Long Patrol Patterns to provide variability to maps?

[Judge Iceberg]

I just finished up the REVENGE 1 scenario by Max-Violence.

I thought of an idea to improve that scenario, and am wondering if it is any good.

:idea: Once you have worked through one part of a map, the place doesn't change, and can safely be moved through after that. How about using very long patrol patterns to create a return presence in such areas?

Using REVENGE 1 as an example, there is an underground area and a surface area. You clean out the underground one first, then use it as a convenient way to safely move from one part of the map to another.

Suppose you created a group of surface patrolers whose path SOMETIMES caused them to go into the underground area. Use waypoints 1 and 2 above ground with a 90% chance when getting to 2 the guy heads back to 1. But 10 % of the time the guy heads off to waypoints 3-to-99. That way the initial population of patrollers would slowly "LEAK" down into the the other part of the map. Those who didn't would still be in play when the player gets to the surface level of the map.

Many of the maps in basic FoT could benefit from this, as you tend to end up working over a map like mowing a lawn, do the area once and don't worry about it again.

So I am curious if anyone has experimented with VERY LONG waypoint complexes that cover an entire large map? What happened when you tried to play it? :

Judge_Iceberg

[Max-Violence]

The problem with making patrol routes like that is, technically, you could have WAY too many actors in one actor with no real way of "fixing" it.

Taking your example - could you imagine how hard it would be if there was 3x as many enemies in the underground area of the map, with only Shadow to use?

Of course, one could use triggers to set the waypoints to change when X enemies are dead...

:idea:

Too bad I'm nowhere NEAR ready for entities and triggers in Revenge2

[Viktor]

If something like this "variable" patrol route idea could be implemented without having all the sentries turning up in one area at once, I agree it would add a whole extra level of realism to a map.

Triggers could work like this (excuse non-Map speak!):-

- Patrol A approaches sentry point 1.

- Patrol A detects that Sentry B is not at his/her post.

- Patrol A now follows Sentry B's route for one circuit.

- If no enemies are detected/engaged, Patrol A returns to assigned route and moves to sentry point 2.

Would this work??

[Max-Violence]

With a fair bit of extra scripting per patrol route + "meeting point" combination, yep, it could work.

Unfortunately, I'm not even halfway done laying tiles for Revenge2, so it'll be a while till I get to play around with this...

[Judge Iceberg]

I tried setting up a little version of what I had in mind, just to gather some numbers. Then I ran some "time trials" to see how fast the system would feed additional walkers into a patrol pattern.

Waypoints ZERO and ONE are 40 tiles apart
Waypoint ZERO always goes to Waypoint ONE
Waypoint ONE goes to: 0,0,0,0,2 (20% of the time on to Waypoint TWO, 80% of the time back to Waypoint ZERO)

Waypoint TWO and THREE are 40 tiles apart, 5 tiles from ONE to TWO.
Waypoint TWO always goes to Waypoint THREE
Waypoint THREE goes to: 2,2,2,2,4 (20% onward, 80% back)

Waypoint Four is within the field of fire of a guy with a minigun.

The Guy with the minigun shot each patroller when he came into his field of fire. Once all the patrollers were dead the scenario ended, exiting to the main menu. I started TEN patrollers, evenly spaced between Waypoints ZERO and ONE, walking continuously with no pauses at the waypoints. The initial state is everyone is between waypoints ZERO and ONE, walking back and forth. Every time someone makes his 20% role at Waypoint ONE he goes on and starts walking back and forth between waypoints TWO and THREE. Everytime he makes the 20% role at waypoint THREE he walks over to Waypoint FOUR and dies. This allowed me to just start it up and watch out of the corner of my eye for the game to exit to the main menu.


ZERO --- --- ONE
THREE --- --- TWO
---
FOUR <- Guy with GUN


I have only run the scenario three times, and it took between 20 and 25 minutes to complete. The first time through, when I was watching it closely went as follows:
.................................ZERO-ONE........TWO-THREE..........KIA
After 5 minutes: ...............8............................................2
After 10 minutes:..............4......................2....................4
After 15 minutes:..............1......................3....................6
After 20 minutes:......................................2....................8
Last guy KIA at 23 minutes.

So you can see that even a very simple setup like this can feed people into a patrol pattern gradually. Even simple additions like sending folks back to an earlier piece of the system could delay how fast folks are sent out on partol. Pauses at the Waypoints would also delay that.

I set this up as a testbed scenario, with a neutral human controlled character and two computer controlled ones. Do any of you every set up things like this, just to see how the game works?

In the case of the setup I made here you could have Waypoint FOUR feed into a long patrol pattern that, at its end, feeds the patrollers back into the Waypoint ZERO to ONE holding pattern, waiting for their next tour of duty in the line of fire. This should keep too many patrollers from getting out onto the patrol pattern and making life dangerous for the players.

J I