An sharp increase in "Playing Emperor" questions...

[yavoon]

wacken do u have anything similar for units? or is it assumed ur making units constantly and would only interrupt making units if the payoff was quick enough?

also I was wondering if there is any thought in ur games on the surity of the win as well as the speed or do u assume ur gna win all ur games and are just interested in how quickly u can do it.

 

[WackenOpenAir]

wacken do u have anything similar for units? or is it assumed ur making units constantly and would only interrupt making units if the payoff was quick enough?

also I was wondering if there is any thought in ur games on the surity of the win as well as the speed or do u assume ur gna win all ur games and are just interested in how quickly u can do it.

As i said, for units it is very difficult to decide their return on investment.
In conquest games, building units is the default. I can pretty properly estimate how many tuns i will need to conquer the world if i just produce units. This estimate makes the expected game length to be used in the decision what other investments to make. The simple but most important example is that you stop research all together when you reach a tech level needed to conquer the world and then spend all gold on upgrading and shortrushing units.
In science games it is about the difficult balance between building units and libraries. It is difficult to have numbers for this. It is one of the few things (i prefer to go by calculations) i do by feeling, estimate and experience.
The picture of return on investment is always there though and even though i can't use it with numbers here, it does help me realise not to build libraries to early. You can make simple calculations like how fast do you expect to conquer cities with an x number of units and how many units do you expect to lose per city, but still, it would be very difficult to calculate the real reward as your conquered cities are gonna breed into ICS land but will take a lot of time to do so. While also it takes quite some time for your units to conquer the land (compared to a building that is operational as soon as it finishes)
I did do some attempts to make such calculations in the past, but decided i had to do with feeling, estimate and experience.

In GOTM games there is no real doubt about the win. They are never above deity except once, and those that are deity+ are always made easy with a good start. Although i did lose one gotm where i was going for 20k and was attacked while i had only 1 defender. And i am aware there is always the small risk of being attacked during the rex phase. It is a microscopic risk though, as this 20k game is the only time i was attacked before i expected it.

Self made games i didn't play a lot. Basically i have had 2 periods of playing my own games. The first was when i just started (PTW) and practiced a while to beat deity. The second was when C3C was there a while already and i wanted to beat SID level. obviously, in both cases winning was not the least bit sure when i started. By the year zero however, it should normally be clear if i am gonna win.
Lately i don't play any self made games because my interest for the game has faded and i always like to think about games much more than to play them. Playing the games is just to test and confirm the things i have thought about.

You can sometimes make some calculations with units. For example, that first attack. Very often, i would upgrade 15-20 swords around 1000BC. If they are upgraded from warriors or not doesn't matter, they are worth 30 shields each. So if i buy 20 swords for 600 shields total and i expect to conquer my closest neighbour with it, you can try to give a value to that conquest. For example, his land contains 2 luxury resources. That is basically 2 gold per turn per city larger than the amount of luxes you already have for you. Then, you can add the value of any wonders, you can add the value of the techonology you are planning to extord for peace. You can add the value of the citizens over 1 in the captured cities (being conservative, just value them 20 shields each, as they could at least be used for rush building) and the value of cities (food surplus + free unit support if you need it + gold, 1 from commerce and 1 from wealth)
These are not accurate calculations, therefore, as i said i have done some in the past but now don't use them in my games. It is nice to do some of these just to get the idea.

 

[zerksees]

Economy
Understanding the return on investment on everything you do, and how it relates to the expected game length (in both ways. Many investments cause the game to last longer and thus make themselve seem good while others may make the game shorter and may make some long term investments obsolete). And how the return on investment, together with exponential growth cause "devaluation" or "interest" just like in a real economy. 100 shields on turn1 is worth more than 100 shields on turn 100.

Your insight on this area is eye opening.

Economy is one area you mention where I could use some improvement. I have drastically cut down on what improvements I build, but am probably still building too much infrastructure.

A few questions if you don't mind:
1) What is your typical government progression and how does the ROI affect it?
2) What civ traits best suit you when considering ROI? Besides the obvious agricultural and possibly commercial, I would think militaristic would have a lot more value to improve the return on investment of military units.
3) If you are not building cathedrals, how do you offset the unhappiness at higher difficulty levels - particularly when you are low on luxuries?

 

[WackenOpenAir]

1) What is your typical government progression and how does the ROI affect it?

The investment for any government is the same: an anarchy period. So for the choise between governments you only need to look at what it returns. In almost all situations republic is the best of the early options (rep, monarchy, feudalism). In some exceptional cases like you could read in the text file to that sid game i linked, another government may be better. (feudalism there)
Then only remains the choise to swich or not to swich. In order to change governments, the game must be long enough that the ROI on the investmet more than pays back for itself.
The cost of changing at higher levels for a non religious civ is so high that multiple changes are not an option. The change from despotism to republic is pretty big and pays back rather fast (may vary from like 20 to 50 turns or so) The extra value of democracy over republic is so minimal that no game will be long enough to pay back the investment. A late change to communism or feudalism could be worthy if you need it for pop rushing in a 100k game for example, but not otherwise.
The reward of non communal governments is always easy to calculate. Just see how many citizens will make an extra commerce, how much you are gonna pay for happiness and for unit support. There is little or no difference in corruption between republic, monarchy and feudalism, so i don't look at that.


2) What civ traits best suit you when considering ROI? Besides the obvious agricultural and possibly commercial, I would think militaristic would have a lot more value to improve the return on investment of military units.

To be honest, i didn't do a lot of calculation on the traits because it aint really needed. Agricultural is so immensely powerfull one doesn't need to calculate. In a recent post i made a simple calculation that resulted in it being 4 times as powerfull as industrious. Inustrious and commercial are commonly accepted to be the next best traits, but i didn't really calculate on it. I simply believe this common knowledge. (not that it is my habbit to just believe common knowledge, but in this case i feel it is correct without extensive calculations) The industrious value can be counted in workers, you need 33% fewer workers. To quantify the commercial trait, you need to do some work, calculating how many %increase you get in your shields and commerce due to the OCN bonus.
Maybe if i am bored someday in the near future, i will do an attempt to analyse and compare all traits


3) If you are not building cathedrals, how do you offset the unhappiness at higher difficulty levels - particularly when you are low on luxuries?

My first goal is to conquer more luxuries, those that are out of reach may be traded. If all that fails, the lux slider is my only method. In republic the lux slider works pretty well. Each citizen produces enough gold so that a 30% lux slider will usually be enough even with few luxuries. With some more luxes, 10 or 20% will usually do. I try not to use marketplaces either since then you need them in all your big cities which makes it a pretty huge investment. My games normally don't take long enough to make that investment worth while at all. And even if they do, units are usually a better investment (to conquer luxes amongst others)

 

[Ansar]

Allocating more space than 12 tiles per town means paying a big price. Workers have more tiles per cities to move over and road. Troops have farther to go to get any places and towns cannot defend each other with slow moving troops.

Above emperor you will not have the culture to fill the gaps in time and the AI will walk right in. I am not even talking about war, just scouts or settler pairs as they do not have to cross you borders to have a look around, unless you did spend the time to make temples and you should not have.

I do not really see much upside to wider spacing at any level. Maybe some Monarch game that you just want to mess around some, I don't know.

So...basically, if you are not settling in a CxxC pattern, then you are setttling in a wrong pattern?

 

[vmxa]

So...basically, if you are not settling in a CxxC pattern, then you are setttling in a wrong pattern?

As far as I am concern, yes. There are special games settings where that may not be the best, such as some HoF runs.

You want to have a 19 tile pattern for fun on Monarch, sure go ahead, but even then it is a waste. I know some players just have to have all structures and 21 tiles in the core cities, but they pay a price for it.