How should units surrendering to a player affect the combat units? Also, what about information that is gleaned from captured units? In RL, if a unit surrenders, the unit that made it surrender is sort of out of commission too, since now it has to safeguard the prisoners. But I think the more interesting problem deals with gaining information. In real life, plans are not totally made on the fly. Troop numbers, troop movements and objectives are all thought out before the battle begins. There have been several cases in history of one side gaining the battle plans for the other (Antietam and Market Garden come clearly to mind). In both cases though, the advantage was not followed through, for in McLellan''s case, he was too timid to take advantge of the foreknowledge (Lee still almost beat him), and at MarketGarden, the Germans thought the plans were fake and meant as disinformation. However, the way most games play, it is impossible to have espionage or intelligence (or counter intelligence for that matter) work. This is because the battleplan is entirely in the player''s head. Now in my system it is a hybrid turn/rt system, so I thought that it might be possible for a player to "see" the orders that the computer or other player''s issue as an advantage. This simply isn''t possible in a full RT system, but can be used in turn based systems. Any other ideas on how to model information that is gleaned from stolen plans or captured units?

Well, in an RTS, you''d have a hard time showing unit "destinations" - as in, where or how units were moving, but you could reval parts of the map and the current location of the capture-ee''s other units or buildings. Perhaps seeing everything in an area around the captured unit might be a good compromise in an RTS.

Similarly, in a turn-based 4E or simulation game, you could be given a look at a unit''s home city, or the area near the unit in question, or a look at the last 20 moves the unit made (perhaps the distance back could vary depending on how successful you were at gaining information from the captured unit- depending on factors like the unit''s loyalty, your relative strengths, etc.).

- HC

-- EMail: cloweh@rpi.edu
-- AIM: SeigfriedH

The problems you guys have mentioned are primarily caused by the discrepancy between the scale of most RTSs and the scale of the operations mentioned by Dauntless. The Battle of Antietam and Operation Market Garden were both huge, involving thousands of men (and tanks, aircraft, etc. in Market Garden).

Most RTSs are nowhere near this scale. They generally have a map of a few square miles, with as many as 200 units on each side, max. Such a tactical situation would not find espionage or interrogation very useful, beyond learning where particular units or buildings might be.

But if we were to expand the discussion to large scale, non-turn-based strategy games that are not the cliched Command & Conquer clones that we are so accustomed to. Imagine a game in which time moves along at a time scale of 1440:1 (1 game day = 1 minute in the real world) and where the map stretches for hundreds of miles (think Civilization or Empire). Instead of building a massive base in an abstract time of five minutes and churning out massive armies from nothing but crystal/gold/etc., we could make a much more realistic, detailed, exciting strategy game, in which espionage could play an important factor. With a good AI engine, the player could delegate tasks to his subordinates, like assigning a given number of units to a General and telling him to capture a certain group of cities. These orders could be intercepted by the enemy and acted upon with great effect.

-Mike

[edited by - doctorsixstring on January 20, 2003 10:46:41 AM]