Starting a game with less competition

[klasanov]

There has been a lot of talk about the number of competitors, and how the AI is handling the reduction in population. So far it doesnt seem to fare well, with 40 competitors. In my current game, 20% of the companies were private and only 6 of the publicly traded companies seemed to start faring well, after 16 years. By that time my avatar was owner of a billion dollar company.

I am going to start a new game, in it, it shall have

1) 4 competitors
2) competitors will be highly aggressive and have high starting capital
2) player company will own no subsidaries, its only industries will be real estate and media
3) dividend shall be 50%
4) corporation shall not invest in stocks, but player shall
5) player shall hold no more than 49% stake in any one company, other than its own (so as to have no controlling factors of the AI companies)
6) maybe anything else i havent thought of

I would like to see if the issue is with the AI not being optimized for ces, or if the issue is with the number of competitors.

[eleaza]

There has been a lot of talk about the number of competitors, and how the AI is handling the reduction in population. So far it doesnt seem to fare well, with 40 competitors. In my current game, 20% of the companies were private and only 6 of the publicly traded companies seemed to start faring well, after 16 years. By that time my avatar was owner of a billion dollar company.

I am going to start a new game, in it, it shall have

1) 4 competitors
2) competitors will be highly aggressive and have high starting capital
3) player company will own no subsidaries, its only industries will be real estate and media
4) dividend shall be 50%
5) corporation shall not invest in stocks, but player shall
6) player shall hold no more than 49% stake in any one company, other than its own (so as to have no controlling factors of the AI companies)
7) maybe anything else i havent thought of

I would like to see if the issue is with the AI not being optimized for ces, or if the issue is with the number of competitors.

I've replied some of my observation in another reply http://www.capitalismlab.com/forum/view ... =30#p17406 So I'll just list them briefly and added some other interesting issues I found.

1) The number really isn't that important, if by chance say all 4 are conservative personality, they will also have hard time expanding. But on contract if most 40 persons have aggressive character, and their expertise not overlapped much (where those do overlapped cluster into a single class), then most of them will still be profitable. Those moderate personality ones are more affected by this, since they tend to stick to single product class.

2) Setting to very aggressive at the start in game setting is just prerequisite to have aggressive AIs to show up. But if those persons don't open up their own businesses, then it doesn't' count for anything. And ironically give them too high start-up capital isn't a good idea, since they will die very slowly, and won't let those more aggressive personality persons to join-in, once the unsuccessful old ones go bankrupt. Constant renew of different personality corporations with different expertise is more important.

3) This I don't think this matter much, if you just set them up with the same amount close to AI start-up, and let subsidiaries roam free, they essentially just behave like AI corporations. And it's a good way of adding and testing different personality persons on the fly.

4) Don't interfere with the stock market, and it's best for players' corporation to do nothing, no dividend, and pure private. It's better to observe how AIs behave without interference

5) I don't know how you can "ask" other AIs not to invest in stock market. And as I said before players should stay away to make observation more objective.

6) I found a very interesting thing, if players every purchase an AI corporation's stock and as long as it's amount is more than its CEO's stock amount, over time if it's profitable, they will purchase back their stock eventually. And if players happened to own more than its CEO, you'll "accidentally" take over, when it purchases back too much its own stock. (but if the corporation is not profitable, they won't do this)

7) I found that AI tend to take some time to research techs to a reasonable level like 2 to 3, even 5 years before they start the business, and their start-up capital is mostly to sustain the losses during this period. And it would take like 10 years for a moderate personality AI expands to all products in its expertise class (without interference), and probably 15 to 20 years to fully dominate the class.

It's much faster for aggressive personality AIs, sometimes they don't even wait for technology to finish (and on occasion losses big, if it's more than its starting capital can buffer, it will end-up bankrupted), and if they are lucky where it doesn't choose a product class already have presence, they will continuously grow into different product classes. Even if there are competitors its own class, it will venture into other products to survive. But once there's competitors, they might go suicide in price war with opponents, and lose their profitability.

The conservative personality AI takes even longer than modest one to dominate its product class. Modest AI sometimes venture into other related product class, but not conservative AI. Conservative one will stay in a profitable product, and once there's competition, it will quickly drop out of never go into it, even the product is its own expertise, it doesn't purchase tech like crazy aggressive AI. Overall, it last for a very long time, and rarely go bankrupt, basically just a passive target in normal gameplay.

One thing curious is how AIs handle media firms. They tend to open new ones once they are profitable, also they are not good at dealing with outdated products. Over time some AI corporations will slowly "aging" with lots of useless junk factories, or unnecessary retail stores. And if it's still make profit, they would just put they on-sale, and stay that way for years. I called them the zombie AIs, don't go bankrupt, but also hard to die, and usually have decent market value not to get merger, they tend to boggle down the game a lot, and start to appear like 40 or 50 years in the game.