Delay AI tech trading (more realistic)

More of a change request and I haven’t real tested it, but:

I have a feeling that AIs directly trade after one of them has researched something. Human players also take some time to react/negociate the trade before they actually send someone a technology. So I think there should be a slight delay for those trades.

By the way, when/how does the AI spend its money? And does it pay the exact amount of transfer cost to its AI traders? I’d like to understand how the AI thinks in this respect.

1 Like

BUMP. This is still insane. Human players don’t automatically trade EVERY tech to every other player. It is beyond unfair when facing a bunch of ai. A friend and I stuck around in an 8 player game where the other 6 quit. Every single ai shares every tech. It’s daunting.

1 Like

That’s not necessarily true. I’ve played games with people that seem to never sleep and send it immediately. I try to as well, but there are also players who just don’t understand what alliance really means and will wait to be asked before sending.
Also, you can trade with the AI immediately so as long as you’ve got the cash, it’s not a problem.
We once did 2 humans vs 30 AI and our biggest hurdle was the AI getting all techs when we could only get 2 at a time.
We were able to buy the tech to remain even.
That was too expenive though.
We then noticed the AI will stop sharing tech to other AI that starts to fall behind. We exploited that hard. Once we broke the AI into factions, we started testing to see if we could get AI allies to turn on AI allies. The answer is yes.
If you stall long enough and keep your neighbor friendly, it’s allies that can’t reach you will get impatient and attack the one that borders you. You can then take advantage of their hard work and sweep in. Once we figured this out 30 AI were no longer a problem, we had them fighting each other as we cleaned up the mess.
It’s just a matter of testing different strategies and understanding what the AI is meant to do.

Coasting below the starcount leader enables a decent amount of tech trades. My last game, the AI leader held 33 stars, I held 31 stars, the heavy trading AI clique members held 30, 29, 29, and 25 stars. Lagging AI left had 12, 10, 4, and 0 stars.

ticks 1088-2336
Leaderboard AI 33 stars always researched Weapons easy to trade with.
Myself 31 stars research mostly TF and XP
AI 30 stars Top trading tier
AI 29 stars Top trading tier
AI 29 stars Top trading tier
AI 25 stars Top trading tier
AI 12 stars 2nd trading tier - also easy to trade with
AI 10 stars 2nd trading tier - also easy to trade with
AI 4 stars orphaned. lagging, not worth trading with
AI 0 stars KO

Game settings:
10 players total, 24 far distance stars per player hexgrid, tb 8 tick jumps, $15/trade whole galaxy.

While there were noticeable attacks and leaderboard changes between the AI for the initial 600 ticks, leaderboard positions stablized between ticks 1088 to 2336.

The 2nd tier trading empires allowed me to keep up with banking and Manufacturing tech. Sometimes Mfg tech from 2nd tier allowed me to trade up scanning and HR with a few top tier members.

The top tier trading empires concentrated on Scan, HR, Weapons, and TF.
The 2nd tier trading empires concentrated on Weapons, Bk, and Mfg.
I concentrated on TF, and Exp.

My spending concentrated on Econ, lagging a bit on Sci, lagging a lot on Ind. After 800 ticks, I was able to quickly ramp up Sci and Ind builds from all of the TF tech advances.

tick 1800-2328
I eventually built up enough military to start raiding the 25 star neighbor. The raids evolved into star capture campaigns. My edge position allowed me to abandon 1 star for every star captured from that 25 star neighbor. Lack of Ind builds over all that time made me vulnerable to attack, so I needed to stay off of the top leaderboard position until I could afford to bribe or properly combat my way out of trouble. The range and HR techs allowed me to travel all the way to the far edges of the entire galaxy reaching unclaimed stars. Reconnecting with 17 Unclaimed and 11 abandoned stars will help with Econ and sci as my vault to nearly double the former starcount leader. The main combantant AI has been reduced to 14 stars and 30% of my shipcount. That empire’s Sci losses lead to ejection from the top tier trading group. Loss of a 4th member slows down their trades a little.

Tick 2432
That former 25 star AI is on its last star. I cheaply bribed that empire 8 times to 0 regard, then many more times to gain 30 techs - mostly HR , and scan. Two payday cycles later, I had enough Stars and Sci to stay ahead on every key tech.

Trading with an AI on its last star with any econ:
Getting a former AI top trading tier empire to a last single star gives a really good opportunity to bribe for techs cheaply.
I gained scan 50-64, HR56-67, Bk71-73
This takes a bit of guesswork on how much credits they have to pay for the trades following each bribe.
I used my phone calculator and gave about 40% of his cycle income as a starting base.
Then I added the amount of every bribe, and deducted the amount to trade any tech over to me. You want to keep a calculator balance of at least $200 to ensure trades would come through. Eventually the dollar amounts should escalate to approximate the expected trade cost for high level techs.

The costs of bribing other empires at that time was about 5x the trade cost. You will need to have lots of reserve credits to maximize tech gains. My overinvestment of Econ made this possible. The heavy bribe spending prevented me from buying econ until the next pay cycle.

This will not work if the AI just took a star, and has not yet built any econ.
This opportunity works whether you or some other AI is at war with the dying AI. Your timing control is better if you are the one sending the fleets to the combat zone.



Perhaps it is my lack of experience with this game but… The game we are currently playing has all the ai at the same tech level. These are ai that were fighting previously. That just doesn’t feel right…

If you can’t create your own games, I’d be happy to make some training games vs AI
In your current game, allow the AI to take the lead, befriend bordering AI and be patient.
You can also set traps for the AI by leaving stars unguarded with ships on the way that will arrive before the AI can to give you a defending bonus and leave it’s stars defenseless. Once you get the hang of it, even AI’s that should roll over you will only pose a minimal threat.
If you wait long enough, the AI goes crazy and just starts moving ships around. These make openings for advancement into their territories. They’re also very bad at noticing when you keep jumping to easy stars to destroy infrastructure and make them constantly rebuild and therefore eventually too broke to do much