Total War thread

[Big Brown Schlub]

Its been that way for a while. CA can't help but make OP ranged units. You ever see what some archers can do in Three Kingdoms?

In Total War it's either ranged units that are king or cav that is king, often times it's both. 3K is a good example, a lot of archer/crossbow focused factions do ridiculous damage, like for example Kong Rong's unique crossbow units, though generic archers that everyone has access to can be pretty ridiculous too. For example Azure Dragons, they're both a really good polearm unit and a really good archer. Cav is also ridiculous in 3K, even the lightest cav has ridiculous mass, they make cav in Attila seem weak and cav in Attila was also ridiculous, my horse archer units would get more kills charging into the back of someone than shooting them, even if they were underfed steppe boiz armed with a knife and armored with a glorified burlap sack that was charging into Herculi Seniores.

Ranged units were weak as a rule in Attila except if they were crossbows or if they were Slavs with bows, and in the latter's case it's not that they had powerful arrows but because the poison mechanic was fucking ridiculous. They already have more killing power than a regular archer but they also inflict 100% exhaustion permanently for the rest of that battle makes them unbelievably strong and all it takes is for a single arrow to merely touch someone in a formation.

 

[Scale Smerch]

I know it'll never happen but I have dreams of CA giving the Chorfs some further love. My boy Zhatan needs new tools to crush Cathay skulls and make Vilitch's life further hell.

But in all seriousness I do like the vamp changes and Neferata is fun.

Neferata is very fun. Her starting location also gives some reason for multiple campaigns because there are several natural expansion routes.

 

[OutInTheRain]

Neferata is very fun. Her starting location also gives some reason for multiple campaigns because there are several natural expansion routes.

I also find Thanquol changes expansion plans. Drawing Empire campaigns south and into Gorbad rather than north and into Chaos.

 

[Ayynon69]

Things like halberd variants make sense, but the heavy armor hybrid units that you recruit turn 2 and use for the entire campaign are boring. I was also thinking how useless magic buffs are, almost universally worse than damage spells. They really need to last longer to be viable most of the time.

I suppose the real problem is retards struggle playing factions with a difficult start. These mouth-droolers leave negative reviews or seethe on the tw forums about the balance being terrible until CA "fixes" it. Essentially catering to DSP, but the tw faggot version.

They literally fucking design games around DSP now and he still sucks at them.

For anyone not familiar with DSP_Gaming, he is a famous total war streamer and PC builder. Lookup "DSP incident" to learn more.

 

[I Know 孤独のせい]

I was also thinking how useless magic buffs are, almost universally worse than damage spells. They really need to last longer to be viable most of the time.

For the most part you are right unfortunately but it can depend on the school of magic, some can be okay but there often some weird combinations that don't make much sense. No idea why they can't buff them across the board.

I suppose the real problem is retards struggle playing factions with a difficult start. These mouth-droolers leave negative reviews or seethe on the tw forums about the balance being terrible until CA "fixes" it. Essentially catering to DSP, but the tw faggot version.

Even if the start can be difficult that's nothing when looking at an entire campaign. The entire difficulty curve is completely fucked so no idea why people are complaining about that particular thing.

he is a famous total war streamer

DSP played TW ?!?!?! No way, he wouldn't even survive the first battle against the rebel army.

 

[Big Brown Schlub]

Even if the start can be difficult that's nothing when looking at an entire campaign. The entire difficulty curve is completely fucked so no idea why people are complaining about that particular thing.

Seriously, they are completely unable of making campaigns that stay challenging throughout, or at least have it not be some snowballfest. If campaigns are difficult, they're always in the beginning. In Rome and Medieval 2 if you started on one ass end of the map you'd at least know you'd have at least one huge faction to fight against when you got to the other ass end of the map. Hell, in Rome if you play as one of the Romans and don't trigger the civil war early and let it develop naturally you always end up with at least 2 big factions facing you. And it's not like Rome and Medieval 2 are those 2tuff4u megadifficult games, they're probably among the easier ones in the series. The only way they're able to prevent snowballs from happening is if they put in artificial stopgaps like Realm Divide in Shogun 2 or having public order problems if you as much as sneeze too hard like in Attila.

 

[I Know 孤独のせい]

Seriously, they are completely unable of making campaigns that stay challenging throughout, or at least have it not be some snowballfest. If campaigns are difficult, they're always in the beginning.

I have to do the unimaginable and run defense for CA for just a moment. From my personal experience barely any developer and/or game is or stays difficult/challenging throughout its entire runtime. The difficulty curve always ramps up in the begging until it evens out quickly and often drops down for the rest of any game. Which often comes down to developers not understand what difficulty is and equating it with the usual busywork ( stat imbalances, spongey enemies etc). And of course the bigges culprit is AI which has seen no tangible improvements in decades and is often incapable of playing the game at all thus needing to cheat to impede the rapid progess of the player.

CA don't bother trying to improve the situation because why should they ? Im sure all metrics that developers get nowadys point to people playing games on the easier difficulty settings and often even dont finish the game or in the case of TW ( and something like Civ) only playing for a couple of turns until starting a new campaign.

AI cheats it is than to create the artificial difficulty. I bet if the AI would actually understand and know how to play both the strategic and the tactical battles we would have this conversations. But alas...

In Rome and Medieval 2 if you started on one ass end of the map you'd at least know you'd have at least one huge faction to fight against when you got to the other ass end of the map. Hell, in Rome if you play as one of the Romans and don't trigger the civil war early and let it develop naturally you always end up with at least 2 big factions facing you.

Even though it feels more natural and less on the nose it is still an artificial system even if it is a bit more elegant than Realm Divided.

The only way they're able to prevent snowballs from happening is if they put in artificial stopgaps like Realm Divide in Shogun 2 or having public order problems if you as much as sneeze too hard like in Attila.

Speaking of, Realm Divided sucks in its current from, they might have been able to tweak it but never bothered. But apart from the aforementioned big issues (AI) I think CA has been really inconsistent with this kind of stopgap mechanic because most players just want to paint the map for a couple of hours.

And since I am an ardent Attila defender, which really pissed off the whiny baby TW players, I have to say that its systems were great on paper and some even actually good in the game. Having to pay attention to public order, something that can be ignored in all other games ( or even exploited) was nice, I think Attila had the most fitting stopgap systems implemented that also felt natural to the game.

I think CA never really figured out of to implement artificial stopgaps and barriers that felt immersive and facilitated people to engage with the game without feeling cheap or forced.

 

[Kane Lives]

And since I am an ardent Attila defender, which really pissed off the whiny baby TW players, I have to say that its systems were great on paper and some even actually good in the game. Having to pay attention to public order, something that can be ignored in all other games ( or even exploited) was nice, I think Attila had the most fitting stopgap systems implemented that also felt natural to the game.

They didn't even need to look that far back. TWW2 had far better implementation of public order and corruption compared to the utterly retarded system they settled with in TWW3. Corruption especially.

 

[I Know 孤独のせい]

They didn't even need to look that far back. TWW2 had far better implementation of public order

Don't know if I would call it better but it actually did something. But you could force and farm the rebellions which CA hated so that why they neutered it. And mentioned that they were looking into it and how to rework it.

and corruption compared to the utterly retarded system they settled with in TWW3. Corruption especially.

Does the AI have to deal with corruption ? Pretty sure the effects are kept to a minimum not don't apply at all like so many other systems. As for WH3 if I recall correctly they wanted to rework it since there are so many forms of corruption now but nothing came out of it. Maybe before 40K comes out someone will remember it.

 

[Kane Lives]

Does the AI have to deal with corruption ? Pretty sure the effects are kept to a minimum not don't apply at all like so many other systems. As for WH3 if I recall correctly they wanted to rework it since there are so many forms of corruption now but nothing came out of it. Maybe before 40K comes out someone will remember it.

They do actually, though they were dumb about it. The major one being that their armies all take attrition (unlike TWW3) so they're pretty much weakened by it, and rebellions due to corruption were also not uncommon (though due to AI bonuses you may need to force it via raiding. Unlike TWW3, however, it's actually possible without investing 3-4 heroes for it.)

 

[Scale Smerch]

Which often comes down to developers not understand what difficulty is and equating it with the usual busywork ( stat imbalances, spongey enemies etc). And of course the bigges culprit is AI which has seen no tangible improvements in decades and is often incapable of playing the game at all thus needing to cheat to impede the rapid progess of the player.

Disagree, it's not so much developers not knowing as much as time interfering with it. In general it's not hard to make an AI which can keep up with the player (see the spell dodging at very hard difficulty), but for a lot of systems it takes time to implement and given current release cycles and expectations it results in bare minimum approaches - i.e. stat boosting, free resources, bullet sponge enemies - since that takes negligible effort to implement. Tack on CA having each faction be more or less unique in terms of mechanics (faction or otherwise) and it makes for a lot of work that overall few people would be willing to pay for.

I don't fault CA for the AI because at least they've thrown some effort into it unlike Firaxis or Paradox, have documented their efforts via dev diaries, and provided a suite of toggles to make for a decent experience. Late game snowballing is another matter entirely since it's still something 4x struggles with across the board.

 

[Snekposter]

Disagree, it's not so much developers not knowing as much as time interfering with it. In general it's not hard to make an AI which can keep up with the player (see the spell dodging at very hard difficulty), but for a lot of systems it takes time to implement and given current release cycles and expectations it results in bare minimum approaches - i.e. stat boosting, free resources, bullet sponge enemies - since that takes negligible effort to implement. Tack on CA having each faction be more or less unique in terms of mechanics (faction or otherwise) and it makes for a lot of work that overall few people would be willing to pay for.

I don't fault CA for the AI because at least they've thrown some effort into it unlike Firaxis or Paradox, have documented their efforts via dev diaries, and provided a suite of toggles to make for a decent experience. Late game snowballing is another matter entirely since it's still something 4x struggles with across the board.

I don't think any 4x or 4x-like game has ever managed to successfully address the snowball problem in a way that isn't just a giant headache for the player. Hell, I hated Realm Divide in S2 because it made guys who were my loyal vassals and would have gotten in on the ground floor of my new rule suddenly hate my guts and decide to take a shot at becoming Shogun themselves.

 

[Big Brown Schlub]

I don't think any 4x or 4x-like game has ever managed to successfully address the snowball problem in a way that isn't just a giant headache for the player. Hell, I hated Realm Divide in S2 because it made guys who were my loyal vassals and would have gotten in on the ground floor of my new rule suddenly hate my guts and decide to take a shot at becoming Shogun themselves.

Realm Divide would've made more sense if it worked like it does in FotS where dudes on the same side headed by the strongest daimyo in the pack join up to take on everyone else (and it'd even be historically accurate!) but obviously it'd be a pain to implement, compared to just pro-Imperial or pro-Shogunate they'd have to make some sort of dynamic way of building coalitions and alliances and advanced diplomatic shit like that like you can in Nobunaga's Ambition or even in Total War's very own Three Kingdoms. I think overall 3K has the best campaign. It's got issues, sure, but the AI actually knows somewhat how to use all mechanics at their disposal plus they get affected by all the shit a player can get affected by. Hell, if the player sucks too hard or if they just let it happen on purpose, the AI can trigger the 3 kingdom divide if one of them reaches enough prestige required to form a kingdom before the player does. The thing that sucked the most about Realm Divide in Shogun 2 is that the AI was not affected by it at all, just like how it doesn't get affected by movement attrition in Attila for example, or how it doesn't get affected by having to pay army upkeep in almost every game since Shogun 2.

 

[I Know 孤独のせい]

Disagree, it's not so much developers not knowing as much as time interfering with it. In general it's not hard to make an AI which can keep up with the player (see the spell dodging at very hard difficulty), but for a lot of systems it takes time to implement and given current release cycles and expectations it results in bare minimum approaches - i.e. stat boosting, free resources, bullet sponge enemies - since that takes negligible effort to implement. Tack on CA having each faction be more or less unique in terms of mechanics (faction or otherwise) and it makes for a lot of work that overall few people would be willing to pay for.

Nobody even made the attempt to program an AI to keep up with the players. And clearly it must be difficult since the only thing we get across almost all games are band aid solutions, mechanics not applying to AI and cheats. Stuff like spell dodging or not wasting ammunition are band aid solutions that follow a clear if/else logic. It's a set behavior that can be circumvented easily for example the aforementioned spell dodging immediately breaks down if you cancel the spell and try again because the AI reacts to the first attempt. Nothing about this is intelligent and that's the issue. The "AI" doesn't understand what is happening and even regular mechanics are too much. Fog of war does not apply to the AI, the snipe trait that some units have doesn't make them invisible but the AI purposely ignores those units. And this shit has been in TW forever. In Shogun 2 the foreign occupation debuff to public order doesn't not apply to the AI. The AI can also convert low tier units into high tier ones.

And yes time and money clearly are the biggest issues why this particular issues keeps persisting but we had the warscape engine for probably 20ish years. That is enough time to actually work on that and make sure that the AI at least understands the bare minimum of whats going on. And we are paying for it, the AI plays the DLC faction with their numerous faction mechanics

I don't expect the AI to be Sun Tzu but the AI can only do three things in a land battle in WH3: Rush me, stand their ground or rush me but send cavalry into my flank. This isnt intelligent or smart or thinking, its if/else. There is no surface level understanding of strategy and tactics or anything to be honest.

It was CA that told its paying costumer base after the disastrous launch of WH3 that its AI is a spaghetti code, I think you are giving them too much credit when its clear that they are over their heads with programming the AI.

I don't fault CA for the AI because at least they've thrown some effort into it unlike Firaxis or Paradox, have documented their efforts via dev diaries, and provided a suite of toggles to make for a decent experience. Late game snowballing is another matter entirely since it's still something 4x struggles with across the board.

I can't really comment on Firaxis or Paradox and the toggles are nice options but still band aids which I have to play around with until I find the setting that suits me. Should be the developers job to fine-tune the difficulty of a game. This mentality of outsourcing balancing to players is something that really grinds my gears and is just plain fucking lazy.

Snowballing will remain an issue until AI understands wtf is going on because every attempt to offload this onto the players always backfires ( Civ 7's soft reset between ages).

 

[Average Russian Spy]

DSP played TW ?!?!?! No way, he wouldn't even survive the first battle against the rebel army.

>DSP in the first battle as Karl Franz. "Okay chat, our first battle, let's go!"
>Battle preparation begins. DSP gets confused for 20 minutes as he tries to figure out how to spin units and make them face certain directions.
>After help from the chat, combat starts. DSP just sends all forces forward without any thought, missile units included.
>Melee troops enter into combat, but missile units immediately flee as enemies approach since DSP never turned off skirmish mode.
>"Uhhh what the fuck??? Why the hell are they running?! Bro!... STOP RUNNING!" DSP repeatedly tries to force missile units to stop. Doesn't work.
>Completely forgets about melee troops as he has been hyper focused on his retreating missile units. Melee troops are routed while Karl Franz stands around doing nothing.
>Army losses kicks in and all of DSP's units insta rout. "What the fuck! HOW DID I LOSE??? Dude, this game is fucking broken. My guys didn't even fight anyone! This game is so fucking stupid, dude..."

 

[Scale Smerch]

Nobody even made the attempt to program an AI to keep up with the players. And clearly it must be difficult since the only thing we get across almost all games are band aid solutions, mechanics not applying to AI and cheats. Stuff like spell dodging or not wasting ammunition are band aid solutions that follow a clear if/else logic. It's a set behavior that can be circumvented easily for example the aforementioned spell dodging immediately breaks down if you cancel the spell and try again because the AI reacts to the first attempt. Nothing about this is intelligent and that's the issue. The "AI" doesn't understand what is happening and even regular mechanics are too much. Fog of war does not apply to the AI, the snipe trait that some units have doesn't make them invisible but the AI purposely ignores those units. And this shit has been in TW forever. In Shogun 2 the foreign occupation debuff to public order doesn't not apply to the AI. The AI can also convert low tier units into high tier ones.

I know where you're coming from, but IMO the anger is still misplaced. To be pedantic current game AI is nothing but conditionals under the hood, and the problem for strategy games is that you quickly rack up enough mechanics and systems that typical AI design approaches (e.g. state machines, GOAP systems, behaviour trees) quickly take significant developer time to tune, balance, and test as a whole - time that you as a dev may not have when the product manager gives you a firm delivery deadline.

Even the RTS side is a relatively solved problem where results are gated by effort: Starcraft I maintains a decent AI (and even has a new flavour in Pluto AI built by a fan), Planetary Annihilation Titans has seen some impressive work over the years, and within my own interests Call to Arms GOH Ostfront has a mod using dev tools that has the AI through movement alone prove a decent threat. None of them are perfect, but they show what's possible for it with dedicated effort, and I guarantee even the AI devs at CA could get close if they had a few uninterrupted months to play catch up and properly account for every new mechanic and interaction permutation.

TL;DR: blame CA management and the company's shareholders for focusing on fluff over mechanical substance; it all comes down to where they're paying their devs to focus their time.

 

[Tsumugi Kotobuki]

It's been a while since the last Skaven DLC but this most recent one has once again reinforced the precept that Skaven are CA and GW's darlings and they're fundamentally incapable of releasing a Skaven DLC that isnt wildly overtuned. No idea what Thanquol is like in campaign but in the vaccum of multiplayer basically every single Skaven unit introduced is so wildly above the powercurve (especially Thanquol, Skreech, and the Stormfiends) that Skaven rocketed up from one of the weakest factions in the game to easily second-strongest behind Undead Legions (which is only strong because its essentially 3 rosters in one).
Thanquol especially is just personally offensive to me: Highest leadership of any Skaven lord on his mount (15 more than Queek lmao), dueling stats to beat Archaon + a powerful ranged attack, free healing for spellcasting that lets him healcap without any investment or decison-making, more WoM generation that Kairos Fateweaver, two of the best spells the skaven have access to, and all for just over 2k gold (most hardcarry MP lords are 3k+).

 

[Doctor 0]

Give me late stickers for just using search to see there's only a few posts referencing it, but is everyone playing REx and M2Ex?

 

[Big Brown Schlub]

Give me late stickers for just using search to see there's only a few posts referencing it, but is everyone playing REx and M2Ex?

I've played REX and it's the first time I could play Rome TW without having to fuck around with it and still have it run like crap, a problem I've had with the game since 2016. The QoL stuff they added is fine, I still haven't tried the new difficulties they added but making it run like butter on modern machines is the best part about it, the game is running the best it's ever been. End turn times are literally instant, for example. I love its potential the most, because they made it run on 64-bit they've also managed to remove some hard-coded limiters that the game had, like the hard faction limit, now just like Rome Remastered you can technically have unlimited factions, they also made scripting much easier which should allow modders even more freedom.

When it comes to M2EX I haven't played it simply because Medieval 2 still runs very well on my current PC but if there's ever a mod that requires M2EX (because it removes the same limitations just like REX does for Rome) then I'll get it.

 

[I Know 孤独のせい]

Even the RTS side is a relatively solved problem where results are gated by effort: Starcraft I maintains a decent AI (and even has a new flavour in Pluto AI built by a fan), Planetary Annihilation Titans has seen some impressive work over the years, and within my own interests Call to Arms GOH Ostfront has a mod using dev tools that has the AI through movement alone prove a decent threat. None of them are perfect, but they show what's possible for it with dedicated effort

I haven't had the time to thoroughly test other games and their respective AI's but it good to hear that there are some examples where the devs are at least trying.

TL;DR: blame CA management and the company's shareholders for focusing on fluff over mechanical substance; it all comes down to where they're paying their devs to focus their time.

Well we will have to agree to disagree here. This is plain incompetency on CA's part because they had Warscape development, Empire, Napoleon, Shogun 2, Rome 2, Attila, Britannia, WH 1 to do work on the AI since those games didn't have a billion different and often marginally impactful mechanics that they didn't have to teach the AI. There was enough time and money involved to get the basics right, Rome 2's infamous marketing also touched upon the AI. They don't have the time and money right now with Hyena's cancellation, WH3 disatrous launch and years dedicaded to fixing it instead of monetazing it,work on a new engine, WH 40K and M3. They have been stretched thin for a while so I might bemoan the state of the AI right now but I can understand it. But that wasn't the case in the past.

And I know this isn't serious at all and we are just shooting the shit about something we generally like but I sincerely hope that when WH 40K comes out and it displays the same issues that we had for the past two decades you don't try to argue that it is again an issue with time and money. It's a new engine that they have worked on for years, meant to carry the franchise for another decade or two without all the tech bloat from Warscape. There is no excuse for the AI to lack a basic understanding what it is meant to do in the game when 40K launches.