When is bulbing the right choice

[cseanny]

First gs is usually an academy. Palace has eight commerce....doesn't take much for academy to beat out settling.

This. Especially with deficit research and then big gold surpluses once your Academy is done. Save that 2nd GS for a huge beaker bulb that will save you 5-7 turns while settling an early GS wouldn't net you anything more than perhaps 1 turn saved. Big waste.

 

[TheWilltoAct]

This. Especially with deficit research and then big gold surpluses once your Academy is done. Save that 2nd GS for a huge beaker bulb that will save you 5-7 turns while settling an early GS wouldn't net you anything more than perhaps 1 turn saved. Big waste.

Deficit research is a good argument, but I don't use it myself. Maybe I should!

I understand you are eager to rule out the settling option, but I'm not convinced that a second Great Scientist will be able to bulb something so advantageous as to blow settling out of the water. You advocate saving a Great Scientist, if you mean that literally then we have to factor in some wastage from it not being used. Not sure you mean that though.

Settling nets you 1 turn saved? Let's avoid concrete numbers, we are discussing hypotheticals.

 

[noto2]

You say bulb a tech that takes 7 turns to research and that blows settling out of the water? This is at a point in the game when your total beaker output is going to be less than 60, so settling increases research output by 10%. Over the course of the game that settled scientist will do far more than save you just 7 turns of research. That's my general rule of thumb. If settling would increase my GNP by less than 10%, I won't settle, otherwise I might. Perhaps I REX more than most people, I don't know, but when I hit writing there's no way I'm running deficit research because that's the point in the game when I'm pumping out settlers. If I don't, the AI will box me in and I'll only have 2-3 cities total, which is too few for nearly any strategy.

 

[FlyinJohnnyL]

If you bult something, you aren't usually just saving the 7 turns. It's the fact that you bulb something you can trade around to other civs. Obviously this making Pangea more attractive for bulbing, whereas an isolated start may skip early bulbing altogether.

 

[cseanny]

Perhaps I REX more than most people, I don't know, but when I hit writing there's no way I'm running deficit research because that's the point in the game when I'm pumping out settlers. If I don't, the AI will box me in and I'll only have 2-3 cities total, which is too few for nearly any strategy.

I don't see what deficit research has to do with ones ability to pump out settlers. You're growing cottages by sharing tiles for the Capital or with a good Capital site working cottages in conjunction with working good luxuries/tiles. On Immortal and below its safe enough to grab your military resource with your 3rd city which gives you more freedom with city placement and planning.

My point being, it's very easy to expand, grow cottages in Capital and run some scientist (my norm lately is in my 2nd city after I've made a settler there). With a few riverside cottages its pretty normal to expand to 5 cities nlt 1000 BC and already have finished a tech like Alpha, Math, or Aesthetics (usually 5-8 turns or even more before 1000 BC) and then you find yourself waiting for your 1st GS to finish.

For me that is usually between 1000 BC and 800 BC at which point you will have a gold surplus, build your Academy and then run at 100% research for quite some time. Picking up more luxuries, trade, and techs like Monarchy (trade too) let you continually grow and work more and more cottages compounding the effect of that Academy.

A settled GS earning 6 beakers x 40 turns gives 240 beakers. An early Academy giving +20 beakers x 10 turns (100% research from deficit gold surplus) = 200 beakers. With further Capital and tile growth that Academy could easily gain you an additional 400-900 beakers over the course of 30 more turns, even taking into account time needed to build gold reserves again before running at 100% research again. Something which makes early Currency even more powerful.

So imo, settling your 1st GS over an early Academy is costing you at least 500 beakers.

 

[§L¥ Gµ¥]

if you're going to bulb, make sure it's part of a greater plan. Knowing when, and what you're going to pop before it happens is key. Read and know the GP tech preferences, and make sure your own research is in-line with what you want to bulb and from whom.

Bulbing provides 1 tech or up to what is it, like 1000 beakers? (more for GS, I believe), so you'd be getting best value around the time of Philosophy.

Consequently, with GS's (the GP I pop most often), my tech bulbs are usually one for Philosophy [great trade chip. Far enough away from liberalism to not be a worry in the race], Education [won't get the whole thing, but around this time Education is a bit of a stretch at 2800 beakers], potentially paper if my GS's are slightly mis-timed.

Early tech just aren't worth bulbing, IMO. So you bulb something like iron working.....saves you 15 turns. That translates into what? Like 1.5-2 axes, maybe 3 if you're going all out? Not game changing when you consider the 15 turns you save on education, and the 10 more you save from getting liberalism, especially when you're on your way to a dominating tech 50-75 turns ahead of it's time.

 

[noto2]

I don't see what deficit research has to do with ones ability to pump out settlers. You're growing cottages by sharing tiles for the Capital or with a good Capital site working cottages in conjunction with working good luxuries/tiles. On Immortal and below its safe enough to grab your military resource with your 3rd city which gives you more freedom with city placement and planning.

My point being, it's very easy to expand, grow cottages in Capital and run some scientist (my norm lately is in my 2nd city after I've made a settler there). With a few riverside cottages its pretty normal to expand to 5 cities nlt 1000 BC and already have finished a tech like Alpha, Math, or Aesthetics (usually 5-8 turns or even more before 1000 BC) and then you find yourself waiting for your 1st GS to finish.

For me that is usually between 1000 BC and 800 BC at which point you will have a gold surplus, build your Academy and then run at 100% research for quite some time. Picking up more luxuries, trade, and techs like Monarchy (trade too) let you continually grow and work more and more cottages compounding the effect of that Academy.

A settled GS earning 6 beakers x 40 turns gives 240 beakers. An early Academy giving +20 beakers x 10 turns (100% research from deficit gold surplus) = 200 beakers. With further Capital and tile growth that Academy could easily gain you an additional 400-900 beakers over the course of 30 more turns, even taking into account time needed to build gold reserves again before running at 100% research again. Something which makes early Currency even more powerful.

So imo, settling your 1st GS over an early Academy is costing you at least 500 beakers.

sorry, my bad. I've been posting so much in the Kmod thread I got confused and thought this was Kmod. I haven't played the retail version of BTS in years, it's a huge difference.

 

[TheMeInTeam]

Deficit research is a good argument, but I don't use it myself. Maybe I should!

Trade techs for gold. You have better research multipliers early, especially with an academy.

I understand you are eager to rule out the settling option, but I'm not convinced that a second Great Scientist will be able to bulb something so advantageous as to blow settling out of the water. You advocate saving a Great Scientist, if you mean that literally then we have to factor in some wastage from it not being used. Not sure you mean that though.

If you are saving it, it isn't for long. Both CoL and Drama open up Philosphy, which is well over 1000 beakers and an incredible trade chip on all difficulties. If you're going bur (then getting CoL early) then it's a good choice...but if you need the to spam specs you might get drama earlier than usual too.

Philo is >1000 . Even with a library + academy it would take 95 (!) turns of settled GS to catch 1000 , but Philo is MORE than that! But not only does the bulb get you that research up front, it offers better trades (more beakers), faster access to liberalism/better civics/multipliers such as universities/oxford (if going for the long-term games), and faster access to military techs which can change the game.

Settling just doesn't make sense unless you're in financial trouble after a tremendously successful early rush or something. Even then, you'd prefer to bulb + trade if you had someone to trade with.

 

[Iranon]

Also, the AI undervalues gold. It thinks 1000 for a 1000 tech is fair, disregarding that the one selling the tech doesn't lose it (500 would be a fair price, maybe adjusted a bit for different modifiers depending on era).

As TMIT pointed out, settling takes quite long to recover the head start from one-off bonuses. If you settle, don't do so for efficiency but because there are overriding reasons to play from behind.
I find it easier to leech from AIs than to keep them from leeching off me while maintaining stable diplomacy, and am generally more comfortable with a slow global tech pace... settling just suits me better even when it's theoretically inefficient.

 

[noto2]

Yes, that's the main reason I ever settle - is the rubber band effect. In civ, you get a research discount for every civ you know who already has that tech. When you're behind a few AI's this really adds up. If you settle a lot in the early game and other AI's bulb and get 3/4 techs ahead of you, you can catch up very easily. You can then bulb/golden age later on in the game when it matters more, or you can go for representation. The thing is, if you bulb right away you do get to make trades to try to stay on top, but you're also giving free research discounts to the AI's, and eventually you run out of bulbing power and then later in the game the AI catches up with you. Especially on higher difficulties, the AI gets more bonuses as the game gets later. So I either try to win the game before the industrial age, or I want to have some GP saved up for golden ages or representation then.

 

[cseanny]

The thing is, if you bulb right away you do get to make trades to try to stay on top, but you're also giving free research discounts to the AI's, and eventually you run out of bulbing power and then later in the game the AI catches up with you. Especially on higher difficulties, the AI gets more bonuses as the game gets later. So I either try to win the game before the industrial age, or I want to have some GP saved up for golden ages or representation then.

I disagree. Imo its more beneficial to trade/bulb your way to a powerful renaissance tech asap which you translate into an enormous advantage by taking over the world. I play Deity/Immortal and win regularly at both. The game, if not already won with Cuirs/Cav/or Rifles is effectively over at this point regardless because you have put yourself in a much stronger position quite a bit earlier.

 

[noto2]

I hear this from immortal/deity players all the time. Sure, this works in retail BTS on a pangea map. I play with Kmod where the AI is so far improved, no one can win on deity and almost no one can win on immortal. It's effectively 2 difficulty levels higher than retail, 1 diff level higher than better BTS. Retail BTS AI is so horrible but then playing on deity where it gets ridiculous handycaps really distorts the game. Certain strategies are useless and others are a necessity, you end up having to do the same thing every game (drafting rifles).
But imagine you weren't playing against someone who got handicap cheats but instead was a good opponent because they were smart, like in MP or in Kmod. I don't think the average person on these boards plays immortal/deity. If you're still playing retail BTS I urge you to try one of the AI improving mods - JUST ONE GAME. Seriously just try one game. Obviously I'd recommend Kmod. Better BTS is the bare minimum you would be playing with, it doesn't alter gameplay at all, just improves the AI. Kmod is as close as you can get to multiplayer without actually playing multiplayer. Then again, some people like the fact that the AI is stupid and they can beat it on deity through manipulation and using tricks. To each, their own.

 

[noto2]

I suppose if we're talking about retail BTS (as in, programmed in 2004 or something) and immortal/deity then yes, it goes like this: Never bother building a wonder unless it's the GLH or Glibrary, bulb all your GP to liberalism, trade, trade, trade, get nationalism, trade, trade, get rifles, get cannons, draft massive rifle armies, try to win the game with a rifle/cannon war. That's the only viable strategy. All other strategies are rendered basically useless. Yawn.

 

[TheWilltoAct]

^ That's why it's fun to hear Iranon's perspective, and that of other outliers.

 

[Doshin]

I suppose if we're talking about retail BTS (as in, programmed in 2004 or something) and immortal/deity then yes, it goes like this: Never bother building a wonder unless it's the GLH or Glibrary, bulb all your GP to liberalism, trade, trade, trade, get nationalism, trade, trade, get rifles, get cannons, draft massive rifle armies, try to win the game with a rifle/cannon war. That's the only viable strategy. All other strategies are rendered basically useless. Yawn.

Posts like this irritate me no end. Lower difficulty players ask the forum how to improve their game, and higher difficulty players offer advice in as succinct a form as Civ IV will allow. Libing a powerful military tech with the help of bulbs is the easiest way to win on Immortal or Deity, but it is only easy to the extent that an individual game allows you to pursue this option. If you try to play this way in every game, you are totally dependent upon a whole host of random factors going your way: no Marble? Your chance of winning TGL plummets. Shaka and Ragnar are your neighbors? Your chance of getting an early DOW skyrockets. Boxed in? How are you going to reach Lib first, or whip/draft a sufficiently powerful army? Mansa Musa is building research with eight cities? You need to stop him. You have Horses and Lincoln is your neighbor? Take advantage now and build some Horse Archers. I could go on. It is one thing to offer an archetypal strategy, another thing to actually play the map.

BTW, I like the standard BTS AI, even though it is flawed and makes a large number of bad decisions. You should watch Soren Johnson's talk regarding Civ IV AI (http://www.youtube.com/watch?v=IJcuQQ1eWWI), in which he distinguishes between "good" and "fun" AI personality types. Religion is the most obvious manifestation of the difference: the AI actually cares about their neighbors' religion(s), and will declare war on this basis alone. Human players don't care, and will be as opportunistic as possible. This isn't to denigrate Kmod (or any others for that matter). I simply want to point out that many people enjoy the leader personalities and manipulating their biases.

 

[Manco Capac]

I hear this from immortal/deity players all the time. Sure, this works in retail BTS on a pangea map. I play with Kmod where the AI is so far improved, no one can win on deity and almost no one can win on immortal. It's effectively 2 difficulty levels higher than retail, 1 diff level higher than better BTS. Retail BTS AI is so horrible but then playing on deity where it gets ridiculous handycaps really distorts the game. Certain strategies are useless and others are a necessity, you end up having to do the same thing every game (drafting rifles).
But imagine you weren't playing against someone who got handicap cheats but instead was a good opponent because they were smart, like in MP or in Kmod. I don't think the average person on these boards plays immortal/deity. If you're still playing retail BTS I urge you to try one of the AI improving mods - JUST ONE GAME. Seriously just try one game. Obviously I'd recommend Kmod. Better BTS is the bare minimum you would be playing with, it doesn't alter gameplay at all, just improves the AI. Kmod is as close as you can get to multiplayer without actually playing multiplayer. Then again, some people like the fact that the AI is stupid and they can beat it on deity through manipulation and using tricks. To each, their own.

K-Mod is playing with psychopathic AI's that can handle multiple wars at once.
Most people won't feel good in this environment. And there is slight changes in game mechanics like culture global thing that even dismissed one of the toppest player here.

In K-mod there is no deity unless Karadoc finally decides to give slack on AI bonuses. AI bonuses are for asinine AI's to mask their idiocy; improved AI's with the same bonuses are simply skyrocketing insanity.

Finally, k-mod is about changing AI persona.
But there is another aspect in retail BTS: its own economy.
Improving its own economical skills is independent of AI's surrounding you (except some trades and feeling secured (which doesn't happen at all in K-mod IIRC)).

 

[noto2]

There are good things and bad things about BTS retail and Kmod or BBTS mod. I just don't like trying to make a game more difficult by giving the AI handicaps, as opposed to making it play more competitively. Giving the AI handicaps distorts the game. In monarch difficulty you can win the game with a small empire, teching to space, and still manage to defend yourself. On deity that is impossible. You want to talk about roleplaying? Civ4 has so many aspects to it. You should be able to play a game with a religious economy, or by turtling to space with just 6 cities, by settling GP in a super capital and wonderspamming, OR by being a warmonger. You shouldn't have to draft rifles and mass cannons to win, EVERY. SINGLE. GAME. AS. THE. ONE. AND. ONLY. STRATEGY. THAT. CAN. POSSIBLY. WIN.
If it weren't for Kmod, or modding in general, I would have stopped playing BTS back in 2008 when I won my first immortal game and lost 20 deity games in a row. I just didn't care anymore, the game wasn't fun because many strategies were completely ruled out, you had to do the same thing every time.

But I have gone off topic. I was saying that settling can be useful, especially early game, and especially if you plan on running representation, which is more powerful with a smaller empire that has good land, (quality land but not as much quantity) with a lot of food. This tangent came about when other said that settling is never worthwhile and you must always bulb. That's nonsense. In emperor and below you certaintly don't have to always bulb. And I suspect most people play emp or lower (probably more than 90%). If you want more of a challenge, rather than switch to immortal difficulty, I would STRONGLY recommend you simply download a mod that makes the AI play smarter, such as BBTS or Kmod.

 

[noto2]

Kmod doesn't change AI personality completely, the AI still roleplays quite a bit, but it is more competitive. If you are religious buddies with Isabella, she still loves you in Kmod, the only difference is if she has conquered the rest of the world and you are the only nation left standing between her and victory, yes, she will declare war on you in Kmod. How can you not find that more interesting? It means in Kmod you can't win simply by looking at a spreadsheet (who attacks at friendly), you have to balance power and not let an AI get 6 vassals.

 

[TheMeInTeam]

Soren Johnson's talk regarding Civ IV AI (http://www.youtube.com/watch?v=IJcuQQ1eWWI), in which he distinguishes between "good" and "fun" AI personality types.

Sorry, but Soren's opinion on this completely sucks. In a balanced game, an AI that plays well *is* fun. You don't create an idiotic AI then heap bonuses on it and pidgeon hole the player and call that "fun". That's called "lazy". Nice try though, Soren.

kmod AI is a lot better than standard and I would use it nearly exclusively if it become popular here in S&T. It runs BBAI in there with some improvements.

Stock AI: 1 Victory condition it can plan (and bts only), and it does it poorly.

Kmod/BBAI: AI can and will attempt to win using any of the major victory conditions, depending on flavor and position.

In other words, in one game you have actual opponents, and in the other you have pathetic walls of bonuses that don't actually try to win. I'm supposed to buy that the latter is "more fun"? How so? Just because deity in bts doesn't completely drain away viable strategies doesn't mean that it's ideal. Stronger-playing AI is always better, with the only caveat being that doing so in a poorly designed game will expose imbalances.

Imbalances, such as speed teching renaissance. Prior to rifles/cavarly/cuirassers, there exists exactly 0 units that are not countered or defended by a previous era unit (excepting the very first units, because there is nothing before them at all). Even the questionable noobiphants are cost-ineffective against spears.

But failaxis, instead of addressing this while tweaking the AI, instead decided to simply hike AI bonuses, further emphasizing the importance of a single strat. They have a long history of taking things away rather than tweaking the AI to play better. Look at their disgrace behavior with overflow ---> bug in their FINAL patch (where they even admitted an unintended consequence...then NEVER FIXED IT!). How about when they bugged espionage spread culture just to switch it back?

When it comes to game quality and post-release support, Firaxis = EA. Soren's is one of a list of names that have contributed to that reality.

There were better user interfaces by 1995 than either of the last 2 civ games, too, but that's another matter entirely and kmod even helps fix the broken controls that failaxis never managed .

 

[noto2]

^^amen