A Commercial trait commercial

[Arrian]

Add to that the similarity between Aryan and Arrian and... Ack!

But yeah, since nationalism tends to end the golden age of Cavalry (sure you can still fight Cav vs. Rifle wars, but it isn't knife-through-butter anymore). Riflemen do not require resources, so cutting iron & saltpeter will no longer reduce the AI to spearmen and longbowmen. Add to that MPPs... I hate that tech.

Hey, speaking of Scientific civs, in my most recent game (Japan), the Koreas got Engineering, not Monotheism, upon entering the Medieval Age. That's like a 1% chance, right?

I play too much Civ.

-Arrian

[joncnunn]

I'm not sure what the exact percentages are, but there is a chance to get any of the first level techs for sci civs.

The problem is that the RNG formula is buggy and picks the first option much more often, which results in:

Over 90% chance : Monothesism
Over 90% chance : Nationalism
Over 90% chance : Rocketry

The next PTW patch is expected to fix this giving nearly equal odds for all first level techs.

[Brizey]

I am almost finsihed with an experimental game. I am playing on Regent to give myself a chance. I am playing Korea on a 80% water, standard, arch. map. The game is almost over. I was on my own island, which I borged. Then I did a massive catapult to 'Cha upgrade (~50). I had a gazillion caravels and a few cavalry and musketmen. I pounded the crap out of India, my nearest neighbor and biggest threat.

Meanwhile, England got MOWs and went berzerk. When I reached the modern age, they were actually bigger than me, but I had a really strong tech lead (probably due to Regent setting and Scientific and Commercial on a inherently spread out map). I invaded England's home island with 50 tanks and 70 artillery. i captured four big cities, but they are STILL ahead in points. England! I now have MA and MI, so I can crush them at will, and with 70 artillery, with almost no casualties.

Obervations:

The extra firepower and range of Artillery makes a HUGE difference. It took me forever to pound down India's capital on hills with musketmen with 'Chas. Artillery pounded down riflemen more easily on hills.

Lots of artillery (70+) makes any battle winnable. I woudl rather have 150 artillery than 150 MA against MI on hills with CviDef. You just park them on some hills, defend with about 10 MI, and pound away for five or six turns. Eventually you destroy the barracks and the population. I need to set this up as an experiment.

Korea is better than I thought. It's been awile since I played scientific, and I really appreciated the free Nationalism and Rocketry. And Commerical made a giant difference on this map. I had two landmasses at full production.

[Artifex]

How tight do you need your cities to be in order to take advantage of commercial?

1. city-tile-tile-city?

2. city-tile-tile-tile-city?

3. city-tile-tile-tile-tile-city?

I usually build my cities like #2 or even #3. So I guess commercial stinks in that case right?

Can option #2 work with commercial or not?

[Artifex]

With Rome, when you use the warrior build/upgrade/legion rush strategy. When is a good time to upgrade and rush? How many cities/warriors should you have when you do this?

[yxhuvud]

artifex:

Regarding city placement. Place your cities so you can get production from as many good spots as possible in the short run, and from as many spots as possible in the long run. This includes water tiles. Usually this makes for a combination of 1 and 2 for me, sometimes I will use even shorter distances. It depends solely on how the map looks also on how close your opponents are. The closer your oppenents are, the tighter your city placement should be, since you will want to have as muhc production as possible when you go to war.

Which takes us to your second question. As a general rule, you cannot have too much troops. On the other side, you cannot attack too early if you have sufficent troops. This also heavily depends on how close your opponents are, and if you are going to war with one or two stacks. Which also comes down to how the map looks and where your opponents are. If you are doing early (around 1500BC) war, and going to use an oscillating strategy, I would recommend having at least six or eight warriors to upgrade. Later wars naturally takes more troops.

[alexman]

Originally posted by Artifex
How tight do you need your cities to be in order to take advantage of commercial?

First of all, the commercial trait benefit kicks in even if you have no overlap. In a test empire (Emperor, Monarchy, standard) of 15 cities with no courthouses I got 6.8% more income. But the benefit becomes greater as corruption due to number of cities becomes more significant. A test empire with the same population as the one described above, but with 25 cities 3 tiles apart, got a 9.1% increase in income from the commercial trait.

So the commercial trait shines when you have enough cities so that most of your corruption is due to number of cities (as opposed to distance from the capital).

This can happen in Despotism without courthouses, but it generally means very tight spacing (at least city-tile-tile-city). It can also happen with loose spacing in a Democracy with courthouses and police stations, but you will really see the benefit when you are way over the OCN.

[badman]

I hate the fact that close city spacing is more efficient. :/

Anyways, I've been wondering if you guys prefer tight city spacing in decent terrain or if you rather pack your cities if you start in crappy terrain.
I tend to pack my cities close together if the terrain sucks but place them rather loose if the terrain is good.

[Dominae]

The reasons you pack your cities in close when the terrain sucks are the very same reasons you should pack them in close when the terrain is nice.

I could go into a big long explanation, but just think about it and let me know if you agree.


Dominae

[Catt]

First, I'm sorry . . . I last posted here saying I'd try to find the tests I had conducted some time ago, and then forgot all about the subject until this thread jumped back up the board today.

I am still a bit befuddled by the view that the principal strength of the commercial trait is the ability to engage in tighter builds without resulting painful corruption. While I see that the tighter build pattern allows the benefits of the trait to shine more clearly, I still can't get over the view that the +2 gold in the city center for cities, and the +3 in metros, is a far larger distinguishing factor that separates the commercial from the non-commercial civs.

I couldn't find my old screenshots, but I took a few new ones. It is difficult to capture the relevant details in a screenshot, so, a few posts down, I'll also attach the scenarios constructed which generated the screenshots (and, of course, the data) should anyone care to take a look at the scenario maps and "empire set-up."

My original tests, weeks or months ago, were run using an imaginary empire -- it employed a loose build pattern -- not the "optimal" of no overlap, but also not a rigid city-tile-tile-city pattern (or denser pattern). It was "set" in the Industrial Age, with a very well-developed infrastructure -- banks (and stock exchanges in some cases) and universities being quite common. It in fact was an "overly builder" empire in that it was paying maintenance for improvements in a few 95% corrupt cities. It had a well-centered Palace and FP; was played on Emperor; standard map. Under such conditions, the OCN, as modified by the difficulty, should be 9.6 (12 x 80%). For a commercial civ, the OCN should be 12 (25% boost). This imaginary empire (one of several) contained 32 cities. Just about all (if not all) cities were Cities (<13 pop) or Metropolises (>12 pop).

Below is a screenshot of the summary F1 information from 2 versions of the scenario -- the first when played as a commercial civ and the second when played as a non-commercial civ. You can see that the commercial civ enjoys about 8.3% more income from its cities about 170 gold). What you can't capture in a screenshot is that the vast majority of this extra gold comes from: (1) the city tiles of cities and metros within the core; (2) as modified by the improvements (markets and banks).

A detailed review of the F1 screen shows in fact that 9 commercial civ cities have had their corruption reduced by 1 gold compared to the noncommercial civ's cities -- this 9 gold makes it to the treasury as 18 gold (markets and banks). The remainder of the excess gold generated by the commercial civ all comes from the city center tiles -- and most of it comes from the core cities -- the reason that the raw corruption numbers seem to be about the same is that the extra city-tile gold afforded the commercial civ, in terminally corrupt cities, goes right to corruption itself, to some extent artificially inflating the commercial civ's raw corruption total.

[Catt]

After re-reading the thread (and deferring always to alexman's far superior understanding of corruption), I decided to test an ICS-dense pattern to get a feel for the effects. I built a second sceanrio with ICS spacing (city-tile-city); again Emperor; standard map; again with a solid Palace and FP placement.

But this time I didn't bother to "construct" an empire with city improvements, etc. -- it was just a series of cities (several cities had harbors for food). I also set-up 2 different versions: version (1) has all "cities" at size 4 (meaning no city center bonus); and version (2) has all cities at size 7 (meaning +2 gold per city).

The screenshot below shows again the summary F1 screens for commerical town empire; non-commercial town empire; commercial city empire; and non-commercial city empire, respectively. You can see that, in the "town empire" the commercial civ enjoys about 8.3% more income than its non-commercial peer. But what's interesting is the "city empire" -- the commericla version enjoys a 29% advantage over its non-commercial peer (I forgot to reduce the slider in one shot -- with it at 0%, the income would be 290 instead of 130).

[Catt]

All of which leads me to believe that the key factor that separates a commercial civ's performance from a non-commercial civ's performance has more to do with ensuring larger settlements appear as early as possible. The biggest leap in performance will come when moving from towns to cities, with a smaller step from cities to metros.

But it implies several things to me: (1) those players who often allow very few cities to grow into metros (out of happiness or other concerns) may be significantly short-changing themeselves when playing a commercial civ; (2) when playing a commercial civ, it makes sense to get any needed aqueducts up and running sooner than one might otherwsie -- growth yields big rewards to commercial civs; and (3) it may make sense to alter one's research pattern on entry into the Industrial Age -- placing Sanitation higher on the priority list may make a lot of sense for commercial civs.

What am I missing / being a bonehead about / screwing up in arriving at the conclusion that the principal advantage of a commercial civ is the city-center bonus gold?

Catt

PS - here is a zip with the test scenarios I used -- 2 for the "imagined empire" and 4 for the ICS empire. The ICS empire also is altered so that, when playing on Emperor level, 20 citizens are born content (I didn;t want to have to do something to ensure no riots on the first turn. Also, if you start turn one on the ICS, you will need to manually switch a bunch of citizens from tax collectors / scientists to entertainers for an apples to apples comparison.

[Dominae]

Catt, assuming your analysis is correct (I'm not the one that's going to dispute it!), there's still the fact that a tighter build pattern helps the early-game immensely. Forget about Commerce in the Industrial age, I'm focusing on Shields on the Ancient age. I think what alexman was getting at is that the tight build pattern that is known to work well is actually helped by the Commercial trait, in the early stages of the game. Once cities can get real big (beginning to middle Medieval era), the effects mentioned in your test should start appearing, and will come into full force in the Industrial age. But, as always, getting there is the key. A tight build pattern makes this easier.

It is possible to ICS (city-tile-city) in the early game, and remove cities later on to get some good Metros. Unfortunately, it is not possible to do this with a looser pattern (city-tile-tile-city), which I prefer.


Dominae

[Nor Me]

Originally posted by Catt
What am I missing / being a bonehead about / screwing up in arriving at the conclusion that the principal advantage of a commercial civ is the city-center bonus gold?

Nothing. It's true. The point alexman is trying to make is that commercial has an early advantage over scientific and religious because the tightly spaced ICS and early war without the distraction of improvements strategy is optimal. (Your shooting yourself in the foot by arguing in the other thread that the strategy you play is not ideal.) I pointed out in the 1st reply that while Commercial helps this startegy, it doesn't necessarily encourage it.
This is why this has little relevance to his argument. If you do REX for ICS, it will be some time before any city is size 7 even with rivers. Aqueducts come at the same time as goverment changes and the middle ages when the Commercial advantages are much smaller than those for Scientific and Religious.

As for sanitation, a Commercial civ has no more reason to prioritise it than and Industrious one as the gain is similar. It's city level where Commercial has the advantage. Of course if you only play Carthage and France it doesn't matter.

[Artifex]

I might try city-tile-tile-city, but I can't see myself ever playing city-tile-city..thats just too extreme.

I thought city-tile-tile-city was ICS...and that I thought was pretty extreme...but you call this a loose pattern? wow.