Sawmills

[Nelphine]

For me, the original goal was that the AI were fine with cost of sawmills, and the human wasn't. Therefore, the AI needs no change, but the human does.

Therefore, the extra lairs solve the problem. Giving the AI any compensation (in the form of increased gold, or extra buildings), defeats the purpose of giving the human extra. 

Therefore ai start should not change.

Changing the sawmills (and the associated free city bonuses and housing) in a manner that means the treasure from the lair is not needed for sawmills (or the other basic buildings, but it should be as obviously important as sawmills were) defeats the purpose of adding the lairs. If we make that change, we should instead not add the lairs, and should instead reduce the starting ai gold since they no longer need it for sawmills.


Additional conclusions:
Adding the lairs would be FUN and greatly enhance gameplay. Therefore it should be done.

This means any sawmill or associated change should be done in such a fadhion that we preserve the balance reason for adding the lairs. Therefore any change to sawmills or associated, should still be done in a way that we end up needing to spend the gold to become properly economically functional - this is both why the lairs exist and why the ai starts with so much gold.

Additionally, changing the sawmill and associated will result in more engaging gameplay. Therefore this should be done . But it should not result in the reasonable choice of skipping the sawmill AND basic buildings, and in fact it should still be done in such a way that gold should be spent on at least 1 (but probably more) of those 4 buildings as fast as possible.

[Seravy]

You are trying to have basic buildings that are not required to buy, and basic buildings that are required to buy at the same time. You're contradicting yourself. There is no such thing.

I agree with your conclusions, but having sawmills act as a required gold sink, and having sawmills not act like a required gold sink are mutually exclusive. So we have to pick one or the other and we already agreed we pick the "not required" path.

(Note that "not required" still means it's worth buying. It just doesn't put you into a massive disadvantage if you cannot. So it's not acting like a gold sink and won't eat up the lair gold - you'll still want to spend the gold on something, but it will be a choice of economy or military. "Required" would imply you can't spend it on military and need to spend it on sawmills (or lose the game instantly if you do spend it on military and fail to invade fast enough - like the "barracks first" Warcraft strategy.))

Summary, we can only pick ONE :
-New sawmill and housing, lairs, player has extra gold they can decide to spend any way they want.
-New sawmill and housing, no lairs.
-Lairs, old sawmill and housing, lairs pay for sawmill costs, so player has no extra gold to spend.

[Nelphine]

Right, and I've always stayed on the stance that we should have required buildings.

What I agreed to is that they don't need to be AS required. In other words, the sawmill can change to the level of other basic buildings, and housing should stay better than building for pop 1 or maybe 2; and housing with builders hall better than 3 or maybe 4.

And I believe the other basic buildings are important enough that if you don't build some/all of them before going military, you WILL lose in the long run, if against 3+ AI. 

So I think those buildings already meet that level of requirement, and toning down the sawmill to that level as suggested will also meet that requirement.

That's why I agreed to the partial mix, where the city does a little (1 food, 2 hammers), the sawmill is reduced just a little.

I don't think the sawmill cost should be hugely reduced. Maybe to 75 or 80 but not lower. It's still important, and can still use up hold.


Just doing all that, I believe the potential for spending the gold on military already is risky. I don't think you can afford to do so in the long run. The main thing is to curb housing. Your proposed formula means you CAN do housing and it roughly matches sawmill (and therefore other basic buildings). That means you COULD save the gold for military, which I think is bad. 


So yes, I think we can reduce the swmill from being super required, to simply being as good as the other basic buildings - which are already good enough I would call them required compared to saving gold for military.

Since those other buildings are already required, I think adding the lairs will mean the new treasure will be spent on a combination of sawmills and other basic buildings, which means it won't be spent on military - it will simply let the human get its first few cities running faster to match what the AI already does.

Therefore, the only thing left is to tone down housing to be sure it doesn't provide so much that you can save gold by being just as good as building a sawmill for too long.

[Seravy]

AS required

"Required" is binary though.

In other words, the sawmill can change to the level of other basic buildings,

But this is impossible if it's "Required".
As I already explained about a million times the sawmill is required because the city doesn't have enough hammers to build anything without it. Other buildings don't produce hammers, so they can't possibly be "required" in the first place. It doesn't matter if you wait 5 turns for the marketplace to build itself, or buy it for gold. Same for any building, except the sawmill. And since local production is relevant, marketplaces are inherently inferior because they are designed to only be better if you don't need the gold locally. So they can't compete in this regard. (other basic buildings aren't convertible to production so those even less so.)

And I believe the other basic buildings are important enough that if you don't build some/all of them before going military,

Keyword "BUILD". You can afford to BUILD them, you don't have to BUY them. This wasn't the case with the sawmill and it is as explained above, the only building that can possibly have this trait (unless we break some very fundamental design elements such as gold/hammer conversions - we obviously don't want that.)

I still think your are confused on what is the definition of a "required buy". A required buy has no opportunity cost - you'd get back the investment plus interest before you could have finished building that particular thing without buying. None of the buildings are like that except the old sawmill. I'm a bit tired of repeating myself.

We can start a debate on whether it's important enough to spend X amount of gold on basic buildings before going military, but that's entirely subjective and ideally, the answer should be "sometimes". However based on your statement that you'd NEVER buy a marketplace, you already said you'd never buy a basic building so they obviously aren't required even according to you.

Look, unless you show me actual calculation errors in my housing post, anything you say about toning down housing will just sound like nonsense to me. I did go out of my way to calculate how much housing is NECESSARY for the sawmill to be able to compete with other buildings instead of being required first buy. There is no other way, if there is one, show me numbers.

Numbers that meet all the criteria we have : Sawmill ROI of 15-17, Sawmill turns to build much less than ROI (preferably half or below), total base+sawmill hammers =8, housing ROI<15 for as much population as needed to have the amount of hammers that result in the intended "sawmill turns to build" on all average or better cities.

Right now I sure you want something that's mathematically impossible. (and yes, I also think that housing might be a bit too strong but hey, if math says we can't have less, I have to accept that.)

[Seravy]

btw I'll start working on this on Saturday (roughly 36-40 hours from now ideally) so we need to decide before then.

[zitro1987]

My last remaining thought (and may be a really poor suggestion) is that trade goods and housing are combined into one: 'housing and trading'

Farmers contribute to 'growth'
Production contributes to 'gold'

An added bonus is that this opens up a new building slot.

[Nelphine]

Required means 'to remain competitive'. Despite the marketplace never being a good choice, that doesn't mean the others aren't. That's why I said at least 1. 

However, I think you are correct. I'm using required, when discussing at least expert, and possibly even higher, where you would need to buy some of those buildings or the AI would get too far ahead.

So you may be right that my definition isn't helpful. (You may also very well be right that what I want is mathematically impossible.)

As to exact formula.. I'll see if I can come up with anything tonight. But, I feel like I'm more likely to disagree with your conditions than come up with a better formula than you have. You've always proven to be very solid on that.

[Nelphine]

Is growth formula same as outpost growth formula? Meaning a pop 25 location with no ores or race would be growth 50? That seems low.

Neve mind find it on the MoM wiki. So 25 max pop would have 12.5*10 growth at pop 1.

[Seravy]

That, plus the racial modifier (average races are +6 = +60 people)

[Nelphine]

Oh. I didn't realize +6 was average. Baloogas. Nope, doesn't seem to cause any issues.

Ok, the formula I would suggest is:

Sawmill provides 6 production and only costs 70, city is 1 free food and 2 free hammers. Sawmill ROI is 17 turns. Sawmill build time at pop 1 is also 17 turns; at pop 2 its 14 turns; at pop 3 its 10 turns.

Housing formula is:
(400+10*max pop)/(current pop*current pop) + minimum of (production OR current pop *[2+(max pop-current pop)/(max pop)])*roundup[100*(max pop - current pop)/(max pop)]/2

It's not quite right, as population 4 and particularly population 5 is a little too high with that housing formula.

It also doesn't peak quite high enough when current population is equal to 50% of maximum population - ideally it would peak closer to 60%, but I figure this is already probably too complicated and you won't be able to use it.

So realistically, I'd want to add in an extra multiplier after the /2 at the end, to make the peak higher (in other words, I want to stretch out the second portion of the formula, which would somewhat alieviate population 4 and 5 as well, because that would be below the median for the second portion, so the stretch would make it lower); and I'd want to add in a shift to the right by about 10% for the second portion (which would also help population 4 and 5, while shifting the peak to the desired 60%). Given builders hall, the stretch would actually not be very much. (Maybe the peak goes up by 10%, the bottom goes down by 10%, and then change everything in between to match the new curve.)

If you have remotely enough space for this (which I doubt), then I'll work out those tweaks; otherwise, well, again, we can FIND the formula to do what I want, but that doesn't mean it's actually doable due to coding restrictions.


Result of this formula: population 1, do housing. It would be amazing. Population 2, do housing. Population 3: if on particularly high max pop, or if you have a builders hall, do housing; but usually, you'd start building here. Population 4: if you have extra food (for instance from wild game), do housing; otherwise, do the sawmill (even if you have high max pop or builders hall, you'll still want to start actually building at this point). Generally, for an average city, you'd end up doing housing to population 3 (takes 3-5 turns), then sawmill in 10 turns, saving you 2-4 turns over just building it from population 1.

Secondary result of this formula: When your population gets around 50%, you're actually gaining a rather high amount of population with housing. With builders hall, this could actually be made to be high enough to overcome trade goods (and would be shifted closer to 60%). Therefore, on large cities (more than 20 max pop usually), when you finish all buildings, if you are NOT building military, housing could be worth it for a time; but at a later point, (around 80-90%), it would slow down enough that trade goods would become better again. Both would be close enough that it would only matter for people who REALLY worry about every little bit, making it really only useful to even bother with at master or above. Could put the calculation into the housing and trade goods display? No idea if you can do that kind of thing in a display. Probably not.
Either way, it would make it so that leaving anything on housing accidentally would generally not be a loss compared to choosing trade goods, until you got up very high (90%). And choosing trade goods too early wouldn't be an issue either.