DESIGN: Terrain Values

[tyrantpimp]

Originally posted by MrBaggins
Have you played the Apolyton pack? Do you dislike the Apolyton Pack? Are the terrain effects better or worse than default?

If so, can you explain how the changes in the Apolyton Pack adversely effect your play?

Woe slow down sport have you been getting enough sleep lately? Anyway the apolyton pack is fine i didnt know it had tile tweaks though. I have been playing this playtest version and taking notes to try to help out the efforts of the project so i havent played a mod in a while.

I disagree about the coast tile being made less attractive though . .

[MrBaggins]

If you didn't know it had tile tweaks, and its been tweaked extensively, how do you know that tweaks we'd make now couldn't be a significant improvement over default?

The fact is, that the original values were implemented with comparitively little playtesting. Three years have elapsed since.

Without actually playtesting different values, its very difficult to say whether something is good or bad, although there are a couple of constants...

Production is important early. Starting positions should tend to give roughly equivalent resources. Forests shouldn't be a penalty.

Changing the production value of a beach to make it a moderately negative terrain isn't a significantly unbalancing situation, since there is only one kind of beach... and every costal city is in the same boat.

Given that values are merely relative, decreasing the values from ideal to less than ideal, has relatively little ultimate effect. The tile improvement effects have always been more significant in CTP2, anyway.

Its about PW in CTP2, not whales or cows.

[tyrantpimp]

I guess i take being human for granted i adapt to whatever resources are available. Seems like civs in weak startup areas would have to concentrate on military to survive. There are numerous instances of this occuring in history. If however the AI cannot be made to achieve this then by all means tweak the tiles if it will operate as a stop gap.

[Peter Triggs]

I disagree about the coast tile being made less attractive though . .

Me, too. IIRC, this was something WesW did very early on and I don't think it was one of his better ideas. He wanted to get the AI to build more interior cities, but this isn't the data you should be adjusting to bring that about.

[Maquiladora]

...and every costal city is in the same boat.

Nice.

He wanted to get the AI to build more interior cities, but this isn't the data you should be adjusting to bring that about.

When i made the change of beach tiles i never had that in mind, in fact im against that. My main thought was that beach tiles are too good, its as simple as that. Its takes away choices, if youre stuck in a bad place just plonk it next to as many beach tiles as possible.

On any map more coastal cities than your opponent is almost ALWAYS better. In the case when you dont want anymore coastal cities is when youre on the same continent as your enemy (so you can still attacK) AND they clearly already have naval superiority over you anyway.

I also increased production on forests. In the 3 test games (upto Feudalism) i saw no effect of lower production on beaches, not even early production. The map was max diverse and we all drew a scattering of forests, but i think this is more interesting than simply making beaches a safe settle location.

ALL TI's were made 2 times more expensive, anticipating forests would increase overall production. It did increase, but not alot, maybe 5-10% (even with beaches 0 prod.). It turned out these higher TI costs were good, upto where i played anyway. Especially 400PW per road through forest. I didnt play enough to see the long term effects of these TI costs, though it could be they cripple the AI in defence... as with all these changes its hard to "test" anything with the AI.

Im still thinking about alot of balancing issues though, such as,

- Productive locations (forests/plains/mountains)
vs
- Growth locations (grassland/beaches)

Theres nothing that can be done in terms of TI's (AFAIK) to make production less effective.

In a maze of ideas i thought maybe we could have a "Rations Tax", like regular Tax on production, except you assign a % of food collected to go to PW OR we have a seperate pool for Rations Tax collected and each tile imp costs PW AND Rations to build. Just an idea.

[Tamerlin]

Originally posted by MrBaggins
The fact is, that the original values were implemented with comparitively little playtesting. Three years have elapsed since.

I find the SAP2 values far better as your TI policy is more important than in the original game.

Production is important early. Starting positions should tend to give roughly equivalent resources. Forests shouldn't be a penalty.

IMO the AI settlers should rather be taught how to evaluate the surrounding tiles and to explore a few tiles before building the first city if needed.

[Maquiladora]

I find the SAP2 values far better as your TI policy is more important than in the original game.

In a way the SAP2 values make it even easier to blanket your civ with TI's, because theres more production around due to forests and TI's havent increased in cost at all. The only thing that does make "tile imping" slightly difficult is the constant state of war all game but all you need is one turn on 100% and you have a few cities covered in farms. I still think all TI's need new costs.

IMO the AI settlers should rather be taught how to evaluate the surrounding tiles and to explore a few tiles before building the first city if needed.

It would be best if the AI could look up the 21 tiles of a 2 border city for finding a good city site, then look at the total resources of those 21 tiles, then look at the city square (make sure it has good growth) then make sure they wont overlap with another city, check its adequate for production and growth in the initial 9 tiles (sci comes later), then settle. Or am i dreaming?

[Tamerlin]

Originally posted by Maquiladora

It would be best if the AI could look up the 21 tiles of a 2 border city for finding a good city site, then look at the total resources of those 21 tiles, then look at the city square (make sure it has good growth) then make sure they wont overlap with another city, check its adequate for production and growth in the initial 9 tiles (sci comes later), then settle. Or am i dreaming?

Isn't it just a matter of IF and THEN... ?

[MrBaggins]

Not quite, but something like that.

[MrBaggins]

Originally posted by Maquiladora
Im still thinking about alot of balancing issues though, such as,

- Productive locations (forests/plains/mountains)
vs
- Growth locations (grassland/beaches)

Theres nothing that can be done in terms of TI's (AFAIK) to make production less effective.

In a maze of ideas i thought maybe we could have a "Rations Tax", like regular Tax on production, except you assign a % of food collected to go to PW OR we have a seperate pool for Rations Tax collected and each tile imp costs PW AND Rations to build. Just an idea.

The production issue is... the more production you have, the faster you build settlers you can build, and thus more cities... and more cities is always better... right?

I mentioned above about increasing the city square production value, and then balancing that by lessening the workday multiplier. If you do that it will make the extra squares less significant, in the total production.

The other thing to do, to make growth more important, is to make growth more NECESSARY for Settlers. Its been said before, but make Settlers cost 2 city sizes to build. You might even wish to reduce the actual production cost a tad.

Since we have actual population, we can call it 20000 pop (and have other units cost pop too... maybe 100-1000?.)

If units cost population, then growth will become more important.

[Maquiladora]

Sorry Mr Baggins i missed the workday thing first time around.

So how do i,

make Settlers cost 2 city sizes to build.

I see the flag "BuildingRemovesAPop" in units.txt but thats all. Or were you mentioning it as something to do in the future?

[MrBaggins]

Something to do in the future, although its been mentioned elsewhere, in DESIGN discussions, IIRC.

[The Big Mc]

what everybody think of adding some new terrains to be used by the random map generator

like

forest hills

underwater mountain ranges

[MrBaggins]

The final piece of the early food vs prod (and prod always being better right now) balance issue would be to have, as has been mentioned before, optional food (and gold) upkeeps for units.

If Settlers have a reasonably large food (20-30) upkeep, then a higher production/lower food civ can't pump settlers without regard to food... they need food to support the settlers on their journey, or their city (ies) starves, which defeats the point of building the settler.

Its relatively simple to have the AI keep in check (don't start or finish building a settler that you can't support,) and will increase the early importance of food, which was vital, historically.

[SMIFFGIG]

what everybody think of adding some new terrains to be used by the random map generator

IMO there should be MUCH more terrain, terrain diversity and unique terrain pieces (like the natural wonders mod only with MUCH improved graphics) also remember the places of interest in Alpha Centuri

This game is about realism and of course fun, the more diverse terrain is and varied amounts of terrain adds both realism and fun, and who knows u might end up being the first civilization to discover the the (a) grand canyon or largest mountain in the world

May i also add, naming areas on the map (and already named areas possibly) +re-naming) should be an option in CTP2 now that the source code is available. Exactly like in Alpha Centuri and Civ3

Its something small i missed from both those games that wasnt in CTP2 its fun and does add to realism when you can name all the different provinces in your civilizations territory

I hope both these ideas will be done *crosses fingers*

[MrBaggins]

Originally posted by SMIFFGIG
IMO there should be MUCH more terrain, terrain diversity and unique terrain pieces (like the natural wonders mod only with MUCH improved graphics) also remember the places of interest in Alpha Centuri

Those aren't new terrains, though... just specially impressive versions of terrain... implemented using special "terrain improvements". I don't see any reasons to change that, either... unless there are some specifically general forms of terrain that are obviously missing... which might be just flood plain... and is that modelled ok, already?

We have to work on the map generator with better geology and climatology realism focus, before we decide what, if any terrains are missing.

[The Big Mc]

Originally posted by MrBaggins


unless there are some specifically general forms of terrain that are obviously missing... which might be just flood plain... and is that modelled ok, already?

What about a lot of underwater terrain the way CTP does is makes the sea floor flat and boring :d

[MrBaggins]

True, but whats the value of one deep sea tile vs. another?

We already have the major "different" deep sea features... Trenches, volcanos and rifts are obviously different. Apart from that, every deep tile is similar to another, from a game effect viewpoint.

For shallow water, we also already have appropriate variety.

[The Big Mc]

Is it if I make my sea city on top of a mountain I can build directly into it and use it to cut cost on the city


Also the land and shallow water is good eye candy you don’t really have large tracks of land. Underwater you seam to find a nice spot and park it without really taking into the terrain values because there are no penalties or consciences.

[MrBaggins]

Originally posted by The Big Mc
Is it if I make my sea city on top of a mountain I can build directly into it and use it to cut cost on the city


Also the land and shallow water is good eye candy you don’t really have large tracks of land. Underwater you seam to find a nice spot and park it without really taking into the terrain values because there are no penalties or consciences.

Sea Cities haven't been well balanced because they occur at a point in the game that doesn't matter.

We should obviously improve the AI, and resilience of the game challenge... however, we have to decide how sea cities should work...

E.G.

Should sea cities require a volcano (I.E. geothermic generation source)?
Should sea cities be impractical without an abundant food source (E.G. squids or whales good)?
Should deep sea food TI's be rebalanced? (TI's, not the underlying tiles base value are currently the key component to a cities production)

We need specifics of implementation... not "deep sea should be more varied."

[J Bytheway]

Originally posted by The Big Mc
What about a lot of underwater terrain the way CTP does is makes the sea floor flat and boring :d

To be fair, the sea floor really is very flat and boring, except for the occasional mid ocean ridge.

[SMIFFGIG]

Should sea cities be impractical without an abundant food source (E.G. squids or whales good)?

Too be honest Giant Squid is the most ridiculous trade good in the game and i would be quite happy to see it cut and replaced by something realistic
especially since it can be harvested in times before christ.

Anyway i think this is getting on to realy changin the dynamics of the game

However i do think the terrain should be greatly varied eventually currently its much too uniform and (boring) and the sea floor should be drastically improved, currently it is also very (unrealistically) bland

[MrBaggins]

Originally posted by SMIFFGIG

Too be honest Giant Squid is the most ridiculous trade good in the game and i would be quite happy to see it cut and replaced by something realistic
especially since it can be harvested in times before christ.

Anyway i think this is getting on to realy changin the dynamics of the game

However i do think the terrain should be greatly varied eventually currently its much too uniform and (boring) and the sea floor should be drastically improved, currently it is also very (unrealistically) bland

Again, this is all well and good, but we're not a "Civilization Games General/Future" forum... we're a Source code project group. "We" are the designers... not a bunch of fans giving arbitrary ideas to some other designers to consider.

If you have an actual implementation to consider, I'd love to hear it... otherwise, I don't see the value, unless your point is to remove something specifically unrealistic, without any suggestion of replacement.

E.G. Removing Giant Squid as a Trade Good.

[The Big Mc]

Well MrBaggins you can’t eat Giant Squid it is actually a poison

AS for food you remember ctp in space your city would starve unless you made food tanks so something like that. Well I think a sea city would do with becoming modular i.e. you can’t make a tile imp without it connecting to the city or another tile imp the terrain then obstructs you from achieving this so planting your city in a underwater mountain range is not a good idea.

[MrBaggins]

Originally posted by The Big Mc
Well MrBaggins you can’t eat Giant Squid it is actually a poison

AS for food you remember ctp in space your city would starve unless you made food tanks so something like that. Well I think a sea city would do with becoming modular i.e. you can’t make a tile imp without it connecting to the city or another tile imp the terrain then obstructs you from achieving this so planting your city in a underwater mountain range is not a good idea.

Thats an interesting point about the giant squid... but I think we need to find a replacement good, for some variety of deep sea resources... Maybe swordfish, tuna shoal or something?

Space cities worked that way for a reason. It stands to reason that space tiles have zero food, because they are the definition of dead terrain... no food in the vacuum of space. There are zero restrictions on construction.

If you had a rift preventing construction, I think that'd be a problem. They already restrict undersea tunnels, and I think thats enough, frankly. With a trench (I.E. very very deep) terrain type, I think that should perhaps prevent underwater TI (or city) construction, due to the extreme depth involved.

[tyrantpimp]

Its my understanding that most of the mineral wealth on the planet is under the sea so it should be easy to compensate or add to trade goods if you look at it like that. Of course modern man is unable to tap most of it but then again the future is yet written.

One ti i would like to see is mine fields and unsersea mines that would be most excellent. Of course it wouldnt be cool unless the AI could use it effectively. But if it did that would be sweeet.

[The Big Mc]

I tried to do that in slic ones I got the tile imp from ctp some where.

[Maquiladora]

Okay well i did some tests yesterday with some of the changes Mr Baggins suggested. The test was build 10 cities as quick as possible, but not using rushbuying or farmers, because it can throw things off.

Default values for grass and plains. In Const.txt i changed UNIT_WORKDAY to 0.005 it was 0.015. 5% is still alot later in the game but just low enough to encourage lowering the workday slider for a lower wages slider instead. Still needs testing though, along with changes to happiness.

The table shows how many turns it took to build 10 cities on each terrain with different settler costs each time.

Grassland Time	Settler Cost	Plains Time

113 Turns	740		84 Turns
85 Turns	600		74 Turns
63 Turns	400		55 Turns

Obviously the grassland cities finished the test with bigger cities and obviously the plains cities still had some time to grow and had more production.

I was worried about changing settler costs too much, afterall the costs werent put that high for no reason. I guess it was probably to help prevent rapid early expansion strategy.

In the default game thats the strategy anyway, all it does is slow it down and as the table shows, higher settler cost increases the gap in good and bad production land.

edit: Another rather big change i made was to give city square 30 Production, instead of default 15.

[MrBaggins]

Exactly what I thought would happen... and it proves part of the supposition.

The other part, which we can't model, til I (or someone else) codes it , is increased initial growth cost for settlers (I.E. 2 city sizes for a settler,) and food upkeep costs.

These should push the equation over, even further, to making grassland (and food in general) more beneficial for city growth, than production.

I'd say that squeezing the relative importance of the city square and surrounding tiles (considering that we'll be dealing with mixed terrain, most of the time) plus changes to the cost of settlers to 600-500 ish... are maybe the beggining of a sweet spot...

[Tamerlin]

As far as I am concerned, the earlier part of the game is often the most interesting one, if you slow it down (higher Settler costs...) I fear it could become rather boring (as in Civ3).

I don't understand why these values should be hardcoded as it seems that they can be changed through SLIC. Whatever the case, I hope these new terrain and unit values will not be incorporated into the original game patch or that if they are they will be optional.