Sign in to follow this  
Rob Zacny

Episode 311: Total War: Attila

Recommended Posts

Rob Zacny and Troy "Flava Flavius" Goodfellow examine the latest Total War game, Attila. Following the dizzying heights of Shogun 2 and the despair-filled depths of Rome 2, Attila shows that Creative Assembly can still soar like so many proverbial eagles. The praise is effusive in this show as both hosts are happy to see the series return to form. Rob breaks it down: "Attila... basically awesome."

 

Listen here.

Share this post


Link to post
Share on other sites

So things are looking good for Total Warhammer then!

 

More seriously, great to hear that CA are back on track.

Share this post


Link to post
Share on other sites

I am really looking forward to this episode, because I just got done with twenty-hour hours of Total War: Attila, courtesy of Steam's free weekend, and I found it just to have covered all of the bad design decisions in Rome 2 with a wealth of band-aids. For every step in the right direction, like a slower timescale for strategic gameplay and a more flexible province system, there are steps back towards the wrong, like an interface even more crowded than before and more full of pointless numbers and bars to signify the presence of mostly meaningless systems.

 

The new horde system is great, although after double-digit hours of play I was still having difficulty figuring out the strategic value, in absolute terms addressing an entire campaign's worth of discrete decisions, of sacking versus looting versus razing versus occupying a specific settlement, but at least the presence of a settlement-agnostic gameplay style via hordes makes it a decision with some relevance beyond just making the decision itself. The factional politics are still total crap and I literally cannot believe that Creative Assembly doubled down on them, but it's not surprising that someone there is inspired by Crusader Kings 2. The actual combat is also still terrible -- fast and floaty, with the stupid "kill animation" system of Shogun 2 allowing for ridiculous outcomes because each soldier can only engage one other soldier at a time, therefore theoretically allowing a single soldier from a high-stats unit type to hold off infinite inferior enemies like it's 300 or something. The map's still too big, historically important factions are locked behind DLC from launch, and Creative Assembly still doesn't remember that they used to balance the campaign via faction-wide economics rather than counter-intuitive building chains, so the campaign gets dull before it's even remotely close to the finish.

 

I don't know. I actually look forward to editing this post, after I listen to the podcast, in order to acknowledge the points of history-savvy people who see something good in Attila, because right now it mostly just shows to me that Creative Assembly has the ability to read its own forums and borrow fixes to their design decisions from fan mods. Which, you know, is great, but...

Share this post


Link to post
Share on other sites

"historically important factions are locked behind DLC from launch"

 

I have no problem with this. Paradox and CK2 have shown (to me at least) that it's a perfectly fine business model. I've found the base game great value, and love that I can help support the game and continue to expand the number of interesting factions. It's a sensible business strategy and a good way for a company to spread the risk of big budget game design around a little more.

 

Which isn't to say that it can't be done badly. but just complaining about the concept rather than the implementation is wrong.

 

The rest of your points sound more worrying though.

Share this post


Link to post
Share on other sites

"historically important factions are locked behind DLC from launch"

 

I have no problem with this. Paradox and CK2 have shown (to me at least) that it's a perfectly fine business model. I've found the base game great value, and love that I can help support the game and continue to expand the number of interesting factions. It's a sensible business strategy and a good way for a company to spread the risk of big budget game design around a little more.

 

Which isn't to say that it can't be done badly. but just complaining about the concept rather than the implementation is wrong.

 

The rest of your points sound more worrying though.

 

It just feels weird, I guess? In most cases, except for the two Roman factions, the Huns, the Sassanids, and the major Germanic tribes (in essence, the staple of a game in this period), one faction is usually given as a "taste" for other factions of similar style. The Saxons are in the base game, but the Jutes, Geats, and Danes are in a DLC (the Angles aren't even an option, I don't know why). The Franks and Suebi are in the base game, but the Alemanni, Burgundians, and Longobards are in a DLC (why the Alemanni and Suebi are different factions, I don't know; also, another important faction like the Gepids again isn't an option).

 

The Celtic Culture Pack introduces an entirely new type of faction, to Creative Assembly's debatable credit, and I wouldn't doubt that a Nomadic Culture Pack with the Sarmatians and Roxolani is also forthcoming, but as it stands, the repeated feeling that I got was booting up the game to play one faction, finding it behind a paywall, and having to settle for a not-quite-but-good-enough faction instead. It just bothered me enough to make mention.

Share this post


Link to post
Share on other sites

Good game (as is Rome 2 now) but there are just too many design choices that are hamstringing the series.  Firstly, the province system is ridiculous.  Why shouldn't I be able to build walls around a city just because it's not the symbolic capital of a theoretical province?  Also, why should the populace of my settlement be deeply unhappy purely because another faction is mismanaging a settlement in the same province?  It makes no sense.  Plus the "giant" campaign map is really the smallest one ever used because it is simply a series of channels that armies are shuttled down rather than an expanse of land that you had to monitor with watch towers (a la Rome 1).  

 

Most crucially though are the inadequacies of the Warscape engine.  It may look pretty but the physics are far inferior to the engine used in Rome 1 over 10 years ago.  Then there was weight to the units and they actually fought each other.  Now soldiers can only really fight 1 v 1 and so they will simply mesh together in a big ball rather than retaining cohesion.  God, even terrain doesn't have the same effect as it did in the old game!

 

Basically, if you can tolerate the graphics, Rome 1 remains the superior game to Rome 2 or Attila.

Share this post


Link to post
Share on other sites

Anyone here remember how in original Rome, you could migrate population via building army of peasants then disbanding them on another city?  IDK why but I really miss doing that...  What I also don't miss is the 3 gold chevron peasant rebel stack :x 

 

About the combat and game engine, Turrican, I think you are way over estimating what vanilla Rome engine was capable of, and blaming flaws of Rome 2 and Atilla on the engine when it's poor design issues.  Shogun 2 had far better battles than Rome 2 AND Rome (I went back a year ago, it was atrocious, memories of complete overhauled mods gave false impression of the vanilla version) with older version of same engine so I think clearly the game design (regarding to combat) is at fault here.

 

Like, I agree something is terribly wrong with Rome 2's battles, just that the cause is probably balance/design rather than technical.  I'm pretty sure they can recreate exactly how battles happen in Rome in Warscape engine, but they just don't want to.

 

On the topic of battles in TW games, Gormongous, you mentioned that the 'execution' animation ruined the battle... hard to do a direct comparison (cause AFAIK, you can't just disable those?) but what makes you think that is more than just visual flavor?  Cause I've yet to seen any instance of few lone elite infantry holding off mobs of cheaper units in Shogun 2 at all.

Share this post


Link to post
Share on other sites

On the topic of battles in TW games, Gormongous, you mentioned that the 'execution' animation ruined the battle... hard to do a direct comparison (cause AFAIK, you can't just disable those?) but what makes you think that is more than just visual flavor?  Cause I've yet to seen any instance of few lone elite infantry holding off mobs of cheaper units in Shogun 2 at all.

 

The advantage with the old "battle engine" for Rome: Total War and Medieval 2: Total War is that it was a physics-based engine. In the scripting language of those games, units had weight values that were modified by attack, defense, armor, and shield variables, which allowed for the engine to calculate the outcome of basically any interaction from enemy units attacking each other to them bumping into a friendly unit (within limits, if there was an animation to depict it; I recall that Medieval 2: Total War had no animation for units armed with two-handed weapons to attack mounted units, so they were unable to kill them at all, but mounted units are a special case since they're technically two units with different weight values and stat variables calculated together and...). If a unit lost cohesion, there'd be less "weight" behind its constituent members' attacks, making it a noticeably less effective fighting force.

 

The Warscape engine, created for Empire and now in its fourth or fifth iteration (depending on how you count Total War: Shogun 2 - Fall of the Samurai), was not initially designed for extended animation of melee combat and it shows in how it's designed to calculate inter-unit combat and death. A unit performs an action directly upon another unit via an animation, rather than simply placing an action value out there to affect multiple units, and, based on a comparison between the stats of the two units, an animation is selected for the receiving unit. It's an animations-first system rather than a stats-first system. It's a one-to-one thing that works great with rifle fire and bayonet charges, but it is simply not designed to calculate the effects of adjacent unit actions, which are what makes melee combat more dynamic than just several dozen guys pairing off to fight duels on the front line.

 

The flaws of Warscape are not really detectable in Shogun 2, just because of the individualized nature of samurai-style combat, but it means that highly disciplined units in Rome 2, like elite legionaries or phalangites, will break ranks and blob into a formless circle of bodies because, the way the engine is designed, keeping ranks is actually less efficient, with less of your units are making contact and able to do damage to enemy units. There's even a hidden stat debuff associated with "loose formation" because otherwise it's more powerful than normal ranks, since it lets the same size unit gain more frontage and cover more space. The weight of formed ranks behind the front line means nothing, and that's an enduring problem that Creative Assembly hasn't really been able to overcome with unit tuning and special powers.

Share this post


Link to post
Share on other sites

The flaws of Warscape are not really detectable in Shogun 2, just because of the individualized nature of samurai-style combat, but it means that highly disciplined units in Rome 2, like elite legionaries or phalangites, will break ranks and blob into a formless circle of bodies because, the way the engine is designed, keeping ranks is actually less efficient, with less of your units are making contact and able to do damage to enemy units. There's even a hidden stat debuff associated with "loose formation" because otherwise it's more powerful than normal ranks, since it lets the same size unit gain more frontage and cover more space.

 

I feel like I read this exact paragraph in TWCenter a year ago... and it kinda bugs me a bit (forget history for sec) because in Shogun 2, yari ashigaru lives and dies by having good unit cohesion via yari wall and for good half of the game, bulk of your melee forces are yari ashigaru (one of few reasons that makes Oda absurdly good)... so to say that Shogun 2 is fought in 'individualized' fashion just smacks completely untrue to my experience :/

 

So like, you said loose formation comes with penalty... does that mean yari wall formation gives hidden stat buff?  Because it clearly works wonders.

Share this post


Link to post
Share on other sites

So like, you said loose formation comes with penalty... does that mean yari wall formation gives hidden stat buff?  Because it clearly works wonders.

 

Yes, it does. All formations have hidden stat buffs and debuffs to make them approximate how they're supposed to work in reality. My guess, as someone who did a ton of modding in Rome: Total War and Medieval 2: Total War modding but none in subsequent games, is that it works like pike formation in Rome 2, which is to say that yari ashigaru have huge bonuses if they're using spear animations while their unit is in its special formation, but if an enemy moves past the animation capture point of the spear tips, the nearest member of the unit "drops" its spear by switching to a sword model and attendant animations, thereby losing the bonuses, which en masse recreate the overwhelming and collapse of an organized formation like that.

 

All told, it's a remarkably clever kludge to force unit-scale tactics into a game that doesn't simulate units except as a collection of similar-but-discrete individuals, but it's a kludge nonetheless, since it makes the most effective tactic against a spear or shield wall be forcing your unit to run through the enemy rather than attacking it, in the hopes that the animation parameters can be short-circuited, the bonuses lost, and the formation neutralized.

Share this post


Link to post
Share on other sites

If both games did the same trick then I'm quite surprised by how much ball they dropped in Rome 2 cause in Shogun 2 it completely fooled me cause yari wall was awesome and for all I knew, it was simulating what it should have just fine.  But in Rome 2, I saw those cracks that you mentioned (I guess it helped that yari ashigaru had no secondary weapon to switch to like that lol) and while I didn't know the details, stuff like that (along with crap ton of other stuff Rome 2 got wrong) kinda destroyed the magic for me.

 

But man, Shogun 2 was... like it's the only TW game that I actually completed multiple times without heavy modding (then again, it's partly cause lot of total conversions didn't make it past M2).  Most mods I used for Shogun 2 were few skins and just doubling the unit size, which is pretty freaking amazing how well that game's battle scaled with 2x the bodies to make the clash last little longer.  And to me Shogun 2 felt like a pike and bow/shot game.  A fast one, but still very pike centric and hence unit cohesion felt important and well represented.

 

Edit: Sooo any updated thoughts on Atilla?  It's still on sale for a bit longer so I might grab it if you have change of mind after hearing the podcast but if not, I'll probably pass cause it sounds like battle system is still pretty unsatisfying.

Share this post


Link to post
Share on other sites

The advantage with the old "battle engine" for Rome: Total War and Medieval 2: Total War is that it was a physics-based engine. In the scripting language of those games, units had weight values that were modified by attack, defense, armor, and shield variables, which allowed for the engine to calculate the outcome of basically any interaction from enemy units attacking each other to them bumping into a friendly unit (within limits, if there was an animation to depict it; I recall that Medieval 2: Total War had no animation for units armed with two-handed weapons to attack mounted units, so they were unable to kill them at all, but mounted units are a special case since they're technically two units with different weight values and stat variables calculated together and...). If a unit lost cohesion, there'd be less "weight" behind its constituent members' attacks, making it a noticeably less effective fighting force.

 

The Warscape engine, created for Empire and now in its fourth or fifth iteration (depending on how you count Total War: Shogun 2 - Fall of the Samurai), was not initially designed for extended animation of melee combat and it shows in how it's designed to calculate inter-unit combat and death. A unit performs an action directly upon another unit via an animation, rather than simply placing an action value out there to affect multiple units, and, based on a comparison between the stats of the two units, an animation is selected for the receiving unit. It's an animations-first system rather than a stats-first system. It's a one-to-one thing that works great with rifle fire and bayonet charges, but it is simply not designed to calculate the effects of adjacent unit actions, which are what makes melee combat more dynamic than just several dozen guys pairing off to fight duels on the front line.

 

The flaws of Warscape are not really detectable in Shogun 2, just because of the individualized nature of samurai-style combat, but it means that highly disciplined units in Rome 2, like elite legionaries or phalangites, will break ranks and blob into a formless circle of bodies because, the way the engine is designed, keeping ranks is actually less efficient, with less of your units are making contact and able to do damage to enemy units. There's even a hidden stat debuff associated with "loose formation" because otherwise it's more powerful than normal ranks, since it lets the same size unit gain more frontage and cover more space. The weight of formed ranks behind the front line means nothing, and that's an enduring problem that Creative Assembly hasn't really been able to overcome with unit tuning and special powers.

 

 

Yeah, the problems with the battles all boils down to this.  

Share this post


Link to post
Share on other sites

So what do you guys think of how it worked out in Shogun 2?  Fall of the Samurai has everyone shooting so not much there to talk about in melee combat.  But as I described, my experience of Shogun 2 I didn't want my melee units blobbing so that I can pull them out easier if necessary and yari wall (which packs the units in much tighter and kept cohesion) worked very well... didn't feel like bunch of individuals running around doing their own thing, but is that just my own experience?  Like, what were you guys' experiences with Shogun 2 which uses this same engine but gets much less flak (probably because game was overall much more finished and polished) about its combat?

 

Also is this really warscape engine's fault?  I seriously doubt this feature is something that's so hardcoded in that they would need to overhaul the engine.  More likely, CA probably wanted this regardless of goofy overall larger battle problem so that it looks nicer (fancy animations finishing up properly) close up.  Like, it's easily fixable but they don't want see it as a problem because the tradeoff (uninterrupted animation sequence) is something they priotized and is mutually exclusive (if damage overrode animation, units would drop dead without visual cues (and I recall exactly that situation happening in older TW games, one of reason why I totally buy your explanations Gormongous cause in Rome/M2 I recall seeing soldiers just dropping dead from unseen attacks)), if that makes sense.

 

Sorry it's just my pet peeve talking about game engines (the underdog complex from using Gamemaker lol) because I see them as much more capable than its games lead on.  Like when discussing engine problems, I would consider more technical problems like texture loading in UE3 to be engine problem... thing like this (animation > damage vs damage > animation) seems much more likely to be a conscious design choice.

Share this post


Link to post
Share on other sites

After Rome 2 I was at first "cautious" about Attila, which changed to "Cautious Optimistic" once the dev journals/blogs/video start talking about the game, to a finally "Really Surprised" once the game was out.

 

Got to agree with Troy, Attila again proves the point that CA best ability is with focused titles, which improve my hopes on the Warhammer, since it will be focused on four armies instead of all of them. Also, yeah the part of exploring around and find out was going on is really fun, either when I was playing the Saxons (and leaving Britain), or moving my Vandal hordes toward North Africa, there was this feeling of "what hell is going on around here?" finding either separatist factions, romans besieged everywhere, devastations and other stuff around.

 

On battles (I am mostly talking on visual part) what I felt is that in Shogun 2, the mix of animations, plus unit variations (I mean in sense of units with one-handed weapons/spears/bows/two-handed/ect...) and theme (visually, shogun draw a lot from classic samurai moves, just watch the battle scene in Haven and Earth (1990, Ten to Chi) and compare with the Shogun 2) and even the addition of blood (and the animations, again while over the top,again it did draw a lot from movies) all worked and combined together very well. Rome 2 had less unit variation (mostly where units with either sword or spear and shield, almost no two-handed weapon units), animations where less distinctive (since almost everybody is using the same weapon) and much like in King Arthur, the huge shields made unit blob visually worst. Attila improves a bit on unit variation (slight more two-handed units and other types), color is slight better, animations where improved (actually, unit height variation and facial expression work a lot better).

 

I have been playing the Last Roman campaign, it´s quite fun - since it take you to control Roman "Hordes" (the Roman Expendition) and you can either fullfill Justinian orders (and try to balance around the desires of both your wife and his wife, they both give you missions that might even contradict) or try to became independent.

 

Rob commented on the Sprites: talking on visuals alone, yeah, for Shogun 1 it worked since the small number of units allowed each one to be more distinctive, but while the same kind worked in Medieval 1, I did remember felt (talking from memory) that sometimes units distinction was quite lower due the sheer amount* units that Medieval 1 had. I remember having several kinds of Turkish cavalry that that where very similar in stats/function and visual.

 

* I remember Medieval 1 having lots of units, but sometimes to get many of them you need have either special provinces or very developed ones, meaning that most times when you finally could get them, due the way the game micromanagement of units and how it take a single turn to move one province to another, it wasn´t much helpful, since it often takes too much time to move them to the front (since this provinces might be away from frontlines) and maybe back to refill (I remember feeling that around with Medieval II). By the way, one the changes I mostly enjoyed was the way units could be recruited and reinforced, since, while not very realistic, it take way much of the previous games annoying micromanagement and allowed for more mixing of units.

 

By the way, I guess that everybody saw CA just released again Shogun 1 and Medieval 1 on Steam? I guess they got the rights back, because Shogun was under EA while Medieval was under Activision.

 

Also small curiosity: did anyone ever played that Takeda 3 (http://www.gamersgate.com/DD-TAKEDA3/takeda-3) or Strenght & Honor 2 (http://www.gamersgate.com/DD-SH2/strength-and-honour-2) or even Sango 2 (http://www.gamersgate.com/DD-SANGO2/sango-2) all where some kind of Shogun 1 / Medieval 1 visual/engine (??) like games. I never heard about them (at least until I know the gamersgate site, their player review are quite mixed) and I have no idea how the exactly look or play (I do remember seeing a gameplay video and not begin impressed) but its curious to see them.

 

Got to also agree with Rob, that if the narrator saying which unit was fleeing would be great, but I also felt that would mean a lot of more voice work and could lead to narrator spamming comments. The closest thing I saw to what Rob said, it was on Warhammer: Dark Omen, where each unit was lead by a character and you would see a small window showing the animated head of this character saying lines or asking help. I do however remember that Napoleon would open a small window showing where something the narrator was saying was happening (but the visual of the window wasn´t very good).

Share this post


Link to post
Share on other sites

I didn't play any of those games, but I did play Kessen and Kingdom Under Fire, but both are console exclusive so it's very hard to play them now.


Kessen is more pure tactical battle game, while Kingdom Under Fire is this weird but amazing mix of TW-lite battle with character action game (!) that was just unbelievable at the time.  Like, think bit of dynasty warrior except how you controlled (unlike in DW where they just kinda sit still) your army actually mattered.

 

Oh wow, Warhammer: Dark Omen, bringing those memories back I was too young and stupid (and wasn't capable of reading English) to properly play those games then! :x

Share this post


Link to post
Share on other sites

Yeah, having listened to the episode, Rob and Troy seem to me like two guys for whom a (justified) love of the period leads them to see the Total War: Attila in a more flattering light than I can, as someone who put almost a hundred hours into Rome 2. When a game like Attila is built around a period full of chaos and decay, perennially wonky things from Rome 2 like poor economic tuning for factions and hate-you-forever diplomacy fit the narrative, even if they come from half-baked or defective systems. The whole time that Rob was telling me his admittedly cool story about settling too soon in Salona as the Ostrogothic horde, I was thinking that it was the sort of thing I'd seen in every single Total War game since the first Rome, only it's actually thematic when your expeditionary force is inexplicably pissing off all of its target's neighbors if that force is an unruly Germanic horde. If you can't keep the game's AI from being terrible, force the player into situations where terrible AI seems authentic!

 

Overall, listening to the podcast and playing a final few hours of Attila myself, I think that it's a mediocre game that's brought to life by a setting that's fascinating but under-explored elsewhere in popular media. In particular, the horde mechanics are a great improvement on the original Rome's Barbarian Invasion expansion, but almost everything else is simply jury-rigged in direct response to criticisms of Rome 2. For example, tying the loyalty and revolt mechanics from Barbarian Invasion into the political system from Rome 2 actually gives an in-game consequence to the ludicrous "Your bucket is emptying, do you refill it even though it'll also fill someone else's bucket, too?" decisions of the latter, but it means that the most important values for a general are their loyalty and influence, two values between two and four clicks into Attila's awful interface, while authority, cunning, and zeal continue to be spotlighted as important values even though they directly influence nothing in the game on their own. That's what Total War: Attila is to me: making the most of a broken design by introducing a new setting with new systems that make the majority of Rome 2's awful design decisions feel relevant.

 

Also is this really warscape engine's fault?  I seriously doubt this feature is something that's so hardcoded in that they would need to overhaul the engine.  More likely, CA probably wanted this regardless of goofy overall larger battle problem so that it looks nicer (fancy animations finishing up properly) close up.  Like, it's easily fixable but they don't want see it as a problem because the tradeoff (uninterrupted animation sequence) is something they priotized and is mutually exclusive (if damage overrode animation, units would drop dead without visual cues (and I recall exactly that situation happening in older TW games, one of reason why I totally buy your explanations Gormongous cause in Rome/M2 I recall seeing soldiers just dropping dead from unseen attacks)), if that makes sense.

 

Sorry it's just my pet peeve talking about game engines (the underdog complex from using Gamemaker lol) because I see them as much more capable than its games lead on.  Like when discussing engine problems, I would consider more technical problems like texture loading in UE3 to be engine problem... thing like this (animation > damage vs damage > animation) seems much more likely to be a conscious design choice.

 

I mean, I can't say for certain because I'm neither a game designer nor employed by Creative Assembly, but the latter are adamant that the one-on-one, animation-deterministic battle system is hardcoded into the Warscape engine and that it would be an overwhelming expense of money and time to introduce systems-based physics calculations or even to make the outcome of two units' contact partially independent from the animation library. They could be lying, or at least exaggerating, but either way I don't really blame them, because press and fans will be impressed by highly-choreographed kill animations in a way that they'll never be impressed by the dead body of a horseman whose horse was run through by a pikeman flying through the air and knocking down a different pikeman, allowing a different horseman to penetrate the pike wall (or at least the behind-the-scenes calculations that allow that to happen, even if the animations don't quite reflect it). The latter is more technically impressive, but the former is more comprehensible and therefore easier to sell. There's not really any money in a truly accurate battle system anyway, since almost no one has realistic expectations of how it would be, so flattering their preconceptions with Hollywood-style "cinematic combat" is the more judicious alternative.

 

Why was Shogun 2 better? Each faction shared the same units, with only a single DLC introducing one faction-specific unit to each faction. They were able to be balanced rigorously according to a very strong rock-paper-scissors multichotomy. Cavalry countered archers hard and swords soft, archers countered spears and swords hard, spears countered cavalry hard, swords countered spears hard and cavalry soft. No unit, not even a gunpowder unit, was powerful enough to exist outside of that network of hard and soft counters. Special powers were mostly limited to recreating tactical situations where a unit would have an effect on combat beyond its base stats (usually at the cost of fatigue, which was still incredibly important in Shogun 2; here is where the yari ashigaru's spear wall fits in, although other special powers were already headed to the "instant magic" territory of Rome 2 with the elite units in Fall of the Samurai). In short, it was just a more carefully considered game in which each unit had a clearly defined role that could only be partially replaced by another unit, so there was never an instance of something like in Rome 2 or Attila where I'd throw a mix of Germanic Levies, Freemen, Spear Levies, Bagaudae, and Frankish Spears into a fight because it was unclear which one of them was the proper unit to counter a mix of Germanic Brigands and Saxon Spears, backed up by Germanic Hurlers and Germanic Archers.

 

Discussing why the Shogun 2 campaign is better than the Rome 2 or Attila campaigns is a slightly different matter, though, and much more focused on the inability of the AI to sustain the same rate of expansion as the player on a map of over two hundred provinces. Attila partially fixes that by making the Little Ice Age drastically devalue provinces over the course of the campaign and by allowing settlements to be razed, but both of those also end up disadvantaging the AI more than the player. If the AI actually had to pay upkeep on its armies, which it doesn't have to do in Rome 2 or Attila on any difficulty level, most factions would drive themselves into bankruptcy and elimination before the player even got to them. It's not pretty, not at all.

Share this post


Link to post
Share on other sites

Hmm that sounds like great explanation for why Shogun 2 worked and why Rome 2 felt really floppy.

 

So final verdict on Atilla for you is average, ok I think I'll pass on it then~

 

I hated the zeal-authority-cunning stats from Rome 2 so much, so to hear that it's still in it.... arrgh.

Share this post


Link to post
Share on other sites

So final verdict on Atilla for you is average, ok I think I'll pass on it then~

 

If you have a strong personal connection to the period or to the factions therein, I think it's worth buying. If you are just a fan of Total War games looking for the next hit in the series, I don't think so, no.

 

I hated the zeal-authority-cunning stats from Rome 2 so much, so to hear that it's still in it.... arrgh.

 

I really do not understand why it is in there. Sure, it made a little sense to remove it when Rome 2 de-emphasized generals as provincial governors, so the three-stat system had be repurposed into a different kind of balance between different styles of military leadership, but generals as provincial governors are back in Attila, so now the authority/cunning/zeal system has to be re-repurposed back into multivalent stats... I don't know, it boggles the mind sometimes.

Share this post


Link to post
Share on other sites

Yeah, having listened to the episode, Rob and Troy seem to me like two guys for whom a (justified) love of the period leads them to see the Total War: Attila in a more flattering light than I can, as someone who put almost a hundred hours into Rome 2. When a game like Attila is built around a period full of chaos and decay, perennially wonky things from Rome 2 like poor economic tuning for factions and hate-you-forever diplomacy fit the narrative, even if they come from half-baked or defective systems. The whole time that Rob was telling me his admittedly cool story about settling too soon in Salona as the Ostrogothic horde, I was thinking that it was the sort of thing I'd seen in every single Total War game since the first Rome, only it's actually thematic when your expeditionary force inexplicably pissing off all of its target's neighbors if that force is an unruly Germanic horde. If you can't keep the game's AI from being terrible, force the player into situations where terrible AI seems authentic!

 

Overall, listening to the podcast and playing a final few hours of Attila myself, I think that it's a mediocre game that's brought to life by a setting that's fascinating but under-explored elsewhere in popular media. In particular, the horde mechanics are a great improvement on the original Rome's Barbarian Invasion expansion, but almost everything else is simply jury-rigged in direct response to criticisms of Rome 2. For example, tying the loyalty and revolt mechanics from Barbarian Invasion into the political system from Rome 2 actually gives an in-game consequence to the ludicrous "Your bucket is emptying, do you refill it even though it'll also fill someone else's bucket, too?" decisions of the latter, but it means that the most important values for a general are their loyalty and influence, two values between two and four clicks into Attila's awful interface, while authority, cunning, and zeal continue to be spotlighted as important values even though they directly influence nothing in the game on their own. That's what Total War: Attila is to me: making the most of a broken design by introducing a new setting with new systems that make the majority of Rome 2's awful design decisions feel relevant.

 

 

I mean, I can't say for certain because I'm neither a game designer nor employed by Creative Assembly, but the latter are adamant that the one-on-one, animation-deterministic battle system is hardcoded into the Warscape engine and that it would be an overwhelming expense of money and time to introduce systems-based physics calculations or even to make the outcome of two units' contact partially independent from the animation library. They could be lying, or at least exaggerating, but either way I don't really blame them, because press and fans will be impressed by highly-choreographed kill animations in a way that they'll never be impressed by the dead body of a horseman whose horse was run through by a pikeman flying through the air and knocking down a different pikeman, allowing a different horseman to penetrate the pike wall (or at least the behind-the-scenes calculations that allow that to happen, even if the animations don't quite reflect it). The latter is more technically impressive, but the former is more comprehensible and therefore easier to sell. There's not really any money in a truly accurate battle system anyway, since almost no one has realistic expectations of how it would be, so flattering their preconceptions with Hollywood-style "cinematic combat" is the more judicious alternative.

 

Why was Shogun 2 better? Each faction shared the same units, with only a single DLC introducing one faction-specific unit to each faction. They were able to be balanced rigorously according to a very strong rock-paper-scissors multichotomy. Cavalry countered archers hard and swords soft, archers countered spears and swords hard, spears countered cavalry hard, swords countered spears hard and cavalry soft. No unit, not even a gunpowder unit, was powerful enough to exist outside of that network of hard and soft counters. Special powers were mostly limited to recreating tactical situations where a unit would have an effect on combat beyond its base stats (usually at the cost of fatigue, which was still incredibly important in Shogun 2; here is where the yari ashigaru's spear wall fits in, although other special powers were already headed to the "instant magic" territory of Rome 2 with the elite units in Fall of the Samurai). In short, it was just a more carefully considered game in which each unit had a clearly defined role that could only be partially replaced by another unit, so there was never an instance of something like in Rome 2 or Attila where I'd throw a mix of Germanic Levies, Freemen, Spear Levies, Bagaudae, and Frankish Spears into a fight because it was unclear which one of them was the proper unit to counter a mix of Germanic Brigands and Saxon Spears, backed up by Germanic Hurlers and Germanic Archers.

 

Discussing why the Shogun 2 campaign is better than the Rome 2 or Attila campaigns is a slightly different matter, though, and much more focused on the inability of the AI to sustain the same rate of expansion as the player on a map of over two hundred provinces. Attila partially fixes that by making the Little Ice Age drastically devalue provinces over the course of the campaign and by allowing settlements to be razed, but both of those also end up disadvantaging the AI more than the player. If the AI actually had to pay upkeep on its armies, which it doesn't have to do in Rome 2 or Attila on any difficulty level, most factions would drive themselves into bankruptcy and elimination before the player even got to them. It's not pretty, not at all.

 

It's a really good point- Shogun 2 also had some really subtle interactions that come through when you dig into the game, too.  Ashigaru bowmen, for example, had a hidden debuff against armor, which made most samurai units largely impervious to them unless they got focused down and stood there taking it, and samurai units tended to be more flexible, but more expensive.  You still used ashigaru of all types throughout the campaign, and the optimal play involved using buff-stacked ashigaru the whole way through, which is actually pretty thematic, too.

 

I've said elsewhere that Rome2 and Attila have more variety- but it's skin-deep, just different skins on the same heavy infantry units.

 

That all being said, one of the biggest problems with the TW series' combat is that units don't work together well- it's very cumbersome to try to come up with a formation where your matchlock ashigaru can be protected by yari ashigaru, even though that's exactly what happened.  You basically see flanking, putting bowmen behind someone else, and that's about it.

Share this post


Link to post
Share on other sites

I really appreciate 3MA doing an Attila:Total War podcast.  I am heavily invested in this series and find in general that TW offers something unique and thousands of hours of entertainment.  That said, I find myself often frustrated by the game design choices and the lack of playtesting or polish evident in many releases such as Rome 2.

 

I've read a lot of positive stuff about Attila, and obviously Rob and Troy were pleased with the game, but I find myself mostly in agreement with the negative discussion I'm reading here.  What I haven't seen discussed much though is the campaign AI, which I think is a fatal flaw in the game and something that seems to get a pass from folks.

 

Specifically, I find the campaign AI in Attila is either (i) too passive and running away all the time with armies often taking to the sea to remain offshore for many turns, seemingly adrift, or (ii)  overly aggressive, with certain hostile factions stalking the player across continents, seemingly uninterested in anything but annoying the player for no apparent strategic reason.  In both Attila and Rome 2 games the AI seems challenged to build empires, and certainly doesn't create the strong ones we've seen in other games, such as Shogun 2. 

 

With Attila there is a temptation, becuase of the nice horde mechanic, or the family tree, or the intricate gamey struggle to balance food and squalor, to overlook what is actually going on in the campaign: that is, everything and nothing, all at once.  With all the torching of regions and sudden appearance and disappearance of factions, you might not notice that no one is really consolidating or gretting stronger.  You might not notice the eternally two-region Ebdanians are sending one stack every four turns in your direction, hoping against hope that you might not notice it and leave one of your cities undefended.  How this beneifts the Ebdanians is a mystery of course.  Or you might be willing to forgive the CAI suicidially attacking to take one of your cities when three of your stacks that will annihilate it the next turn are just out of it's range.  You also might not notice how the Huns will continuously respawn unless you kill the game's protagonist three times in battle.

 

I'm not forgiving Attlia, though, as tempted as I was to do so at twenty five hours, by sixty hours I'd seen through the campaign AI and was annoyed by all its extraneousness and silliness.  To its credit I do feel like CA made an attempt here to provide a multidimensional and challenging game.  Anyway the CAI is the reason Attila gets my low score; and I am a bit disappointed to say I am currently finding Rome 2 (modded with Divide et Impera) more fun than this game. 

Share this post


Link to post
Share on other sites

Specifically, I find the campaign AI in Attila is either (i) too passive and running away all the time with armies often taking to the sea to remain offshore for many turns, seemingly adrift, or (ii)  overly aggressive, with certain hostile factions stalking the player across continents, seemingly uninterested in anything but annoying the player for no apparent strategic reason.  In both Attila and Rome 2 games the AI seems challenged to build empires, and certainly doesn't create the strong ones we've seen in other games, such as Shogun 2. 

 

With Attila there is a temptation, becuase of the nice horde mechanic, or the family tree, or the intricate gamey struggle to balance food and squalor, to overlook what is actually going on in the campaign: that is, everything and nothing, all at once.  With all the torching of regions and sudden appearance and disappearance of factions, you might not notice that no one is really consolidating or gretting stronger.  You might not notice the eternally two-region Ebdanians are sending one stack every four turns in your direction, hoping against hope that you might not notice it and leave one of your cities undefended.  How this beneifts the Ebdanians is a mystery of course.  Or you might be willing to forgive the CAI suicidially attacking to take one of your cities when three of your stacks that will annihilate it the next turn are just out of it's range.  You also might not notice how the Huns will continuously respawn unless you kill the game's protagonist three times in battle.

 

This is an especially good point. I'd actually be quite interested in a game where the central challenge is balancing the food you need to feed your people with the money you need to protect them, but only if the other factions in the game are also facing that challenge. In Total War: Attila, they most emphatically aren't. You could be the cleverest sort of player, scouting out which of the AI's settlements are producing all its food, and sack or raze them, but it won't hurt the AI in the slightest, because they get no penalties for running out of food. You could outflank them strategically and trap an AI army in the field throughout winter, but they'll be just as strong come spring, because they get no attrition for the seasons. It's impossible to cripple the AI economically, because they get free money if their income drops below three thousand a turn. It's also impossible to surprise the AI, because they don't have fog of war.

 

If all other things were equal, all of these advantages would make the AI unbeatable, but the AI's been struggling to play Creative Assembly's games since the transition to a 3D campaign map, especially in terms of building armies and coordinating them against the player, so in practice, all of these little "cheats" just cover up that the player's playing Mad Dog McCree with the AI instead of being on a level playing field. Sometimes the illusion is perfect, when the Franks suddenly hit the outlying settlements of your small Saxon kingdom with two armies you had no idea that they had, and sometimes it's not, when a Pictish army crosses the North Sea to attack your small Saxon kingdom because you were trading with the Roman separatists in Britain even though the Picts are losing a war with their immediate neighbors the Caledonians.

 

It's funny, really, because I think some of the biggest design failures in Rome 2 are all the little mechanics and features that don't directly affect anything but that the player has to keep touching and pulling for optimal play. It seems like the AI can't really handle them, either!

Share this post


Link to post
Share on other sites

This is an especially good point. I'd actually be quite interested in a game where the central challenge is balancing the food you need to feed your people with the money you need to protect them, but only if the other factions in the game are also facing that challenge.

 

Basically TW should just learn from Civ games.

 

Oh my dream~  Civ + EU + TW... it would either be the best or worst abomination.

Share this post


Link to post
Share on other sites

Man, you guys should do the next Total War show. A lot of useful information here.

 

You are too kind, we had long time to ponder and rewrite our opinions here, much easier than recording for podcast~ <3

 

BTW how many of you guys use sabotage + auto resolve 'exploit' in TW games?  IDK if it existed in older titles but I figured it out in Shogun 2 and it always kinda kills the final 1/3 of the game (oddly enough, also made civil war lot more palatable cause it was lot easier to deal with) for me because I'm one of those player who is always compelled to pursue the most optimal known path.

 

Basically the exploit is from Shogun 2 and on (maybe in previous titles too but IDK), you can have agents that can cripple enemy army's move.  This also deactivates their capability to reinforce nearby allies.  And the way auto resolve works in recent TW games heavily favor number over quality after some tipping point.

 

So what I do is... get like 2 ~ 3 stack of CHEAPEST units possible, and just chain sabotage auto resolve to pick any armies apart 1 at a time with auto resolve.  This is how I beat legendary difficulty for both Shogun 2 and Rome 2 (TWCenter first for Egypt legendary clear hoooraaaaahhh).

Share this post


Link to post
Share on other sites

You are too kind, we had long time to ponder and rewrite our opinions here, much easier than recording for podcast~ <3

 

BTW how many of you guys use sabotage + auto resolve 'exploit' in TW games?  IDK if it existed in older titles but I figured it out in Shogun 2 and it always kinda kills the final 1/3 of the game (oddly enough, also made civil war lot more palatable cause it was lot easier to deal with) for me because I'm one of those player who is always compelled to pursue the most optimal known path.

 

Basically the exploit is from Shogun 2 and on (maybe in previous titles too but IDK), you can have agents that can cripple enemy army's move.  This also deactivates their capability to reinforce nearby allies.  And the way auto resolve works in recent TW games heavily favor number over quality after some tipping point.

 

So what I do is... get like 2 ~ 3 stack of CHEAPEST units possible, and just chain sabotage auto resolve to pick any armies apart 1 at a time with auto resolve.  This is how I beat legendary difficulty for both Shogun 2 and Rome 2 (TWCenter first for Egypt legendary clear hoooraaaaahhh).

 

That sounds like a recent exploit. I know that, in Rome and Medieval 2, auto-resolve was reviled for how "unfairly" it was calculated. Basically, the computer would take the two stacks and go unit by unit comparing the stats and assigning casualties, starting at the beginning and skipping the generals, until the amount of casualties for one side or the other met a preset threshold, based on total army morale, for a rout. Since units are ordered by default according to their melee combat effectiveness, the most expensive and elite units to retrain (remember retraining to replenish units? They got rid of it because it was apparently impossible to make the AI do it) would always take the brunt of the fighting, while the trash mobs further down the roster barely ever got touched. Ranged and siege units were also vastly undervalued because, even though their ranged value was used, they got their simulated casualties from melee units anyway. Finally, at least in Rome, walls weren't even figured into the calculations when auto-resolving a siege assault. Basically, it was an incredibly clumsy system with a lot of flaws and I generally like the auto-resolve in the Warscape engine better, although it puts out wonky results almost as often.

 

Man, you guys should do the next Total War show. A lot of useful information here.

 

Like Gaizokubanou says, we're just fans of the series with too much time to think. Almost all my knowledge of Total War comes from a semester in college that I wasted as a tester for Rome: Total Realism, and from applying those experiences to criticisms of the differences between the Battle Engine and Warscape. I loved listening to the podcast, because I have a passion for the so-called Dark Ages as a period and hearing other people echo that passion was great, but I really don't think Attila is where it could have been, from my several dozen hours of play. Maybe, as Bruce Geryk might say, my doctoral studies have made me impossible to satisfy when it comes to late antique and the early medieval periods, but I still hope. At the Gates is coming someday, and the team for Europa Barbarorum 2 is planning some vague "summer release" to be their first full-featured version of the mod.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this