REX Tests and Results

[nbarclay]

This thread is for indicating what REXing tests various players plan to try and the results of such tests.

We can use it to figure out what REXing strategy seems to work best, and then post whatever plan we decide on to our topped thread as a reference for the people who actually play the turns.

[nbarclay]

I'm planning to run some tests generally based on the following premises:

1) I want to keep EotS's growth rate as fast as possible, i.e. two turns per pop building workers and an average of 2.5 turns per pop building settlers (or two for settlers where chopping makes it practical). The same philosophy applies to #3 once it goes in settler pump mode, but I'm hoping for straight 2-turn settlers there (not that I'm sure yet that I can get it).

2) At least initially, I'll tend to focus on balanced approaches that build enough workers for tile improvements to at least almost keep up with city production and growth.

3) At least in my initial tests, I'm planning to run two settler pumps in EotS and #3 and rarely build settlers or workers elsewhere until cities reach size 4. (That will slow down REXing a little but leave us with bigger, more productive core cities.)

Nathan

[nbarclay]

Are other people having the same trouble I am with disease striking the first turn of the new test game? The test game may need to be redone.

[DeepO]

Okay, I'll run a test where EotS becomes a pump, and #3 if possible (maybe after a worker to help). Other cities produce a mix of warriors and workers (a settler if productive), but no granaries, and no delaying to get to size 4. One of my objectives is to get a road to the Neck fast, preferably with some extensions onto a mountain. (so a largely worker oriented test).

What will be the end date of this test? or should me measure how long it takes to settle a number of cities, and gain a certain amount of gold? Further, we should build a few improvements as well, to simulate we'll build barracks and larger troops. Could be realistically, of course, like one third of our cities get temples, one third get barracks, the rest nothing.

Oh, and I haven't played the initial turn yet, just looked around a bit. Can't help you with the disease question

DeepO

[DeepO]

Other thing to test: chopping / pyramids question.

DeepO

[nbarclay]

I'd say build what you want where you want, but treat it like a real game. That way we explore various combinations of temples, barracks, and units. And we know how many shields each item costs, so we can translate back and forth to compare performance.

Nathan

[DeepO]

Okay.

End date? maybe best to keep saves all 10 turns or so, so we can compare when needed. I'm going to start on this after I find some breakfast.

DeepO

[nbarclay]

How about if we end when REXing is complete or nearly so, and do saves around the 2500, 2000, 1750, 1500, 1250, and 1000 BC marks and every century thereafter (give or take a little since some of those years aren't hit exactly)? Using years would avoid the need to count turn numbers.

Nathan

[DeepO]

I hate counting turns too, so that one is resolved. As to the end: it won't be the same for everyone, and the results will be different. Some test will have more cities faster, but not producing anything in the new ones. We might get a bit more flexible on this.

Oh well, I'll start, and let you know where I get with this system.

DeepO

[nbarclay]

Almost by definition, ending when REXing is essentially complete means different tests will end different years. But I view the question of "When does REXing end?" as an important one, and as long as people keep a save from the last turn of any test that looks promising, they can go back and play it some more if doing so is needed for comparison purposes.

Nathan

[alva]

Are other people having the same trouble I am with disease striking the first turn of the new test game? The test game may need to be redone.

Started this morning, but was to tired and made too many mistakes, so I forfeited . I didn't have any disease on the first turn though.

BTW: how do guys do these things, do yo keep notes of every turn:worker/labourer/gold etc.??
(I did so with that pyramid test, but, pffffff........)

[DeepO]

I had the same problem, and first I thought I just added a worker after disease... but this is not right, of course, as production was off. To counter it, I added a worker to the city before hitting ENTER, which died of disease immediately... no problems anymore (and certainly not worth it to redo the test!)

As how I run these tests: if you really want to go for it, you need to write down exactly what happens each turn. I've got an alternative plan for the first 8 or so turns (which I'll post very shortly), in which it is critical to be very precise. After that, I intend to follow a few guidelines (like: use the chopped forest in EotS to add to a settler), and don't keep track perfectly. It won't matter anyhow, as by that time the test scenario will be too far from the real situation. But at first, be precise

DeepO

[Dominae]

I believe I stopped the turn after disease struck Eye for the first time. Forgetting about the "2-rule" for disease, I packaged the save and send it off. I used the solution DeepO mentioned to deal with disease. Adding a pop and taking away a Laborer sets things to normal.

Are you guys planning to always remove disease in this way in your tests, or go along with it? What about Barbs: rapid execution with an MA, or not?

As a baseline for our tests, I suggest we define REXing as being "over" when all the city sites in the city placement proposal have been settled. The year at which this is complete is definitely to be taken into consideration when comparing our tests.


Dominae

[DeepO]

Well... maybe it's best to always counter disease in these tests, to get a good comparison. In the real game, it won't be possible, of course. Further: barbs: same thing. Disband a warrior, create a MA, and deal with them. But note how much gold you gain this way.

DeepO

[Dominae]

Very basic question: do you get gold from disbanding a unit "in the field" (i.e. not on a city tile).


Dominae