Sawmills

[Seravy]

Okay, they might have it once in a blue moon if everything is perfect but even then, they won't be able to hit all of the lairs.
I've run a test and none of the 4 AI managed to push out any stack capable of attacking the lair even on turn 8.

Actually, no they won't. The AI sends out units in stacks of 2 from the capital. That's too small and has a stack size penalty to the strategic strength, so for attacking it would only count as good as 1 such creature. That won't be able to attack most of the lairs, and would need to first move to the stackbuilding spot and get more units.

I don't think the turn limit is necessary but I'm willing to add it if, during actual games, you see the AI take your easy lairs more often than acceptable.

I don't agree that changing sawmill is a step in the right direction. I believe it does its role correctly right now.

Why?
-It's not newbie friendly
-It's not interesting gameplay
-It's not providing a relevant type of advantage to the AI (they would still buy the sawmill and get same hammers even if we change it, so AI production remains unchanged), or disadvantage to the human that result in prevented abuse (production in the hamlet can't be used towards accelerating high tier units in the capital efficiently. Yes, you could build a marketplace from it but it would still take you 20 turns...)
-You agree that the extra 200 cost is not necessary as cities are overpriced compared to military - you just argue the 200 difference wouldn't be enough to matter. So it's still not an argument for keeping it the same, nor changing it. It's an argument for it being irrelevant.
-You don't autolose if you go for a non-sawmill strategy and it doesn't work out as well as you expected

I couldn't really find any argument for keeping it unchanged, but we have several for changing it.
(yes I would like it if we didn't change anything, less work, less risk, less testing needed, but I can't find a reason for doing that. Except for maybe flavor - basic production is a bit weird, never really liked the idea. But I can deal with it, if it improves the game.)

[Nelphine]

It splits the cost of a new city out into 2 pieces. (This is good gameplay on its own. The military vs settler cost is a problem, but isn't directly related to this gameplay even if it's directly related to sawmills.)
It provides a straightforward thing for players to learn. Yes, it's a newborn trap to have a required first building, but it's such an advantage that it is something new players can learn relatively easily, and which really indicates there are buildings that are flat out better - which is important since there are other clearly superior building choices, which aren't as straightforward. By having the sawmill as fairly obvious it forces the new player to examine how good buildings are later on.

It keeps the entire production system very transparent, making it easier for players to learn it and use it. This isn't just flavor, it's also important for engaging new players.

It makes it so obvious a choice for the first building, that when you realize for some cities there are better choices, you get a nice 'aha' moment for learning that.

[Seravy]

that when you realize for some cities there are better choices,

Are there? I can't name a single building (unless you're swimming in gold and can afford to buy the sawmill plus other things. Then you are free to do it in whatever order you like...)

Sawmills are strictly superior because they earn back their cost faster than it would take to even build one without buying. No other building does that, so there can't be better choices. You also get your resource in hammers, which will build you any other building you want, so ultimately it's accelerating any type of resource you might need. It's literally as good as getting twice your gold back without any cost, not even an opportunity cost (you earn it back before you could have even built one without buying.)

It keeps the entire production system very transparent, making it easier for players to learn it and use it. This isn't just flavor, it's also important for engaging new players.

That's nice but I don't see how "sawmill produces 6 hammers, city produces 3" is less transparent than "sawmill produces 8".

It's true that having the required sawmill is easy to learn.
But there is a catch. Learning that isn't making you a better player. It simply enables you to play the game - as long as you don't learn it, you just lose. It's no different from learning that you can produce units by clicking on the left side of the production window, or cast spells by clicking the Spell button. Those are easy too, but you can't win or enjoy the game until you know them.
...and compared to those things, "sawmill first" is way less trivial but the consequences are almost as severe.

Learning which to build first, a sawmill, library, marketplace or magic market is harder - but that actually contributes towards playing better.

Oh, well, not changing it is more convenient for me, although I don't think it's ideal. The free lairs already solve the core problem, so this would have been "extra".

[zitro1987]

I like the idea of guaranteed super easy lair next to player's capital A LOT. It could benefit those early strong military races (vs those economic races) like gnolls and early offensive play - contributing to both fun and balance.

I also like adding a couple of production points for every city, so at least less experienced players not going for buying sawmills at least can get something built over 10 turns or so.

[Nelphine]

Sawmills can't contribute to skill production when you have already minimized mana production. Hammers can only convert to hold, and then to mana.

Therefore, anything that produces skill in a city that is not being used for military production, ends up being better IF by buying it instead, you do not slow down overall skill gain.

This generally means that amp towers are better to buy first than sawmills (my late game hamelts all have amp towers and no sawmill). All other power buildings are also better, as long as that won't prevent you from purchasing an amp tower at another city, which usually means no, purchase the sawmill second.

So yes, it's only when swimming in gold, but still, that's an important aha moment, as habit by that point is to just do sawmill first.

[Seravy]

Trying some math for the "basic production" idea.

Ideally we want sawmills to have better ROI than marketplaces. Hammers are city bound while gold is not.
Marketplaces cost 160g, and produce 8, so ROI is 20 turns.

For sawmills we want 10<ROI<20. Let's call this R. (this ROI gets better if you have a high production bonus so we can't have it start too low, probably would need around 15-18.)

So we want for Sawmills, cost = R*(produced hammers-gold maintenance/2)

We also want sawmills to take less time to build than the ROI, preferably significantly less.

So assuming the city produces H hammers, we want cost<<R*H, but cost = the above.
So we get that we want "H>>(produced hammers-gold maintenance/2)".

In other words we want the city to produce (significantly) more hammers by default than the sawmill would be producing (after considering maintenance). For example if we make cities produce 5 hammers, we would need a sawmill that is a net +3 hammers (for example 5 hammers and 4 maintenance). This would cost 45 to build.

5+ hammers in a city is fairly doable though I rather not give out that much in raw hammers, but we have tricks. First of all you already have 1 hammer by default from your farmer. If we give a food, then the first unit of population can be a worker, so you get two hammers. If we give 2 food, we can have 4 hammers, assuming housing is better than building things until at least pop 2 in all cities. In fact, if we give only food, and make sure housing is always "best" until pop 3-4, then we don't need to give out hammers at all. It also doesn't result in "mandatory housing first" - if you can afford to pay money for stuff then buildings will still be your best choice, housing would only be best if you aren't buying. 

What I don't like in that is, building housing will ultimately mean delay, so while you get the same production rate, it happens slower than currently, even if you do have the money to buy the sawmill. So maybe giving hammers is better afterall. I also don't like the weak sawmill output although we can push that up through higher maintenance, but higher maintenance means you end up with "mandatory marketplace" to be able to afford it. (actually, since the player is left with unused gold, they could buy something else to make up for that loss of speed so that could be ideal?)

Pretty much, city hammer production>sawmill output is the result we got and I don't like that result at all. That would either require too weak sawmills or too high base production. I would want to keep city hammers<sawmill output, but math says that's impossible.

So I guess we found the first big problem with this idea - it would require far too much hammers, and that much hammers could accelerate new city production way too much.
(albeit, considering we agree that new cities turn productive too slowly for the military-settler balance, this could be ideal. Problem is, this much extra hammers would be a noticeable boost to overall game pacing, speeding it up even further, and we probably don't want that. Or not? 5+3 are ultimately still the same 8 hammers, you just get the "5" part regardless of having 200 gold to invest. So ultimately, it's not speeding up anything, except leaving those 200 golds in the player's treasury - which still does speed up things somewhat, but not that far.)

At this point I have to conclude this is ultimately a decision based on our preference - I'm 100% sure the 5 base hammers and weaker sawmill would result in more enjoyable, more diverse, but also easier gameplay for experts (player is left with more gold to use freely, so they can get a lot more advantages though correct play and snowballing) while leaving it as is results in less interesting, more repetitive but harder gameplay.

...that totally sounds like leaving it this way is bad. I mean hard is good but not at that cost... Are we too biased towards the powerful sawmill? Part of me doesn't want to lose that but it doesn't seem to be good for the game no matter how I look at it...

Okay, trying to get some good looking numbers...
3 free food per city
Housing to turn pop 1 into pop 3 quickly, probably in less than 6 turns (3 workers = 6 hammers)
Sawmill producing 6 hammers for a maintenance of 4 at a cost of 60.

...no I don't like this one, that extra food would be used to feed armies instead.

5 free hammers per city (total of 6 hammers with 1 farmer)
Sawmill producing 6 hammers for a maintenance of 4 at a cost of 60. (total of 4 hammer output)

...this doesn't sound bad actually, but does leave the city with 3 extra hammers overall. (6+5 instead of 8) I also don't like the flavor - 5 hammers is a bit too much to get away with "yeah, people go out and gather natural resources". Everyone is ultimately also left with 200 extra gold/city, which isn't getting compensated by extra wait time on housing. Eh.

[Nelphine]

Why not a mix? 1 free food (I quite like the free food idea), 2 free hammers. Make housing better, but only up to pop 2; builders hall make it better up to 3 or 4. Drop sawmill to 6 production.

I like what sawmill does, but I have been convinced you're right, it would be fine 'only' being as good as the other 'build first' buildings.

And while my suggestion doesn't quite meet the required math, it comes close, ish.

[zitro1987]

For calculating ROI for sawmill, should we consider what the typical terrain prod bonus is? For example sawmill with 50% terrain produces 12 hammers, not 8.

I would imagine the typical terrain will be in the 15 to 40 range, so around 25-30%

[Seravy]

Since we want the ROI to be in the range of 10-20 in most cases, we are safe if with pick a number high enough in that range and ignore the terrain effect. (Yes, 50+% bonus would result in a 15 ROI going down below 10, but that's ok. Makes sense that a production building is strictly better than a marketplace there.)

A mixture huh. I'm wary of using food - it opens up the possibility of using that food towards unit maintenance, it also makes famine, unrest and city curses weaker (would need more of them to eliminate all the extra food before city starts to shrink) so I think the cons outweight the pros here (opening up the possibility of using the first pop as worker.). It also indirectly makes Granary a much weaker building (you already have the extra worker do building it won't free up any, unless the city is of a much larger size).
However without food, housing will not be effective at generating hammers at all.

So I don't think we have any other option than adding the hammers directly. At best maybe we can get away with 1 extra food? Doing that would however, only result in 1 extra hammer, which isn't really worth splitting it up. (We either need 5, or 4 but then sawmill needs to be 1 weaker, and thus cost 45)

Housing would need to change in a way that the effect is worth more than the 6 (pop 1) or 8 (pop 2) or 8 (pop 3) hammers the city produces, at least on some of the map locations (and preferably not all of them). I don't remember if we had any formula on how hammers convert to population growth?
At the very least I remember wasting the 8-10 hammers from your sawmill was literally never worth it for housing, not even on pop 25 sports with stream of life and whatever else.
So this might not actually have an existing solution.

[Nelphine]

Right, that's why I was suggesting only 1 free food. I also wasn't going to get ROI into the 10-20 range for sawmill - it would end up higher. (What is ROI right now? Assuming, what, 20-25% production from terrain?)