Topic: 'Basic outpost' - low wares start

For some time I've been saying that I'd like to have a starting condition like 'struggling outpost', but a bit faster to get started.

• I like the idea of building up through the layers of the economy, starting with the basics and working towards more advanced buildings. So for that reason, I've been playing on 'struggling outpost'.
• But with 'struggling outpost', it takes a long time to get established, and a lot of that time is just spent waiting for things to happen... eg. waiting for enough trees to grow and get chopped so that you can build the next building.
• I'd like a low resources start that is similar to struggling outpost, but with less waiting in the early game.

Ok... so I've talked about this a few times, but there hasn't been a lot of interest. And recently, in an unrelated thread, hessenfarmer was twisting my arm to do something about it. So fine. I've had a shot at it. I've spent some hours today tweaking numbers, using 'struggling outpost' as a starting point, and doing a very small amount of testing to see if it is roughly right.

I decided that it was probably best to make it possible to get new tools almost immediately. That does short-cut some of the build-up that I enjoy; but I think it makes for more interesting early-game decisions - and helps sets the condition apart from 'struggling outpost'. So basically this starting condition is like struggling outpost, but with enough extra stuff that you can get a few extra tools fairly quickly. So you can choose between having some additional woodcutter, or builders, or farmers, or whatever you think would help you most in your early-game situation. It just means there is a lot less waiting around.

In any case, my first-draft is uploaded and ready to try if you are interested. And if you have any balance feedback, please let me know. The current attempt is only a rough guess at what is fair; and I'm not an experienced PvP player. The add-on is called 'Basic outpost'. (Currently unverified, because I only just uploaded it.)

Top

Quote

I've been testing this starting condition by playing short 1v1 matches vs the AI on "Crater"; me with 'basic outpost', and the AI with standard 'headquarters'. Obviously the size and style of different maps affects the balance of tribes, so testing on the same small map isn't great - but playing different tribes on the same map just helps me to get a sense of comparison.

I'm not finished testing yet, but I'm already confident that Empire is relatively weak in the version I uploaded. I'm going to give them an extra marble column; and I'm wondering if they might still need something else... perhaps a single piece of cloth so that they can build a barracks without having to get a sheep farm and weaving mill. But I'm not sure. I find the balance of Empire to be a bit weird. Other tribes can build up their economy quite a lot while they search or get ready for the mines that they need... but empire basically just has to beeline a marble mine before they can think about building farms etc. (And they also need quite a bit of marble to build the vineyard and winery to support the marble mine!)

I'm thinking that maybe if I increase every tribe's starting soldiers from 6 to 10, or whatever, that might help Empire. Obviously every tribe would be stronger, but maybe it would be of particular benefit to Empire because their colosseum basic training and their quick-build blockhouses for exploration are two of their tribe strengths - both of which kind of need a few extra soldiers. On the other hand, I didn't really want the initial soldiers to be a strong enough army to fight with. They're really just meant for territory expansion; and getting a functioning barracks is meant to be part of the early-game strategy. So I'm not sure what's best.

Anyway, that's the kind of thing I'm thinking about. Empire definitely will get that extra marble column to start; and I'll might tweak a few other things here and there. But I assume that the add-on will need to get re-verified every time I upload a new version, so I'll just sit on these changes for awhile until I've tested a bit more (and perhaps even gotten feedback from someone).

Top

Quote

I was talking about my low wares starting condition. After playing as each of the tribes in a 1v1 against the AI on a small map (me with 'basic outpost' vs the AI with full headquarters), I'd decided that Empire was slower to get started compared to the others (with 'basic outpost' wares). So I've changed it to give them one extra marble column, which I think helps a lot.

But yeah, I was also just pondering and talking generally about the empire economy - thinking about how waiting for a couple of pieces of marble from a quarry can be a long bottle-neck, but then once the mine is operating marble is fairly plentiful. In the map I was playing on, there's a marble mine site very close the headquarters. But I recall that in many of the random maps I've played, sometimes I don't see stone mines for a long time. Other tribes are generally fine with that, even if they need the stone mine, because they can still build up the rest of the economy while I keep looking. But empire can't. That's why I was imagining long chains of blockhouses stretching out like tendrils searching for marble and wondering if a few extra soldiers (for all tribes) would help.

As for the AI using my starting conditions - I've tested it and they can't do it. I thought maybe they'd be able to handle it, because unlike 'struggling outpost', there's a fair bit of flexibility in what order you can build things with 'basic outpost'. So if you just build things that you are able to build then you probably won't get deadlocked... but the AI doesn't do that. I set up a 4 player AI game for me to spectate, with each AI on a different tribe; and they all failed for the same reason: They set up heaps of build sites, and ended up with their builders stuck on sites with no enough resources to complete. When they started out, I thought they were going to be ok - because they had the woodcutter, and the forester, and sawmill (or equivalent) - but their hyper-ambitious expansion still ruined them. Also, on the map they played on, the stone was just barely outside of their starting territory - and for a couple of the AI that was enough for them to screw-up the placement the quarry and run out of stone.

So, I think they AI would need two things to get a successful start with the low wares conditions. Firstly, it really needs the stone to be in the starting territory, so that they definitely get a decent amount of stone when they build their first quarry. (Most maps have that.) And secondly, the AI needs to be a bit more careful with its build sites. I've got two possible suggestions:

1. the number of build sites the AI is willing to create should be capped at number of builders+1. So if they start with two builders, they should only try to have at most 3 active build sites. When they get the tools to get more builders, this will automatically recruit the builders and the number of build sites will increase. But until then, keeping the number of build sites low will help prevent them from locking up all their resources in sites that are just waiting for a builder to arrive.
2. The thing that actually deadlocks the AI is when they try to build something they don't have enough resources to build... But choosing to start building something that you are still waiting gathering resources for is not necessarily a bad choice. It can help get the building partially built while the new resources start coming in. However that only helps if the builder is actually working on the site. There is no benefit whatsoever to starting a build site where you don't have enough resource and don't even have a builder to work on it. So my suggestion is that if the AI wants to create a build site for something that they don't currently have enough wares to complete, they should only be allowed to do this if they have an idle builder. If they don't have an idle builder, and they don't have enough wares to fully stock the build site, then they should just wait.

I reckon one or other or both of those AI suggestions would probably help the AI in general - but especially in low wares situations. With that, they could probably use my new starting conditions. In any case, the AI as it is currently can't successfully start with the new conditions. So I'll add the usually warning to description to say so in the next version.

Incidentally, the description for the standard headquarters should probably be changed in the base game. It currently says "start the game with just your headquarters"; which is a bit unclear in the context of the other descriptions. It should probably say something like "start the game with a headquarters, well stocked with wares and enough workers for a full economy."

Top

Quote

I don't really feel like I'm mixing topics. I'm just trying to contribute to conversation about the game. You mentioned the AI, and so I responded with my thoughts about the AI in the context of low-resource starts. That's all. I'm not trying to make a formal feature proposal about the AI; I'm just speculating about possible ideas to reduce the chance that the AI will fail in low-resource situations. But sure, when we get a flood of posts about AI ideas all mixed in with the posts about low-resource balance, that could make it hard to follow the conversation. We'll definitely be wanting a new thread when that happens.

Top

Quote

@kaputtnik: yes currently the AI is only able to start from starting conditions with enough wares like HQ, trading outpost or sometimes fortified village:

However the suggestions from @blind3rdeye are very good not only to help with lower wares (although I doubt it to be working with basic outpost) but to make the early game more efficient.
So I'll open an issue and try to find some time to implement it.

Top

Quote