newdig: Computer Player should use pre-dug out rooms.

[GoogleCodeExporter]

In r1850 with the newdig option when a mapmaker gives a CP a few empty rooms to 
start out with, the CP should use them and fill them with rooms. Instead he 
just decides to dig new spaces.

See attached map for behavior.

Original issue reported on code.google.com by Loobinex on 19 May 2015 at 12:44

Attachments:

• TC627_1.7z

[GoogleCodeExporter]

Agreed. Will not be hard to do as this was one of the things I wanted to 
accomplish, so will try to do it ASAP

Original comment by eatpajdi...@gmail.com on 20 May 2015 at 7:49

[GoogleCodeExporter]

Priority raised, this will fit easily into existing framework

Original comment by eatpajdi...@gmail.com on 22 May 2015 at 7:16

• Added labels: Priority-High
• Removed labels: Priority-Medium

[GoogleCodeExporter]

Original comment by eatpajdi...@gmail.com on 22 May 2015 at 7:19

• Added labels: Component-newdig

[GoogleCodeExporter]

Original comment by eatpajdi...@gmail.com on 22 May 2015 at 7:22

• Changed state: Started

[GoogleCodeExporter]

This issue was closed by revision r1851.

Original comment by eatpajdi...@gmail.com on 22 May 2015 at 9:28

• Changed state: Fixed

[GoogleCodeExporter]

Addressed in r1851, please test and verify.

Note that the AI will not automatically prefer dug rooms (since it calculates 
what's more favorable, including considerations of existing walls, access to 
hatchery, room size, etc.)

Original comment by eatpajdi...@gmail.com on 22 May 2015 at 9:30

[GoogleCodeExporter]

Will do.

Original comment by Loobinex on 22 May 2015 at 9:50

• Changed state: Started

[GoogleCodeExporter]

Test failed in r1854 with TC627_1.

It is already very good:
- It uses empty spaces it can use
- It expands rooms that are too small

However, it fails to recognize rooms that are too big. See screenshot.

Original comment by Loobinex on 23 May 2015 at 11:55

Attachments:

• 627Alcove.jpg

[GoogleCodeExporter]

I guess this is a matter of setting weights. I could lower weight of using 
existing dug out area compared to lacking walls. Aside from it looking ugly 
(but still something that a player might build for some reasons, so IMO it's 
OK), do you think the lack of walls carries enough importance that it should 
down-prioritize using that space?

Original comment by eatpajdi...@gmail.com on 24 May 2015 at 12:07

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

It is certainly not the end of the world.

A human might do the same, but much more likely would be to either:
-> Make a full rectangle of the empty space and fill it with the room needed. 
(Instead of making a 4x4 torture room, make it a 4x5 torture room)
-> Use the empty space for a room you want that big. (Instead of making a 4x4 
torture room make it a 5x5 lair)
-> Ignore the empty space and make the 4x4 torture room elsewhere.

Room efficiency has a huge impact and inefficient rooms should be avoided. If a 
training room misses walls, it could run on e.g. 60% efficiency. This would 
only be worth it if otherwise there would be no place to build the room or if 
an alternative location would reduce the training time by 40% from trips to 
hatchery/treasure room.

Long story short, yes give it a lower weight. (Perhaps multiply it with 
expected efficiency.)

Original comment by Loobinex on 24 May 2015 at 12:33