When is it better to buy techs than researching them yourself?

[Iskandar Reza]

As in Aeson's game, thread here, he bought most of his techs. But I'm sure that researching it yourself is good too. The question is, at what point?

[trickey]

Anything above regent (or monarch for better players), you cant possible dream of keeping up with the A.I. in the tech race as far as researching your own techs goes.. So most players go with very minimal research. Although, below regent, if you dont do the reasearch, the A.I. sure a hell won't.

[Catt]

I just finished my 3rd game under the new 1.29f regime. Emperor, small map, all random, 6 civs (including me). I ended up with Gandhi (and for the first time ever playing as the Indians, I didn't have access to horses - Yea War Elephants!) - opponents were Iroquois, Aztecs, Japanese (these three all direct neighbors), Chinese and Americans. Map turned out to be pangea, maximum water. A more complete description of the game conditions appears in the General Forum under a thred titled something to the effect of "Anyone finished a 1.29f game yet?"

I actually researched most of my own technologies, occasionally traded for techs, but almost never beat any techs out of anyone. In fact, for a variety of reasons, I had a grand total of 2 wars the entire game, and each lasted no more than 7 turns. It was a true builder's game, something I've never really had on Emperor.

My first war happened early in the age of cavalry. China had already been eliminated, and Japan was on the ropes. As soon as my last deal with Japan ran its course, I declared war and took a meaningless city with a cavalry and a war elephant. Welcome Golden Age. Five turns later, I made peace. My second and last war took place just before Japan's demise, and after only a very short interval between the conclusion of my first war with Japan. For some unknown reason Japan demanded a tech from me and declared war when I didn't offer it up. After 5 - 7 turns of Japanese cavalry approaching my cities, beign bombarded and then killed, I made peace. The other nations wiped Japan out soon thereafter.

I was consistently the smallest nation on the map (territory, not always on pop basis) other than isolated periods when someone like Japan was close to doom. The Iroquois and Aztecs ended up conquering all of the remainder of the world, and each of them was probably twice as large as me. Nonetheless, because of my compact and well-developed empire, I was able to maintain near tech parity all the way through the modern era. At the end of the game (I won via spaceship), I had researched The Laser and then launched, and the AI had researched both Stealth and Recycling (which I ignored). I was probably about 2 techs "behind," but I went straight for the win - at least the Ai wasn't building the Genetics wonders (I think I remember from the chat that Soren indicated that the AI would be better about launching rather than dithering).

After about Motorized Transport, I did not have enough gold, a luxury monopoly, or a tech advantage that I could use for tech trading. I researched on my own, generally following on the slip stream of the AI (the cheaper techs due to devaluation), and through micromanagament and pre-builds was able to win on a research basis.

Long story made short is: I think 1.29f has / will really change the general consensus about the viability of research, even above Monarch.

Catt

[Arrian]

Trickey, methinks you doth exaggerate.

Most of my experience is on Monarch, and I've played some Emperor. I find that it's usually best to buy tech from the AI (or beat it out of them) throughout the ancient age. In order to begin your own research you need marketplaces, libraries and courthouses. In order to get ahead, you need universities and banks. Of course, getting your Forbidden Palace up and running provides a big boost, particularly if it's far away and has good cities around it.

I usually switch research on in the middle ages. When exactly depends on how many civs I have contact with and whether or not I have the Great Library. I often attempt to conquer my whole continent - and achieving this means I'm alone and probably have a good FP location. Therefore, max research. In my most recent game, however, things were different:

I was Rome, and although I select "continents" I got a Pangea with all 8 civs on it. Because of my local geography, I simply had no choice but to attack Egypt early on. I took Thebes, and left Cleo w/1 city (damn thing was on a hill). I spent the rest of the ancient era building my empire, buying tech, and researching toward Monarchy (w/0% science and 1 scientist). Upon discovering Monarchy, I switched, and promptly unleashed 10-15 legionaries with some horsemen backup on Japan. Kyoto had the Great Libary... otherwise Japan was garbage. My golden age finished off my FP (center of my empire at the time, 2 cities over from Rome) and the Hanging Gardens. I now had the Great Library, so I continued with the 1 scientist thing and used all available cash to rush improvements and save for upgrades (mmmm, knights). I milked the GL as long as I could. Upon losing it, I cranked up my own research and have lead the rest of the way (staying a Monarchy until the dawn of the industrial age, when I switched to Demo). It's been a fun game, actually. My palace is now up in former Zululand, and there are only 2 AIs left in the game. It's pretty much all over. I could wipe out England at any time and win via domination.

Anyway, I was able to outresearch the AI using Monarchy as a government with a mediocre palace/FP setup for a while, with the AI's using republic. The GL was huge because it allowed me to pump every cent I earned back into my economy. The game is on ice now because I now have a very good Palace/FP setup and I'm a democracy with a 4-5 tech lead.

-Arrian

[Panzer]

In my last game (Regent, Small, Pangea, 10 civs) I was stuck on the tip of the main continent. I build 6 cities (5 1/2 actually, one city was on a small tundra peninsula). After a few dozen turns it was obvious that I wasn't going to be able to keep up with the other civs (Damn techtrading bast*rds !!). So I set research at zero and bought every Ancient tech, sometimes trading them for other techs from other backwards civs. Within a few thousand years the most advanced Civs had reached the Middle Ages and after another few hundred years I joined them. I achieved tech parity in the early Middle Ages and I had finished building libraries and marketplaces in most of my cities. I cranked up my research and remained with the top three for the rest of the Middle Ages. I achieved tech superiority in the Industrial Age and won by Domination in the Modern Age (Scientific vicory was disabled).

Moral of this story? Buy/conquer your way through the Ancient Age, achieve tech parity in the Middle Ages and achieve tech superiority in the early Industrial Age .

[Zachriel]

In the early ages, I set science at 10% or so and buy most of my techs. But even then, I can usually sell a new discovery every once in a while -- and when I do, I usually get rich quick. Once I have the Republic, I increase science and can usually hold my own, but it is still a good idea to buy most non-essential techs only once they become obsolete, concentrating research on the strategic choices.

But there are many exceptions, of course. In GOTM9, an Emperor level game, Cleopatra set science at 70% with luxuries at 30%. The map was archipelago and Egypt had to discover mapmaking (while completing the Great Pyramids).



** WARNING ** GOTM9 SPOILER **
http://www.zachriel.com/gotm9/bc1000-Pyramids.htm

[OPD]

As soon as the game starts I set science to a minimum and then buy only military techs when needed. As a result I fall stupidly far behind the AI but even when they have the ability ti build pikemen the AI cities still aeem to have spearmen as defence.

I used to concentrate on infastructure and buy almost all techs but with the new patch the AI charges too much and so I usually pick one civ and repeatedly beat the latest advances out of them.

Only once a FP is built and libs/markets are bought does research properly begin.

The great thing about not researching is that you can buy imporvements and build up infrastructure to be able to reserach alot better later on.

One Tactic which is great (I've only pulled this off once) is to not research and just buy millitary techs untill you can buy chivalry and then you capture whatever city has the GL thus getting a stupid amount of techs.

[Catt]

So - please share insights all. I really think that 1.29f has created a "paradigm shift" [ - boy, that's embarassing to use that phrase ]. My experience with 1.29f is limited (3 games on small or tiny maps), but I sense from the chatter on the boards and my own experiences that researching tech, even at Emperor, is a viable strat.

Experiences to share now that 1.29f has been out more than a few days?

Catt

[Cort Haus]

Originally posted by Catt
I was consistently the smallest nation on the map (territory, not always on pop basis) other than isolated periods when someone like Japan was close to doom. The Iroquois and Aztecs ended up conquering all of the remainder of the world, and each of them was probably twice as large as me. Nonetheless, because of my compact and well-developed empire, I was able to maintain near tech parity all the way through the modern era
Catt

Nice tale, Catt ( ), and it's great to see the Peace Dividend of a "compact and well-developed empire" paying off on Emperor!

Question : how did the luxury trade deals go with the big boys? In my one emp game I was tiny at first but was able to catch up in tech by selling lux's for 2-3 techs each, sometimes with a lux thrown in too. I haven't reproduced this in other games yet but I'm intrigued by the trading possibilities of small civs. However, a recent post from Txurce on the Peace Dividend thread suggested that as his civ's power increased (but not the size) the deals became less favourable - implying that size *and* power go into the formula.

[Cort Haus]

Originally posted by Arrian
In order to begin your own research you need marketplaces, libraries and courthouses

Or a good river start and an industrial civ, when you can research effectively off the bat.

Even without loadsa rivers, I think research is more on the cards in 1.29 than before. Picking the right branches & techs is the key, as is making the most of your starting techs, an awareness of the starting techs of rivals, contact with the rivals, and a feel for which tech they're researching ( swingin' merchant trick).

Max out on research and try to get roads on every worked tile to get required 2nd level techs, then turn down and do a 40-turner on a third-level which you reckon the AI won't be doing.

Your 2nd-level can be traded for other 2nd level - or 2-3 1st level - which could also be bought. If you get the 3rd level tech (eg Polytheism) first it'll be worth a lot.

You can sometimes suss out what they're researching by shopping around and trying various combinations with various rivals. If a civ with 250G offers you 9G for a tech that others will pay 150 for, then sell! sell! sell! to the others, as the 9G customer is about to discover it.

I'm beginning to view the diplomacy screen as an alternative battleground for beating the enemy. Yet, a few weeks ago my research slider was on zero until I'd built the kind of infrastructure that Arrian's talking about. With so many strategic possibilities for warfare and diplomacy, this game rocks .

[Arrian]

Cort Haus,

I'd agree that 1.29 makes researching more viable. But I'd still say that it pays to go all-tax early on, buying or beating tech out of the AI and reinvesting your money into the key city improvements. 1.29's slower tech progression means you have more time to reap the rewards of either strategic choice.

The Great Library is more powerful than it's been in a while (going back many patches). One can shut off research and depend on the library for tech for much, much longer now.

One of the key changes in 1.29, which I haven't seen too many people mention, is that fundamental change made to the tech race. It used to be that a given tech would cost 100% of its value for the human, and less for the AI (on Monarch and above). So a tech that cost 100 beakers would cost 90 for the AI on Monarch. Well, now they've changed it so that the tech costs 100 for the AI no matter what, and it costs more for the human. The gameplay translation of that is that the tech race slows down early on, allowing more time for you to get going. Then, once your empire is properly set up, you can still acheive the 4turns/tech rate and blow past the AI.

Since tech progression is slower, particularly early on, you can get a lot more out of the GL in terms of money. Say the GL used to last 30 turns (this is totally arbitrary, bear with me). Let's say it lasts more like 40 turns now. If you run 100% taxes during that period, that's 10 extra turns (or whatever the difference is) of full income straight into your coffers. Enough money to say... upgrade 20 horsemen to knights

The slower tech progression manifests itself in other ways as well. It's not drastic... but a turn here and a turn there really helps out. Your workers have a few extra turns to develop your terrain before you need to worry about those expensive Medieval city builds & wonders. Your military units have longer shelflives (the Knight springs to mind).

I like these changes. Yeah, they probably make the game easier. Whatever. They make it more fun. Plus, if we mess with the AI build preferences enough to produce tougher AIs, we can compensate and have an overall better game to show for it.

-Arrian

[DrSpike]

I play mainly emperor and deity games and used to think that researching before you had libraries and even universities was a waste of time. However some early landing comparison games in the civfanatics tourney have made me reconsider in games where you care about how early you finish. The first tech isn't going to get much better than 40 turns if you are expanding, but after that you can usually find a 'kink' that allows reasonable research without forgoing too much gold. What is more in some games with fewer players it you can have a substantial impact on the overall progress of research within the game fairly early on, even on deity.

[Catt]

Originally posted by Cort Haus


Question : how did the luxury trade deals go with the big boys? In my one emp game I was tiny at first but was able to catch up in tech by selling lux's for 2-3 techs each, sometimes with a lux thrown in too. I haven't reproduced this in other games yet but I'm intrigued by the trading possibilities of small civs. However, a recent post from Txurce on the Peace Dividend thread suggested that as his civ's power increased (but not the size) the deals became less favourable - implying that size *and* power go into the formula.

My game wouldn't be a good indicator of the "power AND size" experiment. In the early game I was able to secure 2 supplies each of 2 different luxuries. Taking one Japanese city netted me a sole source of a third additional luxury, and a culture flip netted me a sole supply of a fourth different luxury. Through about the late middle ages (IIRC) I was able to trade my excess luxuries for other luxuries, generally on a one to one (sometimes better ) basis, occasionally for tech or gold. But the real problem with my game re: the power & size hypothesis is that, fairly early in the game (early Industrial Age, I think) the world was down to 4 civs. Then down to 3 civs. Since I didn't control a monopoly on any luxury item, it was extremely difficult to trade luxuries -- my two rivals either (1) had their own supply, or (2) were already trading amongst themselves.

Originally posted by Arrian
I'd agree that 1.29 makes researching more viable . . . .

I think Arrian is digging in the right place; and I think he's right that the game is a bit easier under 1.29f (though I'm not complaining about that ), although harder to rack up a high score. I also think that the overall slower tech progression has some interesting and somewhat-wider-than-currently-being-discussed implications for gameplay.

The basic commentary I'm seeing here and at CFC is: (1) techs cost more; (2) since techs cost more, it takes longer to research and more money to trade; and (3) the games are longer. Plus of course Arrian's point about the renewed value of the Great Library, with which I wholeheartedly agree.

Techs do cost more, but we're not generting more commerce per turn than we were before. Ergo tech trading is substantially reduced, even among the AI (harder to save up gold and more economically efficient to research). I wonder if this means that the Zulu and Aztecs (when played by the AI) will now be totally hopeless in the late game, as opposed to mostly hopeless. I wonder if an AI civ that neglects commerce generation and libraries / universities will quickly stagnate as a technologically backwards civ, even with a very large empire.

Another interesting result of the longer game times is the (perhaps?) increased opportunities for the human player to "catch up" from a less than stellar game situation. If the AI has to play into the 1900s before a spaceship victory, versus, say the 1400s or 1500s under the previous version, then there are a large number of additional opportunities for the AI to shoot itself in the foot. My game again wasn't a great example (the remaining civs were all religious!), but I expect that during an extra 100+ turns in any given game, the AI tech leader or near leaders will get stuck in / instigate more wars, necessitating government changes, reduced science spending, etc. - I think the human will have much better odds at making come from behind victories.

Finally, one other interesting side effect / hypothesis regarding the effects of longer games. Since it seems clear that it will be much easier for a civ (human or AI) to both fall behind in tech and have a really tough time catching up on its own if it's not a larger/productive empire, I wonder if we're going to see more warfare, and more instances of war as the only option to avoid oblivion - i.e., I'm now four techs behind and falling further back; if I don't make war now the conditions for successful warfare are only going to get worse (bigger tech disparity). Will the patch make it more common to have two or three clear tiers of technological advancement: (1) the leaders; (2) the also-rans; and (3) the hopelessly backwards? Will the AU course on generating an entire age's worth of technological lead be more feasible under 1.29f?

I wish I had more time to play and seek answers to these questions!

[Cort Haus]

Originally posted by Catt
...I wonder if we're going to see more warfare

Yes! Certainly from the AI. With a higher GPT cost, it'll go broke more often and declare war to break the payments. More important than ever to watch the reaction to raising the bar by 1GPT above their offer. If the advisor screws up his face then you're in the danger zone.

Arrian - could be a good way to get a civ to declare war on you to spare the rep hit, but the best way is to settle that luxury closer to his capital than yours. I've had Bismark. Mao and Shaka all kick off within a few turns of planting the city.

[Arrian]

I you don't want to end up in a war, don't squeeze every last gpt out of the AI. If you can get 25gpt, but 26gpt results in "they would never accept this deal" that is the limit of their budget. Drop to 23 or so and take that. Otherwise, if they suffer a dropoff of 1gpt in revenue, they're gonna attack you.

Cort Haus,

Yeah, the AI definitely takes offense if you steal its luxuries. That's pretty much guaranteed to get you into a war. I use it often (planting overseas "colonies" after an AI vs. AI war to steal luxuries and pick a fight).

Catt,

Will the patch make it more common to have two or three clear tiers of technological advancement: (1) the leaders; (2) the also-rans; and (3) the hopelessly backwards?

Yeah, I think so. With the standard settings, the Zulu will indeed be hopeless... since they already were. Let's see what Alexman's changes to the AI build preferences do (AU102 - I have to download that one tonight).

-Arrian

[candybo]

Well, if the Zulus & Aztecs are hopeless, wouldn't that just make it historically accurate? Next patch: The Russians/Soviets self destruct

[Arrian]

Candybo,

The Communism self-destruct syndrome takes care of that, I think.

-Arrian