Talk:Proposals/Scripted Campaign

From UFO:AI
Jump to navigation Jump to search

Introduction

This page is to give an opportunity for the devs to ask mattn what his plans are for the scripted campaign, to better understand his vision and to decide what role it should have in the future. For reference, please see the staticcampaign.ufo file which outlines an example campaign using mattn's new system.

I've also written up a briefing on how the current campaign works for comparison.

Basic Concept

The assumptions I draw from the example campaign are the following (please correct where necessary):

  • Each stage of the game defines a number of mission "sets". Each set includes a list of possible maps and a quota that must be reached before the set is complete. When a set is complete, the player moves to the next set.
  • The number of UFOs which appear are defined for each set.
  • Quota represents some kind of minimum number of successful missions which must be completed for each set.
    • Does this represent some kind of fail condition?
      • The quota counter is only increased if the mission was won. --Mattn 14:41, 6 December 2012 (CET)
    • What happens if the number of UFOs is reached but the quota of successful missions are not complete?
      • The number of ufos is not related to the missions that appear. The ufos will add new missions that are not in mission list. Those are random. --Mattn 15:41, 6 December 2012 (SAST)
  • There appears to be some kind of command callback at the end of each stage. This is used in the alien base stage, but how is the alien base actually created? How does it know to spawn the mission?
    • It does know how to spawn the mission because there is a mapdef with alienbase - it will just create an alienbase mission linked with the mapdef alienbase. --Mattn 14:35, 6 December 2012 (CET)
      • Btw. this is not yet fully implemented - see SCP_CampaignAddMission --Mattn 15:39, 6 December 2012 (SAST)
I'd like to see an explaining description for each tag used in that script. That's better than guessing.--Duke 19:37, 3 December 2012 (SAST)
What are the planned features to extend that basic concept ? --Duke 21:09, 3 December 2012 (SAST)

Questions/Concerns

Sandbox vs Scripted Gameplay

This is my principle concern: how much randomness do you expect to be possible in your system? How far do you want to go in scripting missions? I agree that the current UFO/mission spawning system is very complex and difficult to balance. I am in favor of a simplified system. However, I believe that UFO:AI's greatest strength is it's sandbox gameplay -- we don't force the player down a single path, we don't treat the player too gently and tell them exactly what and when they should be doing things. When a battle begins the conditions are as random as we can make them within certain parameters (alien spawn positions, sometimes map layout, alien equipment, alien AI -- all emerges in unpredictable ways).

This is, IMHO, one of our greatest assets over the new XCOM. Our entire game is built around the idea that when you play it a second, third, fourth time you can not perfect your performance. Each time you play you'll be presented with different problems. There is no "best" move in any battle, no "best" research path, no "best" strategy. There are only better and worse strategies, and this is largely a result of its "sandbox" gameplay. We set some of the broader conditions, but the player and the AI has very wide freedom within those conditions. This means every time I play the game it's a bit different. This is important to me. When I play most modern games my sense of victory is undermined by the feeling that the game was carefully scripted to ensure I could win. I like knowing when I play a game that everything could go horribly wrong -- it's what makes me feel like I accomplished something when it doesn't.

I am in favor of more scripting possibilities for the campaign, but your example campaign makes me a little bit nervous. Here are a few specific concerns: --H-hour 19:18, 3 December 2012 (SAST)

  • Map selection: In the example campaign it seems like you choose exactly what maps are possible during any specific period. If we have 50 maps (for example), I don't want to only have 10 maps possible at any one stage of the game. It would be nice to be able to hold a few maps back for special events (a major terror mission, for example), but I don't want the rest of my maps to be limited. I want it to be as random as possible, so story-elements like special missions or general balancing needs should intrude upon this as little as possible.
    • see the other two points - there is still randomness included. --Mattn 14:46, 6 December 2012 (CET)
  • Mission locations: One of the best things about our UFO spawning system is that there is no clear pattern to where exactly events will happen. The UFOs spawn and fly and its up to the player to decide whether to let them land or not, when to engage them, and where to put his bases/radars to capture as many as possible. How does the scripted campaign handle the UFO appearance and flight mechanism? You appear to tie some missions to some specific cities.
    • This is still true for the ufo spawned missions (see below) --Mattn 14:46, 6 December 2012 (CET)
  • Alien Interest: Every aspect of our campaign is configured according to alien interest at the moment. How can the scripted campaign interact with this?
    • The spawned ufos (and their missions that they spawned) still rely on the interest value - there is no change in the way we handle it. --Mattn 14:46, 6 December 2012 (CET)

Ok, so the UFOs still appear as normal. It looks like the scripted campaign just defines a set of scripted missions which the player must complete before moving to the next level. Now I guess I'm a little confused. The scripted campaign keeps the hard-to-balance mission type interest mechanism, the same unwieldy UFO selection criteria, and just adds as a side element a few different map sets? If we keep the old UFO appearance mechanism, what problems do you think the scripted campaign will solve? In what ways will the scripted campaign be better? --H-hour 18:50, 7 December 2012 (SAST)

And I know you're busy, but it would be really nice if you outlined in more detail what your vision of the scripted campaign is. It's hard to judge something in such an early state if we don't have a clear idea of what it is supposed to look and act like when it is completed. In your view, what were the problems with the old system? What new opportunities will be available in your system? --H-hour 18:50, 7 December 2012 (SAST)

Format description

description