AI DEATH MATCH 2 - Page 21

Master Belisarius · #**201** March 24th, 2003, 06:20 PM

Pyrochette vs Sallega

The Pyrochette had not problems to defeat the Sallega before the turn 248.

Here the link: Pyrochette_vs_Sallega_C2.zip

Rollo · #**202** March 24th, 2003, 06:29 PM

Quote:

Originally posted by Master Belisarius:
...Hugh, honestly I can't recall. Can remember that in the past we discovered that the AI could reach the max number of colony ships, all of them Colony Rock for example, and then the AI unable to colonize the available Ice/Gas planets.

This is something different: check the game between the Aquilaeian and Gron. During near of 100 turns, the Aquilaeian only built Ice colonies. Although the Aquilaeian sent to die lots of colony ships, always the new colony ship constructed were Ice colony ships...

Well, as I said that was like 18 months ago and the evidence was very sketchy... (so, no worries for not recalling

).
But what you describe is what I have seen also back then. An AI would only build one type of colonizer. I manually scrapped some of them and found that the minister would replace again with the wrong type.
I haven't seen that happen in a long time, so I assumed it was patched out during the upgrade to gold. Apparently it wasn't...
The bug seemed more frequent on maps that had not all warp points connected. I can only speculate here, but maybe planets that cannot be reached influence the ministers decisions (in spite of the fact that the AI should have no knowledge of them).

Rollo

[ March 24, 2003, 16:34: Message edited by: Rollo ]

Master Belisarius · #**203** March 24th, 2003, 07:10 PM

Quote:

Originally posted by Rollo:
Well, as I said that was like 18 months ago and the evidence was very sketchy... (so, no worries for not recalling ).
But what you describe is what I have seen also back then. An AI would only build one type of colonizer. I manually scrapped some of them and found that the minister would replace again with the wrong type.

Then, think I misunderstood you... or my neurons are really dying quick!

Quote:

Originally posted by Rollo:

The bug seemed more frequent on maps that had not all warp points connected. I can only speculate here, but maybe planets that cannot be reached influence the ministers decisions (in spite of the fact that the AI should have no knowledge of them).

Could be, really I don't know the reason.
But the interesting is that if using the same map and same AIs, you start a game between the Aquilaeian and Gron, but exchanging starting sides, you will see that the Aquilaeian will have not problems to colonize, but the Gron yes...

Master Belisarius · #**204** March 24th, 2003, 09:45 PM

Fazrah vs Rage

The Fazrah defeated the Rage at the turn 267. They always had the inititative and won without problems (although the Rage used sun destroyers to kill some of their own invaded systems!)

Here the link: Fazrah_vs_Rage_C3.zip

Master Belisarius · #**205** March 24th, 2003, 11:28 PM

United Flora vs Earth Alliance.

The United Flora defeated the EA before the turn 200 and without problems.

Here the link: UFlora_vs_EA_C4.zip

It's pretty obvious that all the Mephisto's races are not performing so well like in the first contest did.
As Mephisto wrote, the map (with specialized systems) plus the "low bonus", are not good game settings for them.
Anyway, IMHO still think that for these races should be increased a bit, the maximum of refinery/farming colonies per system (see my explanation below).

[ March 24, 2003, 21:29: Message edited by: Master Belisarius ]

Master Belisarius · #**206** March 25th, 2003, 04:50 AM

Tessellate vs Orks.

A very interesting game (at least to me!).
The Tessellate expanded fast and started to attack the external Orks' sytems early. During the turns 120 to 170, conquered many systems and destroyed 2 Orks' homeworlds. Then, I believed the game would finish very soon... wrong! Somehow the Orks stopped the Tessellate attacks (don't think they could have done it against an human opponent, hehehe), and recovered several systems.
But, when the Tessellate's DN with Quantum Reactors started to attack, the game was decided.
The Tessellate won at the turn 280.

Here the link: Tessellate_vs_Orks_C5.zip

Master Belisarius · #**207** March 25th, 2003, 03:04 PM

Namovans vs Khrel.

The Namovans offered a good fight to the Khrel, but at the end was unable to stop them...
The Khrel won at the turn 410.

Here the link: Namovans_vs_Khrel_C1.zip

[ March 25, 2003, 18:16: Message edited by: Master Belisarius ]

Master Belisarius · #**208** March 25th, 2003, 08:28 PM

EEE vs Sallega.

The EEE defeated the Sallega at the turn 320.
The Sallega had some success in the early game, attacking the Aargau system, and later colonizing planets at Baksha and AAr.
But slowly the EEE changed the game to their side.

Here the link: EEE_vs_Sallega_C2.zip

Rexxx · #**209** March 25th, 2003, 09:58 PM

I just checked the EEE-Sallega savegames.
The fact that the AI takes the basic (and not the bonus boosted) resource production into account when it decides about colony types leads to many interesting results (to put it mildly).
All those nice research and intelligence compounds abused for mining and refining while the empire already is buried under piles of ore...

This bug reduces any attempt of writing a bonus using AI to estimated guessing (not very estimated but lots of guessing).
Has anyone reported it to MM?

Master Belisarius · #**210** March 25th, 2003, 10:24 PM

Quote:

Originally posted by Rexxx:
I just checked the EEE-Sallega savegames.
The fact that the AI takes the basic (and not the bonus boosted) resource production into account when it decides about colony types leads to many interesting results (to put it mildly).
All those nice research and intelligence compounds abused for mining and refining while the empire already is buried under piles of ore...
This bug reduces any attempt of writing a bonus using AI to estimated guessing (not very estimated but lots of guessing).
Has anyone reported it to MM?

Yes, I have reported it, but think would be good if you and others report it too...