Total War Saga: Thrones of Britannia

[diamutene]

After conquest the choice is very few, where is sack and go (without settlement)? I don't like building cards this type. Attila cards good. Sorry for my bad English.

[zoner16]

I worry that's, unfortunately, going to be a major flaw of the battles in the foreseeable TW future. This is not because of the CA deficiencies, but because the casual players want to have fast battles. I mean, the majority of the players don't want to be military tactical leaders, but just to have some fun. I wouldn't cry wolf at the CA, it's the players fault, the CA is just giving what they want.
What the CA could do, is to enable modding in such a way that somebody would create slower battles in a right way.

Until now, I prefer to stick to the Medieval 2 engine and play battles there (even if they're not perfect, to be sure).

There's also another problem with making the battles longer that I didn't realize until the DEI guys pointed it out some time ago. The AI's ability to conduct a battle wanes considerably as time goes on, while the player's increases. The AI has the edge when it comes to micro and multitasking, but we routinely shoot that in the foot whenever we slow down or push pause.

Conversely, the player has a massive edge when it comes to long term planning, preparing for uncertainties, and dealing with irregularities. The longer a battle lasts, the more of these pop up, and the less the AI is capable of making good decisions. This is probably not going to change any time soon. You can extend the AI's planning capabilities with better modeling, but doing so is a pretty big undertaking, very likely to create odd corner cases, and even more subject to compute bottlenecks.

[Jurand of Cracow]

There's also another problem with making the battles longer that I didn't realize until the DEI guys pointed it out some time ago. The AI's ability to conduct a battle wanes considerably as time goes on, while the player's increases. The AI has the edge when it comes to micro and multitasking, but we routinely shoot that in the foot whenever we slow down or push pause.

Conversely, the player has a massive edge when it comes to long term planning, preparing for uncertainties, and dealing with irregularities. The longer a battle lasts, the more of these pop up, and the less the AI is capable of making good decisions. This is probably not going to change any time soon. You can extend the AI's planning capabilities with better modeling, but doing so is a pretty big undertaking, very likely to create odd corner cases, and even more subject to compute bottlenecks.

With all the advances in the science on the artificial intelligence I don't think it's unreasonable to expect a game producer to provide us with a good AI, even if it's a big undertaking.
I think the CA doesn't do it because (it believes) that's this investment won't pay back because the players want simple games. An advanced AI doesn't make the difference for them.

[zoner16]

With all the advances in the science on the artificial intelligence I don't think it's unreasonable to expect a game producer to provide us with a good AI, even if it's a big undertaking.
I think the CA doesn't do it because (it believes) that's this investment won't pay back because the players want simple games. An advanced AI doesn't make the difference for them.

The advancements in AI tend to be focused on the models and heuristics that can theoretically be applied to general classes of problems. The actual work of implementing them correctly to a specific problem is just as difficult if not moreso than developing the models in the first place.

The problem with applying any of this to Total War is that 1) the problem doesn't stay still and 2) even if you do it right, the performance hit might waste it and 3) the people who do this right are some of the most expensive people in industry.


Regardless, I think that the target time of 12-15 minutes for a full stack, open field battle is reasonable from both an AI perspective and the fact that most people don't have a huge amount of time to dedicate to a single game. Vanilla is built for the general strategy game market, not Total War enthusiasts. Providing good tools for modders to fiddle with this gives most of us what we want in some way.

[geala]

Modding unit speed on the battlefields is quite easy. Other battle behaviour is more tricky to change in a reasonable way. Changes in kv_rules f.e. can do good, nothing or have unexpected results. Add to this the fact that no TW game ever had functional cavalry warfare. Hopefully some experienced modders take care of ToB.

But ok, I'm still more upset about the unit cards, so ugly, not as much as in TW:R2 but bad enough to not-buy the game, perhaps, I'm still considering. I had to change it myself to avoid unreparable eye cancer.

[Sirlion]

>>3rd of May now !<< click!

I see they have read my youtube comment. Hah. I wish. But it did receive 257 approvals. Good for them, fix the shameful animations and collision detection for once. That gameplay video was borderline 2009 at best (30 fps included)

[LinusLinothorax]

10 Attila features removed from ToB

[Pathstrider]

I feel like there is a discussion to be had on whether these changes are good or not beyond purely presenting it ad "cut content". Probably worth quoting the comment from Jack Lusted (from the Reddit thread):

• Population mechanic - With the move to a cross between Shogun 2 and Rome II's province systems, the only use for population surplus would then be to unlock slots in the province capitals. We had that for a while, but didn't feel like it added anything so now slots in major settlements are unlocked the same as in Shogun 2, by upgrading the main building chain.
• Culture/religion - I've talked about this extensively before. Having a religion system the same as Attila, where you must eradicate other religions doesn't fit the time period and the fact that there was no great spread of Paganism in Britain and Ireland as a result of the Viking invasions and the Vikings that did settle did all eventually convert to Christianity. There are still religious themed buildings in the game, special building chains devoted to various Saints cults as well. You'll see references to religion in army names, and in flavour text in the game as well. Some soldiers will wear Christian crosses round their necks, others Thor's hammer. In the Great Viking Army units you'll sometimes see a mix for example. It's not about no religion in the game, just no mechanic like in Attila as it doesn't fit the history of the era. I know people talk about allowing a 'what-if', but the what-if they're talking about goes completely against the history of the time.
• Edicts - No edicts, it's all about your governors in Thrones. Their starting traits, traits they develop over time, and the followers you choose to put points into give them a range of bonuses that they give when they're a governor. Not having edicts means we can make those bonuses more powerful, so the players choice becomes much more about which governor you put where to get the most out of their bonuses, or where their bonuses are most needed right now. Puts the emphasis on the characters and how they differ, which is a big thing in Thrones.
• Army Traditions - Gone, rolled the bonuses you would've got from there into the bonuses generals can give from traits/followers, and into the bonuses granted from the tech tree. Again focus is on characters and what bonuses each general brings to the table, and having stronger bonuses from the tech tree helps tie in nicely with the unlock mechanic which means where the player chooses to research is giving them stronger bonuses.
• Agents - Been talked about quite a bit. A lot of the bonuses agents brought have been moved to followers such as causing negative public order in enemy provinces, increasing movement distance of armies, causing attrition in enemy armies etc. The garrison chain you can construct in province capitals reduces enemy army movement distance so you can still do that to enemies in your territory, and by default you see clearly into all regions that neighbour your own, so you have a better idea of army movements in the surrounding territory.
• Imperium - The hard caps on number of armies from Imperium has been removed, we generally have gone for soft caps in Thrones, so things like available food and money restrict number of units you can field, and the unit pool caps and replenishment rates factor in as well. We also have mechanics related to characters such as Estates that mean it can be trickier to manage characters loyalty in your faction if you have a lot of them as you have less things to give out to keep them happy. There is still a diplomacy impact for having more Fame (what we term Imperium in Thrones)
• No navies - We had these in early builds of the game but they felt wrong. Having Viking forces that couldn't disembark and were stopped by the coast was weird. How transport armies worked in Rome II and Attila much better reflected how navies were used in this era, so it’s what we've gone for.
• Mercenaries - Incorporated into rosters where this was done, for example the availability of Viking style Foreign Warriors for the Irish factions
• Minor city customisation - Again relates back to the change to the province system and it being a hybrid of Shogun 2's and Rome II's. The minor settlements are villages of set types, much like the regional buildings in Shogun 2 were, however now they can be conquered and owned by others without needing to take the province capital first. Allows for players to really target an enemies food supply or income and wear an enemy down that way. And yes that's still true on higher difficulty levels, the AI does get bonuses to food and income on them but they're % bonuses so if you take out a food region it will still have an impact.
• No forced march/ambush - The movement bonus from forced march now comes from Followers and traits, so whether an army can move a long way depends on who's leading it. We also have a technology that increases movement distance on roads by 50% in your own territory, that means later in the game it is easier to protect your own land. The talk about why no ambush requires a bit more depth so will have to wait, it is Easter weekend and so I'm typing this from home, won't be back into the office until Tuesday.

Personally, I'm supportive of a lot of those (No agents), fine more them as long as they are represented elsewhere (mercenaries built into factions, edicts), and only concerned at a couple - mainly the problem of having too many general only armies running about taking minor settlements.

I'd also be curious to here from people who prefered the Shogun and previous settlement system - the removal of population and minor city customisation sounds more like those, so would you feel it to be a positive change (I personally prefer the Rome 2 onwards system)?

[zoner16]

I feel like there is a discussion to be had on whether these changes are good or not beyond purely presenting it ad "cut content". Probably worth quoting the comment from Jack Lusted (from the Reddit thread):
[/LIST]

Personally, I'm supportive of a lot of those (No agents), fine more them as long as they are represented elsewhere (mercenaries built into factions, edicts), and only concerned at a couple - mainly the problem of having too many general only armies running about taking minor settlements.

I'd also be curious to here from people who prefered the Shogun and previous settlement system - the removal of population and minor city customisation sounds more like those, so would you feel it to be a positive change (I personally prefer the Rome 2 onwards system)?

I personally liked the Shogun 2 province system the best out of the series, because it gave provinces some unique character and helped define your strategy given your start position. This was mainly due to the differences in fertility levels between province farms, the constant need to increase the number of ports you had for trade, and the bonuses and unlocks you got when you controlled certain trade resources.

I wasn't a big fan of the Rome 2 system of huge amounts of minor settlements all over the place that all felt fairly homogenous. Some vestiges of the old system remained, but trade goods were almost all just export fodder (with a couple exceptions), farms produced basically the same amount regardless of region, and all in all the map felt less interesting than the map that existed from Empire-Shogun2. The province system itself always felt too restrictive for me, since everything was controlled at the province level, but settlements within it were still taken one at a time, forcing you to either deal with a bunch of negative effects from your neighbors despite your border still being up, or never settle a war that you don't take an entire province with.

I'm not sure how well this new system is going to scratch my itch. On one hand, it's nice that minor settlements have some diversity and reasons to target specific ones other than just proximity or completing a province. On the other, everything is still taken one settlement at a time, regardless of size, and I'm not sure if this is going to alleviate the old "AI raids all the infrastructure and never attacks" syndrome or make it worse by charging in single generals to take your villages. It's nice that rebels can finally do some real damage right away, making public order more important, and you can't just camp your army far from the front and cheese settlement battles until they return. I like that we can garrison the minor settlements if we don't want the enemy to raid them, which is then represented on the battle map. The Attila governor system is also a welcome addition, which I find far better than edicts. The drive to conquer seems to now be focused on getting more estates for your vassals when its not about money, which is also an interesting shift.

All in all, I'm cautiously optimistic that this might be a good marriage between Shogun 2 and Rome 2's systems. Most of the "removals" from the list sound like refocusing for the time period. Some don't go as far as I'd like, but that's an implementation complaint, not a philosophical one. The big ones are the lack of navies and ambush. The former is mostly my irritation that the Viking navies feel exactly the same as the Anglo Saxon ones. Yes, I get that the Vikings should be able to disembark, raid inland, then jump back onto their ships, and the old naval mechanics don't allow that, but couldn't we at least build some dedicated seaborne troops who get better boats when they cast off? The latter I'll wait to hear their explanation, but it better be a good one.

[LinusLinothorax]

Population mechanic - With the move to a cross between Shogun 2 and Rome II's province systems, the only use for population surplus would then be to unlock slots in the province capitals. We had that for a while, but didn't feel like it added anything so now slots in major settlements are unlocked the same as in Shogun 2, by upgrading the main building chain.

How about a population system the modding community developed? That actually adds depth?

Culture/religion - I've talked about this extensively before. Having a religion system the same as Attila, where you must eradicate other religions doesn't fit the time period and the fact that there was no great spread of Paganism in Britain and Ireland as a result of the Viking invasions and the Vikings that did settle did all eventually convert to Christianity. There are still religious themed buildings in the game, special building chains devoted to various Saints cults as well. You'll see references to religion in army names, and in flavour text in the game as well. Some soldiers will wear Christian crosses round their necks, others Thor's hammer. In the Great Viking Army units you'll sometimes see a mix for example. It's not about no religion in the game, just no mechanic like in Attila as it doesn't fit the history of the era. I know people talk about allowing a 'what-if', but the what-if they're talking about goes completely against the history of the time.

Note how he only justifies religion not being in the game, not culture.

Hope that ambushes will be back. Also wondering why the hell they wouldn't increase the building slots to eight in the main province. It's like one of the most requested things for Rome2/Attila and makes now even more sense, considering the minor provinces only have one or two slots.

[Totalheadache]

Wait no ambushes at all? Or I am reading this wrong?

[Pathstrider]

The bit I'm waiting to see is how much existing aspects have just been moved around and how well they do the same thing - for example, CA have said that agent abilities/army abilities/edicts are all concentrated in leader stats.

Take army traditions for example. I can get from a time period pov there are no standing armies that are around long enough to build traditions the same way the Legions did. And personally, army traditions always felt like another un-needed "+x%" source that wasn't sufficiently different to general stats or items - so I could welcome some centralisation of that. However, looking at Legend's video I am concerned it's just centralising the "one true build" in once place - a look is going to depend on the general's traits I think.

Similar case for culture - from the looks of the Northumbyrie lets play, the culture clash between Anglo Saxons and Vikings is part of the faction mechanics rather than an over-arching mechanic. Why that doesn't apply to the Sea Kings is a really good question, or why not to new lands that other factions conquer?

My overall take is that due to the narrower time period and (I think) 12 turns per year, there are a lot more "constants" in the background - so for example, you can't shift the culture of places in that span of time. I could be ok with that depending on how it's done - if it makes it harder to expand out of your core lands mid game and manage a bigger empire in the long run then I'd welcome it.

How about a population system the modding community developed? That actually adds depth?

Can you point me at that? I haven't seen it.

Wait no ambushes at all? Or I am reading this wrong?

It really doesn't look like it, though sounds like we'll know more next week.

[LinusLinothorax]

Can you point me at that? I haven't seen it.

Like this.

[NostalgiaFan]

Not having Culture or religion is ing stupid. The reason for taking it away is even more stupid, "no one turned Pagan in this time so lets not bother" I thought one of TW's selling points was forging history your own way CA? Why not give us the option to bring paganism back to the British isles? Why not add the depth of keeping control of your people's faith? Why not let minor religions like Celtic Paganism or Celtic Christianity become more prominent?

Just sounds stupid and will take away a lot of depth from the campaign. And Culture has even less reason to be removed, there were plenty of involving clash of cultures in the isle at this time from Norse, Saxon, and Celts. Getting rid of that is just CA not wanting to put any more work into something because of laziness.

[Anna_Gein]

Wow the game look so barebone ! It sucks.

Ambush were great in Rome II and Attila

Sad to see CA nonsensical cut content at it again.

[zoner16]

https://www.totalwar.com/blog/thrones-rosters-reveal

Rosters are out.

[geala]

I mostly don't like the removal of army traditions. Maybe it does not fit the time period, but you could also argue that it did not fit most factions in other time periods, too. Too much focus onto generals for me.

I like the DeI system where agents are turned into useful entities. It is nice to have several points where you can develop, not only one. It feels good for example in case I have to use an unexperienced general that I can at least give him experienced troops and advisers. It also increases the feelings to special units and armies.

Not that any real army ever thought about traditions being helpful for morale ...

[zoner16]

Answer on ambushes from Jack:

In the discussion threads that popped up about Legends recent video on Thrones, and on the comments he made on a stream, I replied to many of the concerns raised and explained the thinking behind many of the changes we’ve made. The one exception there was ambushes, where I said an answer would have to wait until I was back in the office. Now I am, so here’s an answer, it just had to wait as my time was limited over the weekend and this is a fairly in-depth answer to write. Plus, I wanted to talk about how we use some of the data that’s available about how people play our games and so needed to make sure my numbers were correct.

Now, before I delve into the detail I feel it’s worth talking again about the way we have approached the design for Thrones. The aim with every Total War game we make is for it to have the right amount of features in it to make it feel and play as a complete whole. Sometimes that will involve a lot of overlap with previous titles, in other cases there will be more differences. For Thrones the design direction has very much been one of greater focus on consolidating the various sources of effects into fewer, but more meaningful/impactful areas. We set out to deliver the same amount of gameplay depth as with any TW game, but with the focus of what a player spends their time on from turn to turn shifted towards the new mechanics in the game. There’s more emphasis on the culture/faction mechanics and choices around those and the narrative events for each faction, as well as on characters who are a key part of the game. There isn’t less to do each turn, the focus is simply different from what it is in say Attila or Warhammer.

A few people made comments about why other people who have had early access to the game hadn’t talked about features that have been ‘removed’. My hope is that what is in Thrones feels like a complete experience, that nothing feels missing from it.

Ambushes, and their absence from Thrones, is perhaps a good example of that. With Thrones being based on the Attila codebase, the way to keep ambushes would be to have it as a distinct stance as it was in Attila, with armies being unable to move in it. The way it works in Warhammer would have been tough and extremely time-consuming to implement. It wasn’t a viable option. So, if we kept ambushes they would be in the game in a limited way. The next step is to take a look at the gameplay data we have available and see just how often ambush battles took place in Attila. Whilst keeping features that existed in Attila can be fairly straightforward, it varies a lot and some elements require more work than you might expect. We had to factor this in to make informed choices about where to invest our time in developing Thrones.

Now, I know this won’t come as much consolation for the people who made use of ambush and considered it to be an important tool, but the data from how people played Attila doesn’t really support that feeling in most players. Ambush battles were only 0.05% of battles fought in campaign in Attila. Not 5%, not 0.5%, 0.05%. There were over 1,750 other battles fought for every ambush battle in Attila. Judging by the statistics a majority of the Attila player base never fought a single ambush battle.

That definitely made us think about whether it was worth keeping them, given the effort to maintain them in Thrones versus putting that work into other parts of the game that people will definitely get to experience. The next stop for us was looking at the history of the era, to see if ambushes were common.

Most battles from this era are only known from brief references from annals of the time, but for a few there is more detailed information: Edington (878), Brunanburh (937), Maldon (991), Clontarf (1014), Fulford (1066), and Hastings (1066). None of these battles are ambushes, they’re all conflicts fought between forces who are definitely aware of the others position. I’m not suggesting that ambushes did not occur at all, just that the historical records we have don’t indicate that they were a massive feature of battles in this era.

Then we considered the other campaign map changes we’ve made, and how they might affect the likeliness of ambush battles. For example, we’ve incorporated the movement speed bonuses that, in Attila, were gained from a forced march stance into traits, followers and certain technologies. This means armies won’t be moving around in a stance that ambush sort of counters. We’ve also incorporated the movement-distance uncertainty of the AI from Warhammer so that its army movement is less precise, and the buildings/followers that reduce enemy movement distance so there are more ways for the player to make sure they catch their enemy in open battle.
So with the data, and considering the history and other changes, we made the choice to take the time that would be put into ambushes and put it into working on normal land battles, improving the look of battlefields and the balancing of them, as we know players fight lots of them. This way we’re making sure more players get to experience the benefits of that effort.

This doesn’t mean that ambushes are out of Total War and never coming back - the focus of some races in Warhammer around them shows that. We will always consider what’s the best for each game and also look at why so few people are playing them. That’s never going to have a simple answer. For those of you who do play ambush battles, we’d like to know what you love and what you loathe about them.

I know not everyone will agree with this change, but again I hope that explaining the rationale behind our decision shows this is not some thoughtless change. Every change for Thrones has had the same level of thought put into it. We want to deliver a game that people play for hours and hours and that they enjoy every minute of, and we believe that the features we’ve chosen and the changes we’ve made will make sure it does. We hope you’ll feel the same when you get to play the game.

I don't like this rationale, and I don't think the rationale makes the cut anyways. The gist of it is that the dev time saved on not porting over ambushes from Attila was better spent on other parts of the game, which is justified by the fact that only 0.5% of battles from Attila were ambushes.

The first issue is that the time it would have taken to port an existing feature of the engine is incredibly unlikely to have been larger than creating a new feature that has as much impact on gameplay. I obviously can't back this up with actual numbers since I haven't seen the code, but from my own intuition, I'm guessing that the time saved was not used to make a replacement and just allocated to ease the time windows on other projects.

The other is that the reason that ambushes in Attila were very hard to pull of due to the way stances worked and how easy it was to bypass them once detected. In the pre-Rome2 system, ambushes only required moving into terrain that was suited to ambushing, and didn't require stances or movement points. Add that to how easy it was to just walk right around them due to lack of good map chokepoints made ambushes a rare beasts. However, that doesn't mean that they were unimpactful. Often times, ambushes became the highlights of a campaign because they created some amazing turnabouts, which is why you really only needed a couple spread out over a ton of battles to make some great memories.

What I'm suspecting is the real reason is that they decided to cut ambushes because they would play poorly with their implementation of shieldwalls and their intended battle pace. Keeping them in would likely exacerbate a bunch of the underlying issues with their solution to sheildwall combat, and they decided that weathering the storm over removing ambush battles was better than letting poorly done ambush battles out in the wild to color opinions about all ThroB battles (much like their reason for removing river crossing battles in Warhammer 1).

[Pathstrider]

I think the numbers say more about how bad the Ambush stance from Atillia was, but I can sort of see that the effort to make it like Warhammer wouldn't have been worth it.

[Anna_Gein]

What a stupid answer !

Gosh I am so angry about ambushes. Ambushes were like 1/4 if not 1/3 of my battles in Atilla. Once in Attila ambushes were are far as 50 % of my battles as the Western Roman Empire. Back then I resorted to a guerrilla strategy of ambushes and defensive siege to hold against line against barbarians and steppe factions.

I do not care if it takes the devs time to implement the feature. Seriously I am so antipathetic when devs are crafting excuses for cut content. As a customer I did not forced CA hand to produce 2 stand alone titles and 2 majors dlc per year.

Also what a stupid way to justify cut content. Statistics ? TW are sandbox game. Player should be free to play as they like. Some like myself like to use ruthlessly all available tactics to win. Other players may find ambushes unworthy and dishonorable. Some others may lack sufficient skills to lay ambushes. All of this is normal and should not lead to cutting features. If CA wants to develop games where all features are used, they should just make some linear action games like CoD.

What I'm suspecting is the real reason is that they decided to cut ambushes because they would play poorly with their implementation of shieldwalls and their intended battle pace. Keeping them in would likely exacerbate a bunch of the underlying issues with their solution to sheildwall combat, and they decided that weathering the storm over removing ambush battles was better than letting poorly done ambush battles out in the wild to color opinions about all ThroB battles (much like their reason for removing river crossing battles in Warhammer 1).

I think the removal of ambushes battles is just so save development time and release their game earlier. From what I have seen, there is no shield-walls in the game. CA just reduced the unit spacing in Thrones of Britannia. It takes like 10 seconds to change the variable on Rome II and Attila ...