Rebalancing Civ4: RtR Mod

[Cyneheard]

Answering your questions about the current no-Espionage system:

Spies cannot be built.
Communism will not give a Great Spy.
Spy specialists can still be run. They do create culture.
Espionage points are first calculated as though it was a "normal" game, and then converted into culture. This means that EP get the Jail's %age bonus, but not Hermitage's.

The other XML-based option would be to end up with 2 different versions of the mod. That seems like a sub-optimal outcome.

[plako]

Totally removing Espionage seems quite complicated, but if it is easy to do then why not. Alternatively you could just try to remove spies and all means to generate Great spies and left everything else intact.

[novice]

1. Great Lighthouse may be unbalanced? Nerf it to +1 trade routes?

2. Police state buff if WW is removed: +1 happy from barracks, stacks with nationhood.

3. Do people really dislike espionage that much? Is it unbalanced? If we could find a way to make spy missions possible (removing the broken missions) without having the unkillable sentry problem, that would be very cool imo. And I like passive espionage effects, so I would suggest at worst to make spies unbuildable.

[regoarrarr]

7 from WP feels too much. Giving e.g. aggresive leaders relatively easy access to Commando promos. Making it just cheaper would be better.

I for one know how useful (or devastating) this can be

[Krill]

Could always move Commando to combat 5...

[Krill]

How about keeping it, but drastically increasing the fade rate? Like to the point where it disappears after about 20 turns. So the WW modifiers still have a use, you'll still incur a penalty if you get a big stack killed in enemy lands, but the hit is only temporary and you won't suffer all game for one bad event.

I like this idea.

How about correctly fixing the No Espionage game setup option? Then you can go either way, spies or no. I see this list of changes:

- remove the silly espionage-to-culture conversion
- and remove the doubled culture thresholds
- make spies unbuildable (is this already true?)
- no Great Spy at Communism (is this already true?)
- no spy slots and EP from buildings (is this already true?)
- fix Great Spy points to produce a proper kind of GPP instead of those weird typeless points. I'd say Merchant, like your analysis of the Great Wall. Merchants are always useful but never broken.

The changes would most likely have to be done in the DLL code. I have the ability to code and compile that.

Help compiling a DLL would be absolutely tremendous...but we'd still need to do the the Great General change for the GW...or change the bulb preferences for GMs. Merchant bulb to CS is the same as a GP bulb, which has been made a lot more difficult because it was so overpowered.

[novice]

Help compiling a DLL would be absolutely tremendous...but we'd still need to do the the Great General change for the GW...or change the bulb preferences for GMs. Merchant bulb to CS is the same as a GP bulb, which has been made a lot more difficult because it was so overpowered.

Great Wall could produce Great General points even with espionage enabled, I think that's an adequate solution.

[SevenSpirits]

I'm still working on putting things together for the next release version, but some of the changes we've made:

Slavery now gives 20h per pop, but no longer gives anger. The other option I see is to do a straight nerf of slavery to 25h per pop, but leaving the anger as it is.

Serfdom, Windmills:
Windmills are back to a base +1/0/+1 improvement, and still get +1h at Rep Parts (Mine still get that boost, as well). However, Serfdom now gives +1h to Windmills and Watermills, and only gives a 75% worker speed bonus (which on Quick or Normal speed, acts more like a 50% bonus than a 100% bonus)

I'm happy to leave if I'm the only one who feels this way, but... don't you think you should be explaining why you think these are good ideas?

[Cyneheard]

Note: The more I think about it, the less I like the 20h/no unhappy penalty version for slavery. This basically means that the happy cap does not matter, as there will always be a painless way to turn that unhappy population into something useful. That is a bad thing, the happy cap needs to provide a real constraint on action. Slavery should be a way to manage the happy cap, not circumvent it.

WARNING: WALL OF TEXT

Justifying Slavery/Serfdom changes:

First, what does slavery do in the game?
Three things:
1) Turn excess food into something useful
2) Generate production
3') Slavery is the only "emergency" source of hammers in the game before the draft, unless you're playing as India.

Why should slavery get a nerf? Most early game production, and often a large portion of midgame production, comes from either food via the whip, or from forests. Mines are important, but under the status quo, within the constraints of the happy cap, slavery is both incredibly effective, and makes 1f so much more valuable than 1h. This makes map balancing all about the food: The best hammer tile in the game is a river corn. Food should be the most important resource (always going to be the case, as food means pop which means hammers or gold, if you want it), but right now I think it goes too far in being the only resource that matters. And when the surest sign of an MP noob is someone who doesn't use Slavery, I think that's a clear sign that Slavery is the only way to go for too much of the game.

At size 5, it takes 30f to grow, 15 with the granary. Size 5 is near a city's average for the early game, and this means that each food is worth about 2 production. So a 2/0/X cottage nets into almost as much production as that 0/4 mine, while providing significant levels of commerce. At one point in PB2, I think lurking Spullla, Sunrise said (paraphrase) "What, you think I should work mines like a chump?" Considering that there are only 3 early-game tile improvements, the idea that cottages are almost strictly better than PH mines, and almost as good at what a Grass mine is best at, is a bad problem for interesting tile management choices. With a nerfed, but viable, Slavery, mines would truly be better hammer producers than cottages, without going too far in requiring a player to use mines to get early hammers.
Most non-mine food to hammer conversions are 1:1 (or sometimes even less: Settler/Worker food is often less valuable). So leveraging slavery effectively gives someone a very large advantage relative to other choices, made even worse by the fact that Slavery's the only emergency source of units until drafting. While micromanaging Slavery is very skill-intensive, and to me and others a fun mechanic, Civ4 will be a better game if, while keeping the micromanaging skill of Slavery as being relevant, players also have to genuinely think about the trade-offs of what a Slavery-based economy means relative to a Serfdom-based or Caste-based one.

In the unmodified game, Slavery is the civic of choice at least 90% of the time between Code of Laws/Feudalism and the Chemistry/State Property era. And that's when there are two or three other choices out there!

So, we'll try a nerf of slavery to 25h per whip, and see how the new civics compare. At 25h, slavery is still the best food-to-hammer conversion in the game for a long time (for that size 5 city, 1.7h per food), but the happy cap constrains it, and we are intentionally narrowing the hammer-creating effectiveness of Slavery relative to the other Labor Civics. Ideally, if each of the three civics are used at least 20% of the time during that era, then I think we've succeeded in making balanced and interesting choices between the three. Caste dominating the State Property era (unless one chooses to use corporations, where Caste is usually needed for all that food) is fine; Slavery has its own exclusive era.

Serfdom: Used very rarely. Mostly Spiritual players will use it, and then only in rather specific situations. Ruff in PBEM1, and Krill in PB3, and maybe a few others that I didn't notice/forgot about. Still we're talking something like 50t out of thousands of player-turns where Serfdom was a possibility. So, how does our modified Serfdom (+75% worker speed, +1h to Windmills and Watermills; note that Watermills are now +2 base hammers, and still gain 2 commerce later in the game at Electricity. Windmills are otherwise unmodified) compare to slavery?
1) Turn excess food into something useful: Well, this means that workers are more effective. So use excess food to build more workers. Serfdom has the relative advantage of being a limitless conversion of food to useful stuff: Worker turns are always valuable, but it's also a very constrained use of excess food. That's fine; Slavery's excess food conversion is limited, but more efficient.
2) Generate production: This was lacking under the old system. Now, if someone wants to be in Serfdom, some of the slower tile improvements, Watermills and Windmills, are viable methods of generating production. While Windmills at +1/+1/+1 makes them a very good improvement, that excess food cannot be simultaneously converted into 1.5-2 hammers via the whip. Watermills are also powerful, but they can't be built everywhere, and still require a fair number of worker turns to build, even under Serfdom (5t @ Normal, 4t @ Quick).
3') Emergency production: Not really. Chops are 2t instead of 3t, so there's some of that (and 1t on Quick if you also build Hagia or post-Steam Power). But chops are a finite resource, instead of a renewable one. Of course, the worker bonus means that a city will have more improved tiles to work, and now the population to do so.

Caste System: Caste is used a fair amount in the base game. Easy border pops, synergy with Merc and Rep, being able to make Great People, particularly Merchants and Artists, and at least post-Chemistry, if not under State Property, as part of a Workshop economy. And we're not changing the bonuses of Caste, but we're making them relatively more powerful by weakening an alternative and by making something it boosts, Workshops, better. I hope this gets the balance right in making Caste System a competitive civic choice in more situations.
1) Turning excess food into something useful. Caste does a very good job of this one. 2f can be turned into either 3 or 6 commerce, or 4 culture and 1 or 4 beakers. Oh, and 3 (or 6 or 9 or 12 or 15) Great People points. However, without Rep, players focusing heavily on a specialist economy tend to lose to cottages + slavery. That cottage will also be producing 3 commerce in a relatively short period of time, and is far better at #2.
2) Generating production. The workshop bonus does this late-game. In the Guilds to Chemistry era, a Caste Workshop is -1f/+3h. Since each food can be turned into almost 2h under Slavery, this makes the workshop only a little better than a +1h improvement. And this is the only production benefit that Caste gives: something that's worse than a mine.
3') Emergency Production: Nope. Instead, though, it allows you to get 3 different types of Great People without being constrained by a lack of buildings. But that really only applies to Merchants and Artists; 2 Scientist slots is usually sufficient for a reasonably quick 1st or 2nd Great Person, and Theaters are cheap, while Markets are useful, and come not too long after COL. By the time a player gets their 3rd or 4th Great Person, you've had plenty of other options beyond having to resort to Caste.

Emancipation: This is a strange civic. The unhappiness penalty makes it a very tempting target in a TT game where allies are likely to be closely coordinating, and where only one player has to spend the beakers for this very expensive tech. The cottage growth is obviously meant to synergize with US and FS, where even saving a few turns on getting +3 commerce and +1hammer is worthwhile. I don't think we want to try to think of a way to tweak Emancipation. The fact that if half the game takes it, the other half needs to get access to Emancipation really really quickly makes it too sui generis to try and balance properly.

Also, we're trying to avoid changing the overall pace of the game too much. That's one of those subtle things that could very easily cause more problems than we solve. Obviously, a Slavery nerf will slow down the first 75t or so of the game, but midgame production now has more options to work with that should, on average, make production similar or slightly better than the status quo.

As a side note (and because forest chops are the other instant production method): I really like the forest chop implementation: worker turns are precious in the early game, chops are limited, and there's that 50% boost to chops with Mathematics, so there are some really interesting tradeoffs of 20h now and 4 worker turns later for 4 worker turns now and 30h later. The only one possibility to consider would be making Lumbermills available at, say, Machinery. If Watermills become too powerful under Serfdom, then one possibility might be to move the Serfdom boost to Lumbermills, and make them available at Machinery.

[Cyneheard]

I've been testing an XML-based solution to the sentry problem of spies. It is possible to make spies only see their own tiles. How to do it:
Create a "Spy" promotion that gives -2 visibility to the spy. All spies get it for free, no other unit can get it. It works, even on hills, next to hills, next to peaks, and on hills overlooking water. If a spy cannot see his own tile, then bad things can happen (namely, the spy can't be selected if he can't be seen by someone else unless unit cycling brings you to the unit, and you only see the selection icon, not the spy itself). Since spies don't get XP, they can't get the Sentry promotion.

So, we've got several options for spies:
1) An XML-set of changes that removes espionage completely
2) If T-Hawk is willing, then a DLL set of changes to fix the No Espionage setting
3) Removing the ability of spies to see anything beyond their own tile, and removing at least the Swap Civic and Swap Religion missions.

2) and 3) could even be combined, since there are some people who like the idea of an espionage economy, but the invisible-sentry problem is a serious one in MP play. Of course, a one-tile sentry (well, 2-4 tiles per turn, depending on roads and era) might still get lucky, but they become a lot more expensive to have enough for a safe net.