Starting location

[Seravy]

There was the discussion about the starting locations in the Myrran thread.

I had some time to think about it, if we want something to be done the cleanest solution would be this :

Create a sum of "value", adding together all tiles adjacent to the location. Then check if the location is in a predefined range, instead of checking for "max pop>8", if not, reroll. This filters out both too poor and too good starts, and still keeps enough random elements to be fun - you might get good ores OR a high pop spot, OR a high production spot, but not all 3 in one.


Continent size should be ignored for now, and probably permanently - access to naval travel is easy enough and this is way too subjective to ever come up with a solution for it - only the tiles belonging to the city itself should be counted.

Defining the "value" of each tile, and the range will be the hard, and subjective part, especially on the ores.

Non-ore tile values are fairly easy to rate I believe :
Ocean, Tundra, Desert : 0 (produces nothing or 3% production only)
Swamp, Hills, Forest, Shore, Mountain, Chaos Node : 1 (Low food, or high production without food - no need to go into much detail, I believe a 3% production bonus is as good as none.)
Grasslands, Sorcery or Nature Node : 2 (high food)

The hard part is ores. Honestly, I think a single Gold Ore weights more on your early game than going from 10 to 20 max pop. I have no idea how to factor these in at all. Coal/Iron, even mithril and adamant are not so bad, they are a great advantage but only after you start producing troops and built some infrastructure. But Silver/Gold/Gems generate money, allowing earlier fighter's guild, or summoning more creatures (through alchemy), etc. right away - they are the same deal as Mana Focusing, you can start pushing out units a good dozen turns earlier than normal.

For the allowed range itself, I have no idea. It needs to be something that can be reliably found on all land size and mineral settings, it might need to be a function itself instead of a constant.

[Catwalk]

I think you value food a bit too highly in your calculation. Are you limited to three different values? If not, I suggest this:
Ocean / Tundra
Desert / Swamp / Shore = 1 point
Forest / Hill = 2 points
Grassland / Mountain / Chaos node = 3 points
River / Sorcery Node / Forest Node = 4 points

Silver / Iron / Nightshade / Game / Mithril = 3 points
Gold / Coal / Adamantium / Quork = 6 points
Gems / Crysx = 9 points

Can you give higher priority to ores by your capital than ores on the rest of the continent? If so, double the value of those.

I don't think you should ignore tiles on the continent. Did you mention that you had easy access to the total food output of tiles away from the capital?

[Nelphine]

I'd be more likely to do 0/1/3/4, combining the forest/hills in with the grassland/mountain.

Ores highly depend on your plan - using sprites to take cities? Gems are best. Planning for a very small empire? Adamamtium. Need lots of power near the start? Crysx. Going crazy settlers? Coal. Draconian bowmen? Wild game.

I'd put all 5 of those as 'best', gems are probably above them all since they help any strategy the most. So gems are best, but only a little above (20% ish) gold/crysx/adamantium/coal; they would be about 30% better than wild game and about 75% better than silver/quark/iron.

Then the 'weak' ores would be about two-four times as good as population.

[Catwalk]

It doesn't matter if it's 100% accurate, we just need an estimation. But yeah, I'm probably underestimating Adamantium in the above values.

I disagree that forests and hills are as good as grasslands and mountains, though. Consider this: With 6 forest tiles you get 3 food and 18% production. With 3 grassland tiles and 3 mountain tiles you get 4.5 food and 30% production.

[Seravy]

I think you value food a bit too highly in your calculation. Are you limited to three different values?

Not at all. I just think 3% production is completely marginal and worthless in the early game - You need about 3 of those to gain even a single hammer extra. Is there a point in including it?

I forgot about rivers, good catch.

Can you give higher priority to ores by your capital than ores on the rest of the continent? If so, double the value of those.

I don't want to include the rest of the continent at all. It's far too subjective and with how cheap a trireme is, (as well as water walking etc) I see no point. I don't think we could decide on a distance where to "draw the line" anyway - some continents are spanning across the entire map in one direction and the area that far has zero impact on the early game.

Even if we wanted to include it, the weighting would be a nightmare - an adamant tile on the starting city is clearly not equal value to one that is 10, nor to one that is 20 tiles away on the continent....and while a larger continent is better in theory, in practice it's only true if it's not bigger than what you can reasonably colonize before others find their way into it. 22 tile island with a 25 max pop starting spot vs 200 tiles but starting spot is pop 7, which is better...I don't even want to think about it.

(this entirely subjectivity is what has been keeping me away from wanting to change anything - as the starting city is of course subjective as well, just not as much as the tiles further away.)

Speaking of subjectivity, I think you undervalue ores somewhat...but let's try to measure it in an objective way if we can instead.

Let's measure value in hammers - gold is easy to convert to those, and population produces hammers as well.


However, population growth over time is hard to convert to hammers - each 1 max pop results in 1 extra person in 200 turns, but by then we are way after the time period we are trying to balance. If we go by the assumption that our goal is to even out conditions on turn 50, that means 4 max pop = 1 citizen  =2 hammers+2 gold in taxes = 3 hammers. So 1 max pop = 0.75 hammer.

3% production is 1/3 hammer - A typical early game city has ~10 base hammers (8 from sawmill and 2 from a worker - yes in reality it's more at turn 50, but it's also a lot less before the sawmill is built so this is the average).

If we put hammers at 24 points each, we'll have these numbers :

Desert : 8 (1/3 hammer)
Swamp or Shore : 9 (0.375 hammer)
Forest or Hills : 17
Mountain/chaos : 24 (1 hammer)
Grassland : 27 (1.125 hammer)
River/Sorcery : 36
Nature : 45
Silver : 48 (buys 2 hammers)
Gold : 96
Gems : 144
Wild Game : 48 (1 free worker  =2 hammers)
Nightshade : No early game value at all, but we should probably assign a low number anyway - spell protection on the capital is useful later
Iron/Coal/Mithril/Adamant : impossible to measure objectively but they are more a midgame effect than early
Quork : 60 (if we assume gold=mana) or 120 (if we assume 2 gold = 1 mana)
Crysx : 120 or 240.

If we can agree that my above calculations are right, then we have a good measure for everything except the military ores.

[Nelphine]

Iron and coal are super important early, because you'll always build SOME kind of early units in your capital - either settlers or troops. Can probably convert them to hammers. (Base it on settler)

Mithral isn't super awesome - you might use another city for troops. Adamantium usually means that's your troops city. Very important. Probably gold level.

[Nelphine]

I'd pit the crystals at non alchemy levels. Gems are generally always better, so that matches.

[Seravy]

Assuming the average production of early turns is 10, we have a total of 500 production. Out of that, probably 200 goes to buildings (sawmill itself, and maybe stables, forester's guild, whatever else needed) leaving 300 on units. So Iron is worth 30 hammers over 50 turns, or 0.6 hammers/ turn.
That makes it a value of Iron 14 in the above system, and coal a 28.

I'd say Mithril is ~75, Adamant is 150. Not necessarily for the lasting effect, but by actually making early unit spam viable and highly effective (bowmen and swordsmen are a much bigger deal if they have mithril or adamant).

Nightshade - 15 will do I guess.

And this leaves only the biggest question unanswered. What is the range we want the city to be at?
200-400 total?

[Catwalk]

But you don't need a 100% accurate formula to determine the value of non-capital tiles. You just need a formula that's better than not having a formula. Which is really not very hard. Do you have easy access to food output of non-capital tiles? You hinted at that previously.

There's no reason at all to consider forests, hills and swamps to have the same value. 3% production is clearly not nothing, it means extra hammers.

I think that your calculations are crude estimations based on many assumptions, so it's difficult to say whether or not they're "correct". For one thing, it leaves out the fact that gold is flexible hammers whereas "fixed" hammers can only be used in the same city. I think going by "feel" may be a better approach here, because that allows us to make use of many hours of game experience. If you insist on using a formula, I think gold should be converted to hammers at a 1:2.5 ratio instead. I think mana should be converted to gold at a 2:1 ratio. I also think you should assume 16 base hammers for an early city, 10 is way too low to use as a baseline.

[Nelphine]

That 500 production is WITHOUT any of the bonuses you've listed on this page. So, when you figure out what range us acceptable, you need to modify the 500 production by that, and then apply iron/coal to that. . probably at least doubles the value.

I once said I think an average city should have 14-17 population, and mid-high range ore.

So I think that means 100 from the ore, and.. I have no idea, average whatever tiles would give 14-17 population.

Then give a range from 20% lower than that to 20% higher?