Questions

[SerWind]

I'll say briefly I am very very excited for the Attila version of this. Just curious though if there's any plans to have the Senate matter at all? I'm guessing the divide mechanic to choose between Empire/Republic is impossible or no?

[Petellius]

We're working on this. We'll give more details once we get to work on it and flesh it out a bit more but we would like to have the senate and transition to empire be a huge thing, just might not happen on release.

[SerWind]

Apologies for the lack of thanks for the reply. It's good to hear, I have to say I was pretty miffed at how lackluster the senate is in Rome 2. If you don't mind me asking, historical accuracy is one of the goals of the mod, I'm assuming this means we won't be seeing Segmentata?

[Causeless]

Apologies for the lack of thanks for the reply. It's good to hear, I have to say I was pretty miffed at how lackluster the senate is in Rome 2. If you don't mind me asking, historical accuracy is one of the goals of the mod, I'm assuming this means we won't be seeing Segmentata?

Segmentata was historically used, just not during the republic.

[Willhelm123]

Apologies for the lack of thanks for the reply. It's good to hear, I have to say I was pretty miffed at how lackluster the senate is in Rome 2. If you don't mind me asking, historical accuracy is one of the goals of the mod, I'm assuming this means we won't be seeing Segmentata?

We're going into Imperial times so you will see it. We intent to portray almost every legion so some Imperial legions will mostly wear it while some wont.

[Petellius]

Yeah the mod is from 202BC to 100AD and will eventually include events and historical units until that period. This means 100 years (400 turns) of heavily researched and scripted gameplay, including events like the Cimbri invasion and the Social Wars all individual legions and Auxiliaries, diverse systems between republic and empire. At least thats what we want to do . You can obviously play past Trajan's reign (100AD) but there wont be any mod detail after that (so no later units or major events will be seen).

[Alannon]

I also have a question - will this mod ever have the variety units have akin to FOTE? I mean, in FOTE units don't have "faction colours" and there are like, dozens of shields and often armours/clothing available in a single unit. Or this isn't really the goal here?

[Willhelm123]

I also have a question - will this mod ever have the variety units have akin to FOTE? I mean, in FOTE units don't have "faction colours" and there are like, dozens of shields and often armours/clothing available in a single unit. Or this isn't really the goal here?

Factions tend to have a theme colour but we do it quite subtly, and there are a lot of varied and common colours within every faction. For example the Iceni are a blue theme but that doesn't mean every unit is bright blue, its a greyish blue and not on every unit, it just contributes to the overall look of the unit as a whole.

[Petellius]

Well we also don't really have faction colour and there is a lot of varied units and even within a unit they are quite varied. I have never played FOTE though so I couldn't tell you, but I do know we tend to have less units than most overhauls.

[Tempest51]

Are their any options for political reform when barbarian factions progress beyond a certain point? It would be strange for the Gauls or the Suebi to maintain a tribal confederation after they take over all of Europe, maybe give them the option of creating a pseudo-feudal state?

[Petellius]

If I remember rightly the only way for barbs to grow outside of gaul and germany is to use existing civilisations and building structures. Also actually making a confederacy will be pretty much the only way you can start taking down big factions like Rome. You have to fight your way to the top of the tribal ladder and then convince them all to attack with you, lots of inter faction politics and client tribes.

[Tempest51]

If I remember rightly the only way for barbs to grow outside of gaul and germany is to use existing civilisations and building structures. Also actually making a confederacy will be pretty much the only way you can start taking down big factions like Rome. You have to fight your way to the top of the tribal ladder and then convince them all to attack with you, lots of inter faction politics and client tribes.

Yes this much I understand. I'm talking about when the faction has expanded its borders beyond what a tribal confederation can feasibly maintain, perhaps it would make sense for them to transition into a more concrete political structure? Maybe like a proto-kingdom or a Romanesque empire? Think of it as the migration era kingdoms appearing 5 centuries early.

[ABH2]

That is already incorporated into our barbarian building chains. At some point we'll preview those, I think, though we aren't there yet. The Germans in particular have later tiers modeled off the later.

To kind of outline this roughly for people - main settlement lines will now represent political association with a region. Or rather how you are consolidating a piece of territory into your empire officially. In general, the player will get to select from options specific to the subculture/faction. But they tend to breakdown roughly as formal consolidation and direct rule, colonies (sometimes garrisons), autonomous/free regions left mostly to their own, and sometimes tributary status.

For some of the barb factions the colony line allows you options to plant your followers or your population into a region. This can be in some cases another chieftain as a vassal, or lower level followers. So the player will have some choice there.

Hopefully in game we can distinguish all of this neatly and give the player a number of options better suited to different circumstances.

I also have a question - will this mod ever have the variety units have akin to FOTE? I mean, in FOTE units don't have "faction colours" and there are like, dozens of shields and often armours/clothing available in a single unit. Or this isn't really the goal here?

I'd add to what William said. Most of our uniform colors are set by the unit. I think we have a massive amount of visual diversity in our mod thanks to help from a lot of others who have allowed us to use their work. If I list the mods/modders, I fear I'll forget someone important. But DeI, Little Legionaire, Miri82's work, Kaziel's, KLA's, Drambelli, Aguire, TWH etc.

The amount of variety depends on the unit, but I feel like we've accomplished a great deal of depth. More will come with later releases. I'm particularly proud of what we've been able to do with our Hellenistic factions, Arabs, Iberians, Africans etc. For many cultures, we have almost entirely unique rosters matched to the faction already.

At some point, I think there are certain groups where we may want to do unique color schemes for units based on faction. These would be for more generic rosters shared across many factions. I like the idea of some units being objectively colored, but think some should be tribal. We also have some further expansions planned for other groups for after we can get a first release out.

So, I think people will be very pleased with the visual diversity in this mod. I'm not really a zoom in and watch type player during battles, but even I've noticed it is pretty impressive and cool to look at.

It's really for others to judge all of this when we can release, however.

[ReaperXM]

Actually, thinking of that, will the AI be any smarter? I know there can't be a whole lot you can do to it since a lot of the ai is hardcoded, but will you just be giving the ai a lot of screwy buffs, or will they actually be smart?

[ReaperXM]

Client states in Attila offer a big advantage for small nations because they give you a much larger (if stupid) military than in you simply conquered that nation. I remember you saying that you hate the client state system and are going to create a system that lets you create "client states" but still have conquered the land. Will you still get that military boost?

[Petellius]

The client states we speak of is a building chain after you conquer the region, though the diplomatic option still exists. On the AI, the BAI is already somewhat improve if only temperamentally, and we will work on the CAI. Look at the aggressive CAI mod on the workshop, we'll likely try to accomplish something like that but toned down and more suited to AE's more relaxed slowed campaign speed.

[ABH2]

Client states in Attila offer a big advantage for small nations because they give you a much larger (if stupid) military than in you simply conquered that nation. I remember you saying that you hate the client state system and are going to create a system that lets you create "client states" but still have conquered the land. Will you still get that military boost?

Depending on the time of 'treaty' you set for the conquered region, you will have different units recruitable. So if you aren't setting up your own population, you will recruit from the native populations with AOR units. You won't have to spend time building up a population of, say, Romans if you want AOR. We may not have any recruitment slots in this mod like in vanilla since we have three other tools we can use to simulate that:
1. Population - you can only recruit so many soldiers
2. Unit cost
3. Perhaps turns to recruit the unit. So, 2 for some instead of 1 - where appropriate.

Actually, thinking of that, will the AI be any smarter? I know there can't be a whole lot you can do to it since a lot of the ai is hardcoded, but will you just be giving the ai a lot of screwy buffs, or will they actually be smart?

Smart is a strong word with AI. I don't want to do what I did in the Rome II submod where I gave the AI buffs. Attila has a better CAI system, but I haven't gotten a chance to mod that yet. The AI will have certain advantages that all mods give it, but which won't really be noticeable to the player (more food and public order as the two obvious ones).

In terms of things already experimented with that *I* believe improve AI decision making, I do want higher movement speeds across the board for armies. This is controversial as some people like smaller movement speeds for their armies, and I will design this so it is very easy to create submods. I'll probably release a few myself. But the CAI is noticeably more aggressive and responsive when it can move farther distances. this isn't a buff as the player will have the same movement. The other aspect to this is its really realistic. Ancient armies could move a lot farther in 4 months than is indicated in game. Even a barbarian horde. Some factions may have advantages in terms of army movement along with certain character effects and such, but that's not unique to our mod.

My belief is that the CAI does best when it has to do less planning. If it's free to respond to the player in one turn, it's going to do better than if it had to plan three turns ahead. Moving one army to attack you in one turn rather than three.

[AugustusRoman]

Depending on the time of 'treaty' you set for the conquered region, you will have different units recruitable. So if you aren't setting up your own population, you will recruit from the native populations with AOR units. You won't have to spend time building up a population of, say, Romans if you want AOR. We may not have any recruitment slots in this mod like in vanilla since we have three other tools we can use to simulate that:
1. Population - you can only recruit so many soldiers
2. Unit cost
3. Perhaps turns to recruit the unit. So, 2 for some instead of 1 - where appropriate.



Smart is a strong word with AI. I don't want to do what I did in the Rome II submod where I gave the AI buffs. Attila has a better CAI system, but I haven't gotten a chance to mod that yet. The AI will have certain advantages that all mods give it, but which won't really be noticeable to the player (more food and public order as the two obvious ones).

In terms of things already experimented with that *I* believe improve AI decision making, I do want higher movement speeds across the board for armies. This is controversial as some people like smaller movement speeds for their armies, and I will design this so it is very easy to create submods. I'll probably release a few myself. But the CAI is noticeably more aggressive and responsive when it can move farther distances. this isn't a buff as the player will have the same movement. The other aspect to this is its really realistic. Ancient armies could move a lot farther in 4 months than is indicated in game. Even a barbarian horde. Some factions may have advantages in terms of army movement along with certain character effects and such, but that's not unique to our mod.

My belief is that the CAI does best when it has to do less planning. If it's free to respond to the player in one turn, it's going to do better than if it had to plan three turns ahead. Moving one army to attack you in one turn rather than three.

Best post I have ever read!!

[Causeless]

Smart is a strong word with AI. I don't want to do what I did in the Rome II submod where I gave the AI buffs. Attila has a better CAI system, but I haven't gotten a chance to mod that yet. The AI will have certain advantages that all mods give it, but which won't really be noticeable to the player (more food and public order as the two obvious ones).

In terms of things already experimented with that *I* believe improve AI decision making, I do want higher movement speeds across the board for armies. This is controversial as some people like smaller movement speeds for their armies, and I will design this so it is very easy to create submods. I'll probably release a few myself. But the CAI is noticeably more aggressive and responsive when it can move farther distances. this isn't a buff as the player will have the same movement. The other aspect to this is its really realistic. Ancient armies could move a lot farther in 4 months than is indicated in game. Even a barbarian horde. Some factions may have advantages in terms of army movement along with certain character effects and such, but that's not unique to our mod.

My belief is that the CAI does best when it has to do less planning. If it's free to respond to the player in one turn, it's going to do better than if it had to plan three turns ahead. Moving one army to attack you in one turn rather than three.

I'd just like to add a little onto this, to add onto the anecdotal evidence presented.

Unlike most strategy games which use decision trees or state machines (as I believe the earlier Total War games did), Rome 2 uses something a lot closer to some of the techniques used in chess AI. It's not quite on the same magnitude of complexity - the game design of Rome 2 is insanely more complex than chess - but your observations upon the aggressiveness and intelligence of the AI when given less options is very much true and completely valid. It's the same with chess AI, where they tend to play better near the end of the game due to more limited options.

If you consider tic-tac-toe, which has a tiny and very easily brute-forceable decision space, it's easily possible to make an AI which can never lose (only ever draw or win). I've written a recursive algorithm to do this myself, as current computers have more than enough speed and memory to hold the entire decision space of tic-tac-toe. The same concept, albeit with insanely larger decision space, applies to AI in chess, Rome 2 and presumably also Attila. So, theoretically speaking, reducing the decision space of the game for the AI will make it more likely to find a good strategy. This is because it is more likely quickly find a high-scoring strategy - especially when it only needs to search a couple of turns deep instead of the larger decision tree depths that shorter movement ranges would entail.

The same should apply to other AI decisions too, such as number of recruitable units, number of military forces like armies and navies, how many building choices the AI can make and things like recruitment/building times etc. This is almost definitely the reason why large factions take more time on the end-turn screen, as the amount of possible decisions they can make is far greater than that of a city-state. However, for obvious gameplay reasons, aggressively reducing the decision count isn't always the best idea. I'd also like to note that this almost definitely would not work the same way in the real-time battles, as undoubtedly they'll use a different AI system.

I'm of the personal opinion that increasing army/navy ranges is a good idea, both due to the AI buffs and the increased realism in such an approach. The one caveat I'd note that sometimes the AI cheats around zone-of-control (or at least, in Rome 2, but this could've been fixed) which could allow armies to just run through another's zone of control. This was a pretty big issue in R2TR if I remember correctly, although may not apply here.

I just wanted to reinforce the anecdotal experiences you've had with stronger evidence, and validate your observations. If you haven't seen it, I'd suggest reading this very interesting article which goes into the AI system the campaign uses.

[Petellius]

You should try the DeI/AE submod for Rome 2, it shows off just how aggressive the CAI can be using the faster marching speeds.