[SPOILERS]PB46: (In Search of) Fine Foreign Dining, by Hannibal of England

[Krill]

Well, on a map like this, it may well be that there is one decent city site adjacent to the capital, and the rest are "OK with attention and growth". May well be that the third city needs to be settled beyond the second city location for decent food resources. Scouting is the only way to know for sure.

Turn 2




Just the one screenshot today. No seafood. Essentially we are going to settle a city in this location, and we are going to have no choice but to throw workers at it and allow it to grow unmolested. It looks like we will need to farm the grass lands W and W-NW of the capital to irrigate the distant rice tile and both capital grains. Maybe we could sacrifice this city as a city to generate GS from, but I reckon the smarter move is either south of the plains ivory for a slow unit pump, or on the ivory and a commerce city. If that is an island to the north, there is no way we will get the scouting we need for a perfect decision any time soon. But hey, easily accessible happiness with trade bait.

Scout move is obvious, NW - SW onto the plains hill and then curve around.

This turn:




(I have low expectations)

[Krill]

BTW, what is the theme for city names this game?

[Krill]

Commodore: Wang Kon [FIN/PRO] (31st pick) of Mongolia (2nd pick)

Mongolia is weird. Commodore went for the easy option to pop borders, but then ignored CHM to actually make use of the XP: There is no difference other than tech requirements between a 30 hammer monument and a 30 hammer stable if you never build mounted in that city. The Ger doesn't do anything different from another stable unless there are other XP modifiers in place. And mounted units aren't draftable, but I suppose he has one eye on amphibious war elephants. Which he can't build without Barracks, Theocracy, and Vassalage stacked on top. He probably didn't see the snake pick falling how it did though, with 3 FIN leaders present, which then got racked up to 5 with his pick and OH's at the end. Still, he had options for Napoleon, or Washington or even Cyrus if he wanted to be faster. Instead he went with PRO, to make him the 4th PRO pick overall. I think he just picked this for the bundle of stats approach at the end and gave up on any synergy. I'm not even sure you can consider the Keshik a UU it's that bland, especially on a Big and Small map. I have to wonder why he didn't pick something different, Zulu or Native America as he started this high up the snake pick. I know for a fact that Commodore is one of the players that values Native America quite highly now (seems like people either think it's a great civ, or junk, doesn't seem to be in between). Wheel/Hunting doesn't help anything except a chariot rush, and this is a map with 200 tiles per person, I've already said chariot rushes appear unwise.

I don't think he is wrong when it came to the leader choice. PRO/FIN, cheap granaries, PRO doesn't synergise with anything now, but you get the extra free strike chances on XBows off boats (but he can't get Amphib through the drill line, but CHM can if they stack everything up). This game is going to be a grind through the tech tree, with so many fast starting but slow teching leaders. 3 ORGs (one of which is Darius), no PHIs, only one SPI. Just a bunch of EXP players speeding through cities 5-10 and then crashing their economies. So he choose to pick his leader to be able to manage the tech cost increases of the Medieval era, and sidestep the problem all the good EXP leaders were gone. And keeps promotions for relevant units alive.

Does he have a plan? Probably not. He might have figured out he can go Granary before settler though, or he has figured out a way to get a quick settler at size 2 and get enough overflow to throw out his granary before reaching size 3 (this would be scarier). Then roll this though horizontal expansion, but without extra happiness he is stuck at 2 pop whips and not stacking the unhappiness that much. Longer term it's a tech lead to Guilds or Engineering and kill someone, via PRO longbows and Xbows for defense. He doesn't care about canned strategies. Build the economy up and a tech lead, and make players react to him whilst looking unappetizing. Commodore probably picked one of the better "No Canned strategy" choices.

[Krill]

Tech Cost Scaling

I've talked about crossbows, and I've talked about the reason for the naval changes (even though to talk about all the effects of the naval changes would probably take twenty thousand words). I don't think either of these will have the same effect on the game as the tech cost scaling. Every time I start to discuss it, I always link back to SevenSpirits original post:

What would you guys think of the following change to the tech cost calculation? The goal here is threefold: 1) Increase later-game tech costs, and do it smoothly. 2) Decrease ancient tech costs a little bit to lessen importance of starting techs 3) Make different world sizes have a more useful effect on tech cost - currently, large world sizes increase the cost of all techs with the justification that players will have more land and thus cap out at a higher tech rate, but having more land doesn't help you with ancient and classical techs.

X is a percent bonus based on world size: 0% for duel up to 50% for huge. Current tech cost modification according to world size is simple: Cost gets multiplied by 1 + X. My proposal is to multiply cost by (1 + the following) instead:

((log(cost of tech) / log (40) - 1) * (1.2X + 0.3)

What does that do? The quick summary is it means the map size multiplier doesn't apply at all to the cheapest techs, applies about as usual to medium-cost techs like paper, and applies by a greater factor to expensive techs.

The first term, ((log(cost of tech) / log (40) - 1), is a number between 0 and 1.5 depending on how far down the tech tree we are. 40-cost techs are 0. Future tech is 1.5 (well, a tiny bit less, but close enough). Paper is approximately in the middle with 0.73. (We are caring about the geometric/multiplicative mean here, not the arithmetic mean. I don't consider advanced flight to be an average-cost tech because it costs half as much as future tech.) The second term, (1.2X + 0.3), is a number between 0.3 and 0.9 depending on the map size.

Multiply these two terms together and you get 0 for a super cheap tech, a number like 20-60% for a tech like paper (depending on map size), and a number like 45-135% for the top end stuff (future tech), depending on map size. Remember this is replacing a number that ranged from 0-50%.

Here's a link to a spreadsheet showing tech costs for variable world size / handicap / etc, compared to how they would look under this formula. https://docs.google.com/spreadsheet/ccc?...sp=sharing
You should be able to make a copy of it so you can try out different variables.

There are various points to this change. One biggie is that Huge maps don't have horrific tech costs in the ancient era, and now play like a 5 player game. Combine it with the tech cost changes for the starting techs (they all cost 45 base beakers) and then the first 50 turns of the game will play out roughly the same, for every game. How the game starts is down to the geography and the leader/civ pick, map size doesn't really enter into it.

But that is just the obvious bit. This change actually cuts through the entire game: It is not the change to the individual techs that matter, it is the windows of aggression, the timeframes that players have to attempt to attack another player have completely changed that need to be considered. Maybe some of the windows appear to be the same, we'll see about that later in the game.

The reasons for this can be explained quite simply: The ancient era techs are cheaper, so compared to base BtS and previous RtR mod games, techs such as IW and HBR can be researched by an earlier turn. So if Swords and HA can be built earlier, that opens up new opportunities to attack. Other techs, such as Guilds, might appear more expensive, and they are. So it would seem more difficult to research Guilds, but the same point applies to Feudalism, and Machinery, which pushes Knights even further out of reach. OTOH, Feudalism is actually no more difficult to research on a beeline than it used to be: the extra cost of getting to Feud is outweighed by the saved cost on the ancient era techs (other changes, such as making Vassalage save twice as much on unit costs and Serfdom gives specialist slots makes this strategy more appealing, but prely on tech cost there is no change in the number of turns taken to reach Feudalism). It is possible that there is now a significantly greater number of turns between the unlocking of longbows, and players reaching the ability to make knights. Knights may not have changed as units, but because of the opportunity for players to have unlocked units to defend with, the effectiveness of knights, if used in what the current meta considers to be usual, may have changed.

As I've said, this runs through the whole tech tree. Any time the unit needed to defend against an invasion, either to sit in a city, to catch invading boats on the sea, or wipe units that wander in as an SoD, any time a specific unit is needed to defend and is available after the attacking unit becomes available, the window of aggression for that unit has widened. That's why Preats are even more terrifying. But it's worse than this in some circumstances, sometimes to defend you need not just a unit, but to also unlock civics, and tech cost scaling may force players to change how they defend. I don't claim to understand all of the effects of this, just that I see that there are going to be effects that may take more than a single game to adjust to.

A further part of Tech Cost Scaling is the change to the tech cost doubling that occured in previous RtR mods. For example, Rifling on a huge map used to cost 7245 beakers, but now it costs 5517 beakers. In previous RtR mods, tech seems to fizzle out after Gunpowder due to tech costs and the prevalence of just workshopping everything. Now that the Rennaisance era techs cost around 15,000 beakers fewer the units unlocked here are actually more usable. Add in Emancipation being usable, and suddenly there are opprortunities for larger games to pass through the Rennaisance era and possible reach tanks and marines in the Industrial era.

The rule of thumb I'm using are this: Ancient era stuff is easier to get. Anything that doesn't scale into the Classical era is junk (looking at you Greek Phalanx and Persian Immortal. Egyptian war chariots are fine as they collateral catapults now, Sumerian Vulture is a high strength unit that always has value due to cost). The Classical era should start around T50 and probably lasts until when it used to finish, so anything that is unlocked right at the start of the Classical era and isn't countered by earlier stuff has increased value (Praetorian, Basic HA. Theoretically the catapult). The Medieval era takes longer to get through as that is when tech cost scaling starts to really kick in (and any bulb plan to opens up Ren era techs should last longer as it takes longer to reach those same bulbed techs). Anything unlocked at the end of the Medieval era is of dubious value, but if something can be beelined and opened up earlier then that may have increased value (Knights are not one right choices, but Engineering can be beelined so Pikes and Castles are of greater use now). And then the Rennaisance era is an unknown quantity due to other changes, but my gut feeling is that the units at teh end of the Ren era are now viable power spikes to end the game.

And, finally, bulbs: Any bulb plans that comes into use at the end of the Medieval era, and the start of the Rennaisance, will give you greater bang for you buck than before, because of the effects on tech costs for other players trying to follow through per turn research. If in doubt, double bulb Astro. Easiest double bulb to set up and because of the lack of optional prerequisites each bulb is worth 20% more and you have to research it anyway to enable frigates.

[Krill]

I don't know, you wait all day for a turn that by rights should have been available 8 hours prior, and this is what I get...

Turn 3




No resources of relevance, incense on a desert tile behind peaks. Peak 3S1W of the scout and next turn I have to move the scout S anyway. Almost certainly ought to move the scout S-SE because heading any further south west doesn't check the plains tile 3S1E of the scout. Nothing much to say, really. I wouldn't mind uncovering a food resource we can use in the next few turns...

EDIT: Actually, it looks like we are going to get mileage out of 50-ish hammer libraries and wide city spacing.

[GermanJoey]

BTW, what is the theme for city names this game?

No idea! I just went with something whimsical for the capital because I didn't know what your theme plans were.

[GermanJoey]

The best thing I can say for that land is that wow, windmills and Fin cottages are the best possible thing you could do with it because GODDAMN is it barren. Workshops + Caste don't seem feasible here because there's not enough food to support them, pre-SP at least. Note: that also goes for the coast; if you're working workshops with low food then you can't work much coast, if at all.

[GermanJoey]

By the way, can you link to that mapscript? I wanna play around with it and the new mod changes.

[Krill]

The map script is full of resources, the second file is the settings file that the mapscript uses to generate the map (goes in the my games/civ4 FoR folder).

Do you know where Borsche is? He is holding the game up.

[Krill]

The best thing I can say for that land is that wow, windmills and Fin cottages are the best possible thing you could do with it because GODDAMN is it barren. Workshops + Caste don't seem feasible here because there's not enough food to support them, pre-SP at least. Note: that also goes for the coast; if you're working workshops with low food then you can't work much coast, if at all.

I think that on low food areas, you have to go more with hybrid cities: farm a little bit extra, grow to a decent size (defined by available happiness, not by available food), and then swap over to work maximum cottages. If the city size goes past size 10, then it makes more sense to level out on specialists or hill mines rather than bother with workshops no, unless the city is completely flat. Use the hammers to help get out what city improvements you need, and once emancipation comes along, the Towns you already have will give you the food boost to grow the extra few pop sizes whilst you pick up new cottages. Essentially players cottage in two (main) bursts, but never really have to stop growing new cottages. One at the start of the game, and then another burst at Emancipation.

By growing new cottages all of the time, I think it limits the weaker period that Emancipation might find itself in when a player swaps to Emancipation. The extra food from the pre-grown cottages cal also be used to power whatever other production tiles are needed. With this in mind, I want to try and settle wide (which it looks like we have to do due to the paucity of food resources) and have every city work a minimum of 4 cottages.