Need help with multimaps

[Catfish]

Don't you mean a bit?

The digits in the Leaders2 table are bytes. They have values of 0-2. That's why it doesn't use reverse programmer notation.

 

[The Person]

I see. A bit can only have the value of 0 or 1, not 2 or more.

 

[Boco]

I see. A bit can only have the value of 0 or 1, not 2 or more.

The answer (in binary) is 1.

No testing to report, yet. Left town, car broke down, and work's a b****.

 

[Boco]

That mask really should read: 'cannot build unit on map'. The mask only prevents a unit from being built on the specified map/s, it doesn't prevent a unit from accessing a map – that's what columns E and F are for.

So, the AI will move a unit to a map on which it's 'forbidden', assuming the E's and F's aren't F'ed?

 

[Catfish]

I see. A bit can only have the value of 0 or 1, not 2 or more.

That's right. As Boco pointed out, the masks in the @UNITS_ADVANCED table use binary. I also know that each digit from Leaders2 is stored in a byte, because I've seen the layout of this table in the .sav file (using a hex-editor).

No testing to report, yet. Left town, car broke down, and work's a b****.

Well that's tragic, Boco. I've been waiting for that report with bated breath.

So, the AI will move a unit to a map on which it's 'forbidden', assuming the E's and F's aren't F'ed?

Yes.

 

[Boco]

Yes.

That's very good to know.

Sadly, you'll have to keep satisfied with 'bating for a while more.

 

[Boco]

D**n, with only 6 landmasses on map1, the AI still isn't moving from Map1 to Map0. Maybe it's because all of my sites are one-ways under enemy-held cities. More tragic testing to do.

 

[Mercator]

I wonder, what happens if you give the landmass on Map 1 the same number as the landmass on Map 0?

 

[Catfish]

D**n, with only 6 landmasses on map1, the AI still isn't moving from Map1 to Map0. Maybe it's because all of my sites are one-ways under enemy-held cities. More tragic testing to do.

Enemy-held? They don't happen to have guys sitting in them, do they?

I wonder, what happens if you give the landmass on Map 1 the same number as the landmass on Map 0?

They'd run around in circles chasing their own arses... Just kidding.

 

[Boco]

Oops, I was fast and inaccurate in my post (now there's a first ). Four one-way sites lie below cities that are controlled by the same civ. Six other sites lie below cities that are enemy (i.e. at war) controlled but unoccupied. And yes, another nine sites lie below occupied enemy cities.

In the latest test, two 2-way sites were ignored as well. I'll check whether and when any intermap movement occurs during several future turns.

[Edit]
On map 1, landmasses 69, 72, 71, 70, 66, and 68 all lie below landmass 2 of map 0. I'm testing for movement from 69, 72, and 66 to 2. Eventually I'll test for movement from 70 and 68.

 

[Boco]

After 3 turns the only intermap movement by the AI was a single unit from Map0 to Map1 -- the exact opposite of what I'm hoping for. Time to take out the 2 way teleporter. Oh hell, time to cut bait and go with a single map solution for this AI problem.

 

[Catfish]

Well, if you want me to take a look at it, the offer's still open – not that I'll have any solutions, mind you - I've got my own AI problems. Sometimes the Civ2 AI is beyond help. If anyone ever gets their hands on the source code, AI scripting, in addition to improvements to the core AI, is definitely something I'd like to see.

 

[Boco]

After several alpha tests, I've seen the following:

Over a 6 turn span, typically only 1, maybe two, units (out of a few dozen)take advantage of 1-way transporters to hop from map1 to map0. Map hopping always occurred from the westernmost map1 continent to a region where the conflict is AI vs AI, and the defending AI has static units only. The 'attacking' AI simply hops a unit over to map 0 to garrison an otherwise empty friendly city.

During one test, the AI hopped a U-Boat (native teleporter) from map3 to map0, water-to-water.

So no flooding map0 with AI units from other maps, so far. Just a trickle.