research, more culture, more influence, more parallel production and (when caring a little about timing and placement) usually more money. Sure, city-spam has been penalized in CivIV compared to CivIII, but the system still works that way. And the question now is not only 'How do I get as many cities as possible?' but more like 'How do I get as many cities as possible which I can still support?'.
Another thing might be that the AI is either not focusing enough on its 'home economy' or that there just isn't enough economic depth to deal with in the first place. That leads to the question 'Every thing is in order at home, what do I do now ?' and nations tend to go for conquest in this game.
But all that is leading away from the question of this thread
That's why a revolutions mechanism is essential to any present or future Civ game, in my opinion. When looking at the real world, the obvious reason for why big empires don't just become exponentially bigger is because they fracture from internal instability (just look at all of the uprisings going on right now!) When playing RevDCM, this sort of thing is happening all the time. It's great fun, and I consider it a huge step backward that Civ5 did not incorporate any of the ideas of RevDCM.
And by the way, thanks for the info about helping me diagnose my random CTDs. It's so maddening to have an almost-perfect mod of my own that I can't be sure of being able to finish a game on because of CTDs. Although if I put "New random seed on reload" on, sometimes I can avoid the CTDs when playing from the autosave a second time. Very strange...and I have no idea how to diagnose the problem. The autolog.txt is useless...at least when I'm reading it.
To diagnose CTDs in Civ4, you have to edit the ini to enable minidumps. Then you'll need the microsoft crash viewer and open up that mindump file in your temp files folder and, in a specific place, it'll tell you the function name that caused the crash (in the DLL). I haven't done this in a long while, so I'm a bit rusty.That's why a revolutions mechanism is essential to any present or future Civ game, in my opinion. When looking at the real world, the obvious reason for why big empires don't just become exponentially bigger is because they fracture from internal instability (just look at all of the uprisings going on right now!) When playing RevDCM, this sort of thing is happening all the time. It's great fun, and I consider it a huge step backward that Civ5 did not incorporate any of the ideas of RevDCM.
And by the way, thanks for the info about helping me diagnose my random CTDs. It's so maddening to have an almost-perfect mod of my own that I can't be sure of being able to finish a game on because of CTDs. Although if I put "New random seed on reload" on, sometimes I can avoid the CTDs when playing from the autosave a second time. Very strange...and I have no idea how to diagnose the problem. The autolog.txt is useless...at least when I'm reading it.
I'm fine with SoD, I don't get why they get so much hate. I'm not a big fan of them, though.
My only problem with it is, let's say, in the situation when two big stacks with soldiers and siege meet, the winner will always be the one who attacks first, all thanks to suicide siege.
I've been thinking about it but I can't find a better battle system for Civ IV. I've heard a lot the idea of limited stacks, but honestly I think that will be worse: too much micro dependant and with both the SoD and CoD drawbacks at the same time. It would be something like a "Carpet of Mini stacks-of-Doom of Doom".
I also considered eliminating Collateral Damage at all, so siege would only be used to bring down walls, but with that taking cities would be next to impossible (though maybe a little tweaking on defense could help). But also in that case the winner will be the one who has a bigger stack and/or more advanced units.
Limiting the number of units per Civ doesn't sounds good to me either: If the cap is set with civ size, the big empires would be unstoppable. On the other hand, making the cap to be the same to every civ independant of it size it's not very realistic but can work, I mean sure, it doesn't make much sense that a huge civ can have the same army than a tiny civ, but the big civ still has a production advantage and, maybe, a tech one. I don't know how that would work, though.
Anyway, maybe setting the cap based on civ size can work after all, only if the difference is not that big, and also we can implement a defense system to help small civs. I don't know, maybe using citizens to defense, pretty much like "defense specialists". Set one citizen to be a Defender and a unit will appear to defend the city. That can be interesting, as it can save a small civ for being rushed but at the cost of stop working tiles (like specialists) and also having the risk of losing a defender and thus losing population. Also, If a city has too much defenders the city will shrink due to low food and with low food it can't run that much defenders at the same time. With that said, the attacker can pillage a source of food to archieve this effect faster.
Ahh, but I don't know how that would work in practice. It's just and idea, that't all.
Still I don't have that much problems with SoD: defense is pretty simple if you have a stack with siege and attack first thanks to your roads, so I don't think SoD are that unfair, at least not uptil factories arrive and armies get insanely huge. I think the real problem is that the AI sucks at managing them: many times I captured cities with hole armies in it, siege included, when the AI could just have used all that siege against me and cripple my forces and preventing my attack.
My only problem with it is, let's say, in the situation when two big stacks with soldiers and siege meet, the winner will always be the one who attacks first, all thanks to suicide siege.
I've been thinking about it but I can't find a better battle system for Civ IV. I've heard a lot the idea of limited stacks, but honestly I think that will be worse: too much micro dependant and with both the SoD and CoD drawbacks at the same time. It would be something like a "Carpet of Mini stacks-of-Doom of Doom".
I also considered eliminating Collateral Damage at all, so siege would only be used to bring down walls, but with that taking cities would be next to impossible (though maybe a little tweaking on defense could help). But also in that case the winner will be the one who has a bigger stack and/or more advanced units.
Limiting the number of units per Civ doesn't sounds good to me either: If the cap is set with civ size, the big empires would be unstoppable. On the other hand, making the cap to be the same to every civ independant of it size it's not very realistic but can work, I mean sure, it doesn't make much sense that a huge civ can have the same army than a tiny civ, but the big civ still has a production advantage and, maybe, a tech one. I don't know how that would work, though.
Anyway, maybe setting the cap based on civ size can work after all, only if the difference is not that big, and also we can implement a defense system to help small civs. I don't know, maybe using citizens to defense, pretty much like "defense specialists". Set one citizen to be a Defender and a unit will appear to defend the city. That can be interesting, as it can save a small civ for being rushed but at the cost of stop working tiles (like specialists) and also having the risk of losing a defender and thus losing population. Also, If a city has too much defenders the city will shrink due to low food and with low food it can't run that much defenders at the same time. With that said, the attacker can pillage a source of food to archieve this effect faster.
Ahh, but I don't know how that would work in practice. It's just and idea, that't all.
Still I don't have that much problems with SoD: defense is pretty simple if you have a stack with siege and attack first thanks to your roads, so I don't think SoD are that unfair, at least not uptil factories arrive and armies get insanely huge. I think the real problem is that the AI sucks at managing them: many times I captured cities with hole armies in it, siege included, when the AI could just have used all that siege against me and cripple my forces and preventing my attack.
I've already addressed on how to limit the number of units in a reasonable manner, simply increasing unit maintenance should prevent an infinite sized army. For a sufficiently large army, chaff stacks may be able to limit the impact of siege: having 3 units in a single tile that gets sieged would mean half of the siege's potential is wasted, and the attacker still has to dedicate at least 3 units to clearing the tile, so that's 4 units that have consumed their action per tile, assuming the attacker and get a good angle on the limited stack (having a stack of 3 means spearmen to counter horsemen, horseman to counter archers, archers to counter infantry), so they may even need 2 catapult to attack without other losses.
It's just that, for all intents and purposes, if your SoD is bigger and you attack first, you don't need to concern yourself with more advanced tactics such as using chaff.
It's just that, for all intents and purposes, if your SoD is bigger and you attack first, you don't need to concern yourself with more advanced tactics such as using chaff.
If you nerf siege enough, then SoDs cease being problematic. In my Civ4 xml modifications, I've given catapults and trebs a max damage cap of 20%, cannons 40%, and artillery 50%, while nerfing their strength (to compensate a bit for their increased tendency to withdraw from combat) and increasing the number of units they do collateral damage to. This nerfs the advantage of the attacker so that it remains only situationally true. Now I only bring along ~20-30% of my army as siege, and if I have a good mix of counter units prepared for meeting what the opponent has, then my units can survive defending while at 80% or 60% or 50% health. Defensive terrain helps as well.
Also, if I start splitting my SoD up into balanced mini-stack army groups towards the time the opponent gets artillery, and I attack along a 10 tile front with 60 units (~6 in each stack), then the opponent doesn't have enough siege to take all of my units down to 50% health. Sure, he can focus on two or three of the stacks and wipe those out, but meanwhile I'm still coming with 7 mini-stacks, and his artillery has to heal. Before long, I'm at the gates of his frontline city.
Of course, this assumes that I'm attacking an opponent with inferior force. If I'm attacking an opponent with superior force, then forget about it. But there are ways, mind you, even on immortal in the age of railroads, to deprive the AI of the forces to use against you. One good way is to get the AI to attacking a third party and then stab that AI in the back as their SoD is stuck trudging through foreign territory.
Another thing that could help balance SoD vs. non-SoD play within the Civ4 mechanics would be to implement a land blockading system. (The Planetfall mod already has this, so it is indeed quite possible). You could give 2-move units the ability to blockade the 8 tiles around themselves (this would represent the ability to sally forth and back during a turn), thus making those tiles unworkable and those roads unusable, both for trade route purposes AND for the enemy's use of units.
In such a case, an attacker would still have reduced movement going through enemy territory, but, assuming the attacker had some cavalry or tank units on hand, the defender would not be able to use the roads around the attacker to get the first strike in, unless using all two-move units (and no siege). This could lead to the development of fronts and some positional warfare. Together with the implementation of militarily-claimable territory (already present in some mods), this could totally wipe out the supremacy of SoDs in Civ4, all without radically changing the gameplay over to 1upt.
Also, if I start splitting my SoD up into balanced mini-stack army groups towards the time the opponent gets artillery, and I attack along a 10 tile front with 60 units (~6 in each stack), then the opponent doesn't have enough siege to take all of my units down to 50% health. Sure, he can focus on two or three of the stacks and wipe those out, but meanwhile I'm still coming with 7 mini-stacks, and his artillery has to heal. Before long, I'm at the gates of his frontline city.
Of course, this assumes that I'm attacking an opponent with inferior force. If I'm attacking an opponent with superior force, then forget about it. But there are ways, mind you, even on immortal in the age of railroads, to deprive the AI of the forces to use against you. One good way is to get the AI to attacking a third party and then stab that AI in the back as their SoD is stuck trudging through foreign territory.
Another thing that could help balance SoD vs. non-SoD play within the Civ4 mechanics would be to implement a land blockading system. (The Planetfall mod already has this, so it is indeed quite possible). You could give 2-move units the ability to blockade the 8 tiles around themselves (this would represent the ability to sally forth and back during a turn), thus making those tiles unworkable and those roads unusable, both for trade route purposes AND for the enemy's use of units.
In such a case, an attacker would still have reduced movement going through enemy territory, but, assuming the attacker had some cavalry or tank units on hand, the defender would not be able to use the roads around the attacker to get the first strike in, unless using all two-move units (and no siege). This could lead to the development of fronts and some positional warfare. Together with the implementation of militarily-claimable territory (already present in some mods), this could totally wipe out the supremacy of SoDs in Civ4, all without radically changing the gameplay over to 1upt.
digitCruncher
Emperor
- Joined
- Oct 28, 2007
- Messages
- 1,012
My problem with SoDs is very similar to AFS: It isn't the huge stacks that are a problem, it is the problem with attacker initiative in Civ IV (and Defender Inititive in Civ III, where the defender would get so many bonuses, the only way to dislodge someone was to bombard. From memory, I haven't played Civ III in donkeys years)
Civ III combat shows the problems with having *no* collateral damage, and Civ II and Civ I show the problem with *all* having collateral damage (although, to be fair, in Civ II and Civ I, if you lost one defender, EVERYONE was destroyed, unless you were in a fort or a city, in which case Civ III rules were applied)
However, Civ V did one thing right: It was the first Civ installment to attempt to solve the problem in a revolutionary way. However, the implementation was difficult to implement AI support, which is, in my opinion, the only problem with Civ V. The big waves across the border is actually more historically accurate (I think) than narrow, huge armies marching across the land, where supplies can be cut off easily. Although, as Civ V showed, AI support seems too difficult to implement.
One option is to blatantly rip off the Total War series. In that version, you have two stacks that battle on a smaller map.
Now, there is a SIGNIFICANT problem with that: It takes about 20 minutes at MINIMUM to resolve small-scale clashes, and it results in one 'turn' taking up to an HOUR.
So, one solution is to implement each stack as an 'army'. Each battle wouldn't be one unit fighting the best defender until one side wins, instead all the units would attack at the same time (and defenders would counter-attack). The obvious problems would be displaying this battle in a meaningful way, so that if the player loses, then they can see WHY they lost (EG. OF COURSE! I didn't have enough cavalry to secure my flanks, which lead to my predominantly siege army to be destroyed by their flanking cavalry.), and having accurate odds (Obviously, the easiest way is to run the simulation 1000 times, and return the results of the 'test' runs).
Oh, and the actual calculations and AI implementation need to be balanced and effective. It does have a rock-paper-scissors effect (where cavalry beat ranged/siege units, which are beaten by infantry, which are beaten by archers. And you might be able to differentiate light and heavy cavalry and infantry too...), which is always good for balance, and can lead (with creativity) to fun game mechanics.
Just a reminder: Having battles resolve on a tactical map is arbitrarily slow and not particularly needed. It also makes it VERY hard to balance properly. But having armies slam into each other as one unit, rather than 62 waves of attackers, is probably an improvement. Of course, it is a SIGNIFICANT change, and is more likely to be implemented as a Civ VI feature, rather than a Civ IV mod.
Civ III combat shows the problems with having *no* collateral damage, and Civ II and Civ I show the problem with *all* having collateral damage (although, to be fair, in Civ II and Civ I, if you lost one defender, EVERYONE was destroyed, unless you were in a fort or a city, in which case Civ III rules were applied)
However, Civ V did one thing right: It was the first Civ installment to attempt to solve the problem in a revolutionary way. However, the implementation was difficult to implement AI support, which is, in my opinion, the only problem with Civ V. The big waves across the border is actually more historically accurate (I think) than narrow, huge armies marching across the land, where supplies can be cut off easily. Although, as Civ V showed, AI support seems too difficult to implement.
One option is to blatantly rip off the Total War series. In that version, you have two stacks that battle on a smaller map.
Now, there is a SIGNIFICANT problem with that: It takes about 20 minutes at MINIMUM to resolve small-scale clashes, and it results in one 'turn' taking up to an HOUR.
So, one solution is to implement each stack as an 'army'. Each battle wouldn't be one unit fighting the best defender until one side wins, instead all the units would attack at the same time (and defenders would counter-attack). The obvious problems would be displaying this battle in a meaningful way, so that if the player loses, then they can see WHY they lost (EG. OF COURSE! I didn't have enough cavalry to secure my flanks, which lead to my predominantly siege army to be destroyed by their flanking cavalry.), and having accurate odds (Obviously, the easiest way is to run the simulation 1000 times, and return the results of the 'test' runs).
Oh, and the actual calculations and AI implementation need to be balanced and effective. It does have a rock-paper-scissors effect (where cavalry beat ranged/siege units, which are beaten by infantry, which are beaten by archers. And you might be able to differentiate light and heavy cavalry and infantry too...), which is always good for balance, and can lead (with creativity) to fun game mechanics.
Just a reminder: Having battles resolve on a tactical map is arbitrarily slow and not particularly needed. It also makes it VERY hard to balance properly. But having armies slam into each other as one unit, rather than 62 waves of attackers, is probably an improvement. Of course, it is a SIGNIFICANT change, and is more likely to be implemented as a Civ VI feature, rather than a Civ IV mod.
Thunderbrd
C2C War Dog
I've never had a problem with SoDs... in fact, the change in Civ V seemed so loathsome to me that I don't really care to get into the new version. I don't see why its considered problematic at all really. If you properly stack your own defenses correctly you shouldn't have trouble with them. And overcoming them when on the offense is simply a matter of attacking in the proper order. I do like some of the forward thinking I'm seeing here though... particularly the land blockade. And I utilize the surround and destroy option in my games because I think it does work nicely to help to balance the system (it works very well with commandos!).
In the late game, Weapons of Mass Destruction does create a definite deterent to a full on SoD strategy... such stacks are easily wiped off the map with a couple of good nuke strikes. And bio-missiles help in that regard as well.
I do see a problem with the troop to population ratio but that's in part a flaw in the food creates population and production creates troops methodology employed in civ IV to which I have proposed some complex and difficult to program solutions.
In the late game, Weapons of Mass Destruction does create a definite deterent to a full on SoD strategy... such stacks are easily wiped off the map with a couple of good nuke strikes. And bio-missiles help in that regard as well.
I do see a problem with the troop to population ratio but that's in part a flaw in the food creates population and production creates troops methodology employed in civ IV to which I have proposed some complex and difficult to program solutions.
General Tso
Panzer General
I've always felt the opposite. I only ever finished 2 or 3 games in Civ4 (although I played it a lot). Once there are to many units on the map the game becomes a combination of being to complex and boring at the same time. What's neat about the Civ series is that different players can get some enjoyment out of the same game while using totally different playing styles.
In my mod (which has been stuck in alpha mod for quite a while) I used a slightly different idea to stop SOD's. I added a UnitSize field to the terrain xml file. This value controls how many units can be located on each plot, depending on it's terrain type. For example in my mod grasslands can hold two units while desert can only hold 1. I then added a UnitSizeModifier to the improvements, features and routes xml files so that things like forts, oasis, and roads can alter the number of units allowed. By tweaking these values anybody can get something that works for them.
In my mod (which has been stuck in alpha mod for quite a while) I used a slightly different idea to stop SOD's. I added a UnitSize field to the terrain xml file. This value controls how many units can be located on each plot, depending on it's terrain type. For example in my mod grasslands can hold two units while desert can only hold 1. I then added a UnitSizeModifier to the improvements, features and routes xml files so that things like forts, oasis, and roads can alter the number of units allowed. By tweaking these values anybody can get something that works for them.
duckstab
Child of Noble Family
I've always thought that the problem is not so much the SoD as the number of units overall. 1UPT doesn't help with that in the least. I'd really like to see a system where the human player could be competitive in the late game without having dozens or even hundreds of units. That's what tends to make the late game in Civ4 tedious at best; even if you have most of your units in big stacks you're going to have to deal with new units that you've built, and you can't usually use your units effectively unless you're willing to deal with them in smaller groups or individually.
Similar threads
