How to Tech Faster?

[GAAAARG]

Hey! I'm a new civfanatics user and I'm fairly new to Civ IV, too. I've been playing for a month or two now and I'm trying my first game at Monarch difficulty.

I've gotten better at teching since I started playing - in my current game, I just completed Assembly Line and started on Railroad in the very early 1700s (Ancient start, Normal speed). Here are some of the things I'm doing:

Setting the research slider as high as I can without running a deficit, except immediately before I expect to spend serious cash modernizing my military.
Running scientists as my default specialists unless I need the city in question to generate a GP of a different type.
Building libraries/universities/observatories/monasteries in cities working wealth-heavy tiles or running one or more scientists. Saving a national wonder spot in my best science city for Oxford.
Winning the lib race.
Prioritizing science-friendly wonders (Great Library, Sankore when I've done a good job spreading my religion).
Running science-friendly civics, especially Representation and Caste System in the midgame.

That's all well and good, but it seems that some more experienced players are able to tech WAY faster than I am - I've seen posts about diplomatic victories in the 1500s, for example! What should I be doing to tech faster?

 

[Seraiel]

There are also diplomatic victories in the 12xx

To tech faster, you have to do less of all the intense stuff, and do more of the basic stuff.

Build many Cottages in your capital and grow them.
Run Burocracy.
Found an Academy there as early as possible.

That alone should get you 150 BPT at 1 AD.

Monestaries are almost useless up to my taste, because when you tech really fast, you have barely time to build a Library, a Forge, an AP-Temple and a University, and monestaries lose their bonus (that isn't even that big) .

Because you try to build so many Wonders, you're probably also not expanding enough, either peacefully, or better, through war. Many cities means a lot of research, once you can pay for them (somewhere between Currency and Corporations) .

Sankore imo is also a useless Wonder.

If you run Caste System, combine it with Pacifism to create as many GPs for bulbing as possible.

That's all I can think of atm.

 

[GAAAARG]

Seraiel - thanks.

So this gets at the more fundamental difficulty that I (and many other newer players, I'd assume) have when playing. If I'm building lots of cottages, how do I get the food I need to run specialists and generate GP? Or do really fast techers prefer to run a CE?

I ask because even when settling near food resources, I still find myself needing lots of farms to keep up with the AI in city population.

The monastery bonus is better with Sankore, and vice-versa. But I'll try it your way next time and see where that gets me.

 

[Seraiel]

Very simple:

If you run a Cottage Economy, you (generally spoken) don't work Scientists on top. To get the necessary GPs, you run 2 Scientists in your Capital (that should normally have enough food for those 14 turns or what) , and use him to create an Academy. In the meantime, you found a food-rich city that you utilize as a GP-Farm when the time is right, meaining, when you either reach a critical moment, or, morely, when the NE is set up and you can start a GA. In that GA, you switch towards Caste + Pacifism and use all cities that are big enough to get out a GP simultaniously as GP-Farms and create 3-5 GPs (mostly depending on if you got the MoM, one of the very few Wonders that are really worth going for) .

This is a strategy for a Space-Race-Victory. If your tech-aims are lower, like i. e. reach Military Tradition as fast as possible, you'll only get out about 2-4 GPs overall anyhow, and then you want those to be GMs to upgrade your Horse Archers.

Regarding Economy itself: The best you can do, is work the land best, meaining, if a city has got Hills + Food, build Wealth / Research, if it has got really much Food, work Specialists, and if you got Rivers, work Cottages.

In general, your cities don't have to be big to produce amazing research. You have to reach Currency as early as possible, to beg Gold every 25 turns, to sell techs when AIs have money (i. e. often after a World Wonder got built) , you have to build Failgold if you got Marble / Stone, Missionaries when you got a Shrine, and the most comes through getting AIs on friendly or having Mansa in the game, so through trade. On Deity i. e. , I quite often trade about 5 techs for the 1 I teched myself.

Good values for Research are 200 BPT at 1 AD, when you should have about 7 cities. This is achievable with Civil-Service + an Academy in the Capital, the other cities can be really small, only that city has to be as large as Happiness and Healthiness cap allow.

Fouding cities in good locations, is a skill that I regard as almost the most important from all, if wanting to advance to the really high difficulties. Cities with much Food will always develop faster then cities with little Food, so that's your main criterium to look for. Rivers are nice, Resources are amazing, but what gets your cities really going, is double strong Food. Single is enough, but double gives you the opportunity to whip aggressively and maybe even run some Specialists on top, so look for those locations. 2nd to 3rd biggest factor (after Food and maybe resources) are btw. trees.

Hth, ask more directly to get more specific advice.

S.

 

[pomthom]

Do you tech-trade a lot?

A major breakthrough I experienced regarding tech'ing is going for techs that the AIs are less likely to research, or at least avoiding techs that all AIs tech early. The trading value of these techs allow you to virtually multiply the beakers you've invested in them.

Some "pre-Lib" examples on the top of my head (this really depends on the research speed of your opponents -- ie the difficulty level mostly. If you're way ahead this doesn't apply obviously. Neither does it if your attempting some sort of rush or bee-line):

Preferably avoid: Iron Working, Math, Monarchy, Monotheism, Calendar, Metal Casting, Construction, Feudalism, Guilds...
Preferably favor: Aesthetics, Horseback Riding, Literature, Code of Laws, Currency, Music, Compass, Philosophy (not to trade lightly because Lib-prereq), Nationalism...​

My examples might be arguable, I'll let more experienced players correct me if necessary. Also it's not always worth going for 1 of these techs if it doesn't fit at all in your macro strategy.

Be carfeul not to get too obsessed by tech speed: you can always go faster, it's drawing the line where reasearch is not boosted to the detriment of expansion, military, growth (etc) that's difficult!

EDIT: cottage economy VS scpecialist economy VS god-knows economy is an obsolete vision. To get a solid economy going you need a little bit of everything (see Seraiel's post about cottaged Bureau capitole, GP farm, building wealth/research, building failgold and whatnot)

 

[GAAAARG]

I should've mentioned that I'm unfortunately not playing with BtS. Soon, I hope. That means no MoM, no civic-switching during a GA, etc.

Thanks for all the advice. A few things I'd like to clarify:

• I don't know what an NE is.
• Seraiel mentioned begging gold every 25 turns - this is something I haven't been doing. Should I infer that the AI is only programmed to grant a request every 25 turns?
• Re: tech trading: I understand researching techs the AI is unlikely to pursue and shopping those for AI-researched techs. I haven't been getting anything close to 5:1 returns, though, mostly because the AI never has five techs that I haven't researched. Is this a strategy that can be employed at Monarch when the AI isn't teching as fast as it would be on, say, Immortal?
• Also on tech trading, Seraiel singled out Mansa specifically and I don't know why.

 

[pomthom]

I should've mentioned that I'm unfortunately not playing with BtS. Soon, I hope. That means no MoM, no civic-switching during a GA, etc.

Thanks for all the advice. A few things I'd like to clarify:

• I don't know what an NE is.
• Seraiel mentioned begging gold every 25 turns - this is something I haven't been doing. Should I infer that the AI is only programmed to grant a request every 25 turns?
• Re: tech trading: I understand researching techs the AI is unlikely to pursue and shopping those for AI-researched techs. I haven't been getting anything close to 5:1 returns, though, mostly because the AI never has five techs that I haven't researched. Is this a strategy that can be employed at Monarch when the AI isn't teching as fast as it would be on, say, Immortal?
• Also on tech trading, Seraiel singled out Mansa specifically and I don't know why.

1. Get BTS!! (are you playing Warlords or Vanilla?)
2. NE = The National Epic (on the same topic HE = Heroic Epic)
3. Begging gold can be done with Pleased or Friendly AIs only. You can beg up to 50/60 gold (or 3/4 gold per turn, I think), and you can do so only every 25 turns or so (I don't know if that count is exact or not but I'll trust Seraiel on that one). This can be very useful in the early game where 50 gold means a lot. Begging also has an other very useful layer: it confers a 10-turn peace treaty with the AI you begged from (this is also the case when you give in to an AI tribute demand). So, for example, if you declare war on someone, you can beg from Pleased AIs so that they can't be bribed against you or accept a peace-vassal from your enemy (for Warlords or BTS).
4. Regarding tech trades: if you tech fast enough so that you don't need to tech-trade means you don't need to tech any faster and yes diff level has something to do with it. Seraiel mentioned Mansa because he's the best trading partner (other AIs will wait until [x]% of all AIs know a tech before offering it up for trade, or need to be Cautious or better. Not Mansa. If you are unfamiliar with this type of parameters, I suggest you take a peep at this thread)

EDIT: and Mansa is a very good tech'er

 

[Seraiel]

All questions answered before I can post, really good

Just a small thing: There's a random-chance every turn that AI will "forget" that you begged from them. After 25 turns, the chance for AI having forgotten the last Begging is 80%. How much you can beg is speed-dependent, generally you can beg less the faster the speed. I usually try amounts of 30g on Quick, 50g on Normal, 70g on Epic and 90g on Marathon. If I didn't beg for a long time, I up those amounts by 10-30g, so roughly 30%, which is about the maximum one can get, but it's risky and the returns would probably be the same if simply begging the amounts I specified.

To make a successful demand, your military must be stronger than the one of the AI you demand from. Denied demands give between -1 and -2 on relations, but it's a good way to fasten up the rush-buying of units near the end of the round i. e.

A special case is Gandhi. You can deny everything to him, and demand whenever you want, without getting Diplo-hits, because he doesn't care. He follows the general mechanics of random chance and military-quotient though.

Hth,

Seraiel

 

[J@ckRB]

Mansa Musa is a very good techer because he trades a lot, always seems to research at a decent rate comparable to the other civs no matter what land. Also, he really helps in that he will trade monopoly techs (a.k.a techs only he has - all other civs want to wait until at least another civ or 2 (depending on the amount of civs) has also researched the technology). Third thing, his attitude threshold to stop trading is at Furious, he's essentially always willing to trade.

And that, is why Mansa is a brilliant techer.

P.S Gandhi doesn't care about demands and the such, but he does have a higher negative diplo hit if you declare on his friends (-2 instead of -1)

 

[MarigoldRan]

At Immortal or Diety, the answer is "you don't," at least in the early game.

You can either expand or tech, but you can't do both even if you start with one or two gold mines. Since expansion is more important, the idea is to fall very behind in tech, trade/research for a couple of essential ones (like writing or Alphabet), and then kill an AI and take his tech.

 

[Noble Zarkon]

Setting the research slider as high as I can without running a deficit

This is a mistake many newbies make, you want to run your research at either 100% or 0% (known as binary researching. This way you get the maximum out of your multipliers as well as having flexibility in choosing what to research e.g. 10 turns at 0% to build up enough gold then turns of researching, you don't have to choose what you will research until the initial 10 turns are gone and someone may trade something etc.

 

[bernds]

3. Begging gold can be done with Pleased or Friendly AIs only. You can beg up to 50/60 gold (or 3/4 gold per turn, I think), and you can do so only every 25 turns or so (I don't know if that count is exact or not but I'll trust Seraiel on that one).

I think there's a random element involved so that there's a chance every turn for them to forget the previous begging request, and in practice it works out to something like the 25 turns that were recommended. Also, how much you can ask for depends on how far into the game you are - start off with 50 or 60, and as the game progresses they should accept larger requests (150 or so).

 

[Seraiel]

This is a mistake many newbies make, you want to run your research at either 100% or 0% (known as binary researching. This way you get the maximum out of your multipliers as well as having flexibility in choosing what to research e.g. 10 turns at 0% to build up enough gold then turns of researching, you don't have to choose what you will research until the initial 10 turns are gone and someone may trade something etc.

If you have Gold-Multiplier-Buildings and Research-Multiplier-Buildings, split-Research like 70% is better than 100/0, but not running a deficit is a mistake if wanting to tech fast.

Usually to tech really fast, one wants to stay at 100% all the time and fill the deficit from selling techs and all the other mentioned tactics.

 

[DMOC]

Hey! I'm a new civfanatics user and I'm fairly new to Civ IV, too. I've been playing for a month or two now and I'm trying my first game at Monarch difficulty.

I've gotten better at teching since I started playing - in my current game, I just completed Assembly Line and started on Railroad in the very early 1700s (Ancient start, Normal speed). Here are some of the things I'm doing:

Setting the research slider as high as I can without running a deficit, except immediately before I expect to spend serious cash modernizing my military.
Running scientists as my default specialists unless I need the city in question to generate a GP of a different type.
Building libraries/universities/observatories/monasteries in cities working wealth-heavy tiles or running one or more scientists. Saving a national wonder spot in my best science city for Oxford.
Winning the lib race.
Prioritizing science-friendly wonders (Great Library, Sankore when I've done a good job spreading my religion).
Running science-friendly civics, especially Representation and Caste System in the midgame.

That's all well and good, but it seems that some more experienced players are able to tech WAY faster than I am - I've seen posts about diplomatic victories in the 1500s, for example! What should I be doing to tech faster?

Here are some of the things I would think about when considering tech speed:

1. Don't try to compare across difficulty levels; On deity, the AI techs the fastest, so it also means that a skilled human player can use that to his advantage by trading, and thus that speeds up the whole game.
2. About your point on setting research sliders high except when upgrading troops, I would advise against simply turning off cash to modernize your military, unless you are seriously production-strapped or time-sensitive. Typically, if I reach a certain age, I try to use whatever's available at that time (the whip, the draft, or standard prodution) to build the army. I'll upgrade a few troops (a few!) but never my whole army, since I think it's far more important to save the gold for technology. Use the obsolete troops for city garrisons.
3. I noticed you did not mention growing your cities in one of the things you're doing. I was playing a deity game earlier and I was using Hereditary Rule to my advantage by having a size 15+ capital just past 1 AD, and most of the citizens were working cottages.
4. If all else fails, why not try and read a walkthrogh about some of our games? I've got a couple that include fast teching (but they're probably normal for deity anyway) in my signature. My favorite is my Sury II game.

A final note: I did not read the previous posts (after yours) in this thread before making this post, so if there's something I said that's repeated, then it's probably important.

 

[DMOC]

I should've mentioned that I'm unfortunately not playing with BtS. Soon, I hope. That means no MoM, no civic-switching during a GA, etc.

Thanks for all the advice. A few things I'd like to clarify:

• Seraiel mentioned begging gold every 25 turns - this is something I haven't been doing. Should I infer that the AI is only programmed to grant a request every 25 turns?

3. Begging gold can be done with Pleased or Friendly AIs only. You can beg up to 50/60 gold (or 3/4 gold per turn, I think), and you can do so only every 25 turns or so (I don't know if that count is exact or not but I'll trust Seraiel on that one). This can be very useful in the early game where 50 gold means a lot. Begging also has an other very useful layer: it confers a 10-turn peace treaty with the AI you begged from (this is also the case when you give in to an AI tribute demand). So, for example, if you declare war on someone, you can beg from Pleased AIs so that they can't be bribed against you or accept a peace-vassal from your enemy (for Warlords or BTS).

All questions answered before I can post, really good

Just a small thing: There's a random-chance every turn that AI will "forget" that you begged from them. After 25 turns, the chance for AI having forgotten the last Begging is 80%. How much you can beg is speed-dependent, generally you can beg less the faster the speed. I usually try amounts of 30g on Quick, 50g on Normal, 70g on Epic and 90g on Marathon. If I didn't beg for a long time, I up those amounts by 10-30g, so roughly 30%, which is about the maximum one can get, but it's risky and the returns would probably be the same if simply begging the amounts I specified.

To make a successful demand, your military must be stronger than the one of the AI you demand from. Denied demands give between -1 and -2 on relations, but it's a good way to fasten up the rush-buying of units near the end of the round i. e.

A special case is Gandhi. You can deny everything to him, and demand whenever you want, without getting Diplo-hits, because he doesn't care. He follows the general mechanics of random chance and military-quotient though.

Hth,

Seraiel

I really don't think GAAAARG should be focusing on begging tactics to increase tech speed. At monarch difficulty level, there are other, far more important factors to teching speed than the amount of gold earned from begging (unless the map is an insane 32 civ world or something, where begging can be done perhaps every turn!). First, focus on expansion/development, how to effectively conduct military, and tech trading.

 

[pomthom]

I really don't think GAAAARG should be focusing on begging tactics to increase tech speed. At monarch difficulty level, there are other, far more important factors to teching speed than the amount of gold earned from begging (unless the map is an insane 32 civ world or something, where begging can be done perhaps every turn!). First, focus on expansion/development, how to effectively conduct military, and tech trading.

I agree (was just answering a question )

 

[Seraiel]

Begging tactics are important. 60g * 6 Civs = 360g = maybe 6 turns of research at 100% with deficit = about 1/4th of the research financed until the next begging can happen. (All standard values of 6-7 cities on a Standard / Normal map with no Courthouses) .

I agree with you DMOC that I also didn't like to see this in CIV, because it's actually not what fits in a strategy game imo, but AI is even worse, sometimes demanding like all of one's gold on T2 after meeting them because they're annoyed from one having traded with their worst enemy of which one could not even be aware. Let's be honest, it's just a game .

 

[pomthom]

On the begging subject (sorry for the off-topic), if you beg something when the AI actually didn't forget you already begged from it (so it refuses), does it affect the probability of forgetting on the following turns?

By that I mean: can I shamelessly beg from the same AI every turn until it accepts or does my insistance harm me? Other question: is the probablility of forgetting a beg (or accepting one...) completely decorrelated from the amount you ask? If I ask 1 Gold, do I have a greater chance of it being granted than if I ask 50 (which is still reasonable) ?

 

[Seraiel]

As far as I know, you can beg every turn and the chance is 1/25 or something like that that AI will have forgotten that you begged it, and also afaik, the AI only decides if it is willing to give in to a beg that is below the threshold it would be willing to give, or deny it in any case if it hasn't forgotten about the last beg.

 

[Sun Tzu Wu]

If you have Gold-Multiplier-Buildings and Research-Multiplier-Buildings, split-Research like 70% is better than 100/0, but not running a deficit is a mistake if wanting to tech fast.

Usually to tech really fast, one wants to stay at 100% all the time and fill the deficit from selling techs and all the other mentioned tactics.

I disagree with this idea of splitting Commerce into Research and Wealth on the same turn. I agree with NobleZarkon and will try to describe the merits of binary research in a different way ...

Binary research (either 00% or 100%) is superior for three reasons:

1) At 00% Research, one always gets a free beaker/turn even when specialists add some amount of beakers/turn.

2) Unless, the commerce rate is evenly divisable by the current Research slider, one loses a commerce/turn due to round-off error that doesn't get converted to either research or wealth.

3) With the exception of specialists generating raw research, one can defer committing to a research path.

Points 1 & 2 are critical at the beginning of the game when commerce rates are usually less than 100 Cpt. One really can't afford to lose 1 Bpt every turn and an additional 1 Bpt for every 00% Research turn, when 100% Research is not sustainable.

Point #3 is far more important at difficulty levels not too far below Deity, when one can easily keep ahead of the AI's research and has to work hard to feed technologies to the AI to gain a return in them researching parallel technology paths. At the lower difficulty levels (definitely at Warlord and maybe Prince), dragging the AIs along technologically hoping they can research some techonology that one doesn't know yet becomes less and less efficient as the difficulty level becomes lower. At some lower dificulty level threshold, one is better off teching beyond the AIs and not helping them at all. Point #3 may not always apply to Deity level, since one might be having difficulty keeping up with the Deity AIs as opposed to waiting for them to complete a desired technology.

Sun Tzu Wu