On SP Scenario Design Principles

Discussion and development of scenarios and campaigns for the game.

Moderator: Forum Moderators

Post Reply
User avatar
skeptical_troll
Posts: 498
Joined: August 31st, 2015, 11:06 pm

On SP Scenario Design Principles

Post by skeptical_troll »

Hello fellow Wesnothers, as I am back into Wesnoth as it periodically happens, I'd like to start a discussion on a topic which I think is at the base of why creating UMC is so much fun: scenario design. I am sure many UMC creators would agree with that and I hope they'd enjoy contributing to the discussion based on their experience and personal ideas.

While I'm aware the risk here is not saying really anything new, as this is a recurrent topic in the forum, I don't recall seeing a specific thread on the matter, at least recently. I know there is a 'how to design a campaign' guide, but not a 'how to design a scenario' one, although a few general useful tips are given in the former. I think it would be nice to collect ideas and draw a more complete set of principles to take into consideration when designing a scenario (I am focusing on SP here), and in particular on how to achieve variety in a campaign.

I'd like to focus the discussion more on the 'strategic' than the 'tactical' aspect. I am not sure if I am using these words correctly, but what I mean is that I don't want to discuss much about variety in unit roster/terrain/type of enemies etc.. which mostly affects micro management. I think that on that front there is already a lot of diversity and it's perhaps the must fun part of the game. I'd prefer to discuss the aspect of high-level decisions and the general flow of battles in SP gameplay. The reason I bring this up is because having played many years, and having created some content of my own (the issue applies to my scenarios as well), I feel that, mainly because of the AI limitation, SP scenarios involves more often than not the following steps, regardless of the actual objective.
  1. recruit the units you need in the first 2-3 turns
  2. Advance all your army to a good position and form a line
  3. Defend against the waves of enemies until they run out of gold (with some ToD retreat+advance dance)
  4. Reach the enemy keep(s) and finish off the enemy leader(s), or accomplish whatever you have to while the enemy is too weak to stop you.


Of course, there are notable exceptions, and even this canonical structure is fun to play, but I think finding ways of avoiding its excessive repetition could greatly improve the game SP experience. I will start by giving some ideas on how to avoid the occurrence of the 4 steps above, some of which have examples in mainline.

Avoid recruitment of the whole army (or the bulk of it) in the first turns
  • Less gold from the beginning, more from the villages to conquer. This means the first part requires a small force, or cheap units to control them, and attack only once the army is built. This is rarer than one may think.
  • Uncertainty on which enemy you will face, so you will need to do some intelligence beforehand, as it's typical in MP. This has the problem of replaying with previous knowledge, so it can only be used with some form of randomisation of the enemies' composition.
  • Difference in terrains which favor recruitment 'in steps'. Example: you can recruit saurian and Dunefolk and your initial keep is in a swamp bordering with a desert. You'd need to use saurian to get out of the swamp and then go heavy on Dunefolk. Of course there must be several keeps in the maps for this.
  • The actual objective is getting gold, so recruitment should be done gradually and with care.
Avoid moving the whole army to form a defensive line on good terrain
  • there is no defensible terrain at all :roll:
  • Necessity to move/attack quickly (this helps for point 3 too) because
    • tight turn limit (obvious, but potentially frustrating)
    • strong enemies on pursuit, must flee
    • need to capture a target (location, unit) before the enemy to accomplish objective or to get a reward
  • Exploration/moving scenario with spread enemies rather than a single battle.
  • Trade off between defense/economy, villages are in one place, good terrain in another.
  • Objective to protect a hardly defensible position or to escort a slow unit
  • Defensible positions are not immediately accessible, need to conquer them by force
  • Army split. This one I particularly like, because it involves thought about balancing and composing the different task forces.
    • By design: you control two armies in different location with different challenges
    • By necessity: multiple fronts/tasks at the same time, e.g. enemies from different directions, chokepoint to defend + enemy to kill..
Avoid keeping the army on the defensive until enemy runs out of gold
  • Enemies with high income, defense only is not sufficient or they'll get too strong.
  • Enemies have a lot of gold, but recruit slowly (from 2-3 tiles)
  • Tweaking AI and using micro_AI to avoid reckless enemies. The standard is for them to be very aggressive.
  • Different AI sides fighting each other, need to take opportunities
  • Leader assassination is a viable and convenient option
  • Puzzle-like scenario: heavily relying on a particular tactic (heavily relying on features/ability of a unit or combination thereof )
Avoid the last 'killing leader' part when the battle is basically over, especially if it has no story relevance.
  • If the objective is 'kill enemy leaders', grant them good income so they don't remain alone at the end.
  • let leaders come forward once low in gold (with AI tweaks), or run away from the battle.
  • killing leaders not necessary, just village control / few remaining enemy units
  • Don't place leaders too far away from the bulk of the action
That's all I could think of for the moment. I'm curious to hear your opinion on this, how do you feel about this problem (or if you see it as a problem at all), and what others ideas you got.
white_haired_uncle
Posts: 1109
Joined: August 26th, 2018, 11:46 pm
Location: A country place, far outside the Wire

Re: On SP Scenario Design Principles

Post by white_haired_uncle »

Random thoughts based loosely on the thread topic/contents...

Earn Your Keep: You don't start in a castle with a bunch of gold. You have to capture a castle. One has lots of gold but is small, so recruiting will be slow. On the other path you gather villages, perhaps finding peasants along the way. It has less starting gold, but perhaps a larger keep. Basically, the player gets to choose more starting gold vs higher income. OR, you start with an okay castle and can play the same old BFW scenario, or fight/run/bribe your way to a much "better" one.

Maybe one path has terrain advantages (speed or defense, villages to heal in), or monsters to battle. Starting with a handful of level 0 units, the player can rush to a safe place to recruit, or risk losing some of his starting units but gaining XP (at least one unit should be on a leadership path, IMNSHO).

Monsters/Warring Factions: The enemy isn't just sitting around waiting for you to show up. Maybe there are monsters - you can fight them to gain XP (and gold, like a bounty?) or avoid them and let them weaken your enemy. Maybe you're after some orcs, and you find they are already engaged with a weak group of elves - you can ally with the elves for half your gold, fight them into submission/death (one option should give the ability to recruit shamans, because I like healers) or ignore them.

Leader Capture: I've seen this tried a few times, but I've never seen it work. When you kill/capture an enemy leader, his remaining troops join your side. Every time I've played one of these, the enemy leaders are always the last to die so I get nothing, but if done right it would be an interesting way to build an army. Maybe instead of the classic enemy leader parks his butt on his keep the whole game, there are a few small factions wandering around the woods you can use to build your army from the ground up. Maybe the one true enemy has to grow his army the same way. Do you go for the easy pickings, gaining a unit or three at a time or try to get the big early score and a commanding lead at the beginning? What will your opponent(s) do?

Advancement: The goal is not to defeat your enemy, but to gain XP, perhaps for a few key hero type units. Do you recruit a few heavy infantry units who can deal nearly death blows so your heroes can pick up the easy kills, or a bunch of cheap cannon fodder? Or marksmen that do less damage? Maybe some (expensive or weak?) healers or leaders? Maybe you have a (potential) healer and a (potential) leader in your hero group - which do you try to advance first?

Be the Enemy: Most of the time, either you take the battle to your enemy, or you have to hold out for X turns against swarms of enemies. Either way, the enemy leader basically sits back and waits for you come to him. Why not reverse the roles? Or even meet in the middle? Can the AI even be programmed to have an enemy leader actually lead his troops into battle?

EDIT: In most of the above, I proposed giving the player multiple ways to approach the scenario. Now that I think about it, I should mention that I don't think there should be a "right" way. Hopefully, each path the player takes would be about equally difficult (for some mythical canonical definition of "equally difficult").
Mawmoocn
Posts: 154
Joined: March 16th, 2019, 3:54 pm

Re: On SP Scenario Design Principles

Post by Mawmoocn »

Hello,

I don't consider myself as a expert player, I played some high difficulties with many restarts ...

In my opinion, most difficulties are due to AI behavior and map design.

When I first started this game, the raw elements of the game haven't sink in yet, but the most important aspect besides AI and map is to learn damage redistribution.

It's a hard concept to do when you're lacking awareness to achieve it or fail to absorb knowledge provided to you, which was the case for me until more trials have been made.

I won't comment on every point you made as they hold similarities when combined on a scenario.
I don't know if it helps but I think my comments lacks ideas, feel free to ignore/question.

I think you want new objectives or interactive scenarios similar to UtBS, I'll suggest some of it maybe but not sure...

@skeptical_troll
Spoiler:

@white_haired_uncle
Spoiler:
I think scenario only balance won't fit on campaign "balance", therefore making campaign considerations a requirement to any scenario adjustment.
Post Reply