next »

[dspencer]

Maquis

This is a game about growing an army, and beating your opponents forces. It is a tactics game, and it is a roguelike game. It is a game you can play over and over, having a different experience each time. It is a game where you can plan far ahead, but need to be able to roll with the punches to win. It is a difficult game, but it is a rewarding game.

It's also not finished. Check this space for updates.

Download v0.0.1

[dspencer]

I present, my yet-unnamed-fire-emblem-inspired-roguelikelike!

Planned features:
~ 30 missions to complete to take over the world! Or something!
~ A number of different varieties of missions (Kill all enemies, seize all gates, protect some friends, get to point D, STOP that other dude from getting to D), all with limited timeframes!
~ Maps with wonderful different terrain features, such as "Forests," "Mountains," and "Forts!"
~ A number of different level generators, that generate maps with different feels! Like the forest island generator, the mountain pass generator, or the inside of the castle generator!
~ Limited consumable items to personalize your crew!
~ A fortune teller that's actually worth paying for, maybe!
~ Secret levels, if you complete secret goals! oooOOOooo.
~ Multiple AI strategies, with the enemy potentially switching strategies mid fight!
~ And if you die, your army might live on as a legendary "Ghost Army" to kill your later attempts of glory! You'll be real mad about it!


All of this in glorious, ascii beauty. Aww yiss.


Before the scenario, select a new member to grow your ranks. In this case, a fighter fills in a missing skill for our army.

Our Leader, Lord Oser, is preparing to take on an enemy in battle.

Success!


Ok, so here's the lowdown. I enjoy roguelike games; I also enjoy the games I've played in the fire emblem series (FE6,7,8). At some point, I realized that I liked both of them for many of the same reasons: specifically, the sense of urgency and finality of your actions, as any death is permanent for the rest of the game. Unfortunately, the fire emblem games are limited, and once you play through them you generally know what's going to happen.

This doesn't need to be!

To start, I'm going crib stuff from the fire emblem series; they did a good job of creating a bunch of interesting mechanics that work well together. However, as that stuff changes from game to game, I'll eventually migrate away as I balance things around the progression in this game. The idea, as in any roguelike game, is that it's very hard to win; so it won't be as easy as just selecting pegasus knight any time it comes up in the "recruitable units" area. You'll need to keep careful track of your funds, and move deftly around the map in order to complete your objectives in time!

I have a rather long list of planned features, as seen above, but  the ascii interface allows me to iterate fast. It would be nice to have a pretty graphical version of this game, but as I'm not particularly art inclined (and, even as I'm trying to learn how to draw, it'll take a while to fix that) it'll stay this way until someone awesome decides to donate some time to it .

The end goal is a terribly difficult game that you can play again and again without tiring of that preserves the fun blend of tactics + RPG character building from Fire Emblem.

Progress:
~ Multiple types of terrain are available, as opposed to spaces simply being open or closed. Forest costs more movement to move over, generally; classes can override this (such as pegasus knights, who can now rightfully move over mountains). This is the major progress.
~ Fixed a number of issues regarding fog of war - it still leaks a little info when setting paths (because I haven't yet implemented units getting intercepted when moving into fog of war) but it blocks most other info. Awesome
~ Pathfinding was rewritten, and actually works now!
~ Added a number of enemy classes. Brigands, Soldiers, wielding axes and spears respectively. Probably going to make a Bandit, which is the same idea but for swords.

Todos:
~ Generalize level rewards. Instead of the player getting the choice of 1 out of 3 units after every level, they sometimes instead get a troupe of units, or some money, or some items.
~ On that note: money, shops, consumable items, unit inventory.

[dspencer]

Ended up getting sidetracked, but the game is starting to feel closer to being playable. Accomplishments, alongwith related work that that still needs doing:

~ Better inventory. Weapons still don't have durability, because I haven't added shops so that seems unfair, but it'll be easy to change that. Consumables exist, at least in the form of the vulnerary, which is entirely useable at this point. Trading, items, and money are still todo.
~ Added most of the rest of the classes, except troubadour cause I got lazy. Magic users now exist, and the magic weapon triangle works. Added the resistance stat, and fighting uses the correct stat of resistance or defense.
~ It's becoming clear that balancing is going to be super hard. I made some improvements to the level generator to make it generate better levels, but it's not quite there. The boss sometimes generates right near the start, and sometimes the level awkwardly requires running around a maze with everyone waiting in one spot for you. Soon I'll need to add time limits to levels, and alternate goals like "survive" or "seize gate".
~ Critical hits now happen. For a while I was playing with everyone getting a free 5% crit rate, but that's a bad idea as it turns out. I haven't added luck to this game, so I'm trying to decide if I want some sort of critical avoid stat... In addition to this, I've added the rails for *killer weapons to be added to the game. Right now this only manifests itself in bows (which are flying killers) but now I have the code to make Hammers, HorseKillers, Rapiers, etc.

Let's make a new team, following our good friend Lord Mustony, the Mercenary! This game we're lucky, because we've got a shaman in our ranks. He'll make short work of anyone who isn't resistant to magic.


As you can see: more units, more spread out. Unfortunately, the boss is nearby, and there's no logic to end the level when he dies. Oh well!


I forgot to take a screenshot of us destroying cout raddy, the level 5 knight. But here's one of my fighters moving - now you can very easily see all the squares they can get to, and all the square they can attack to.


Still quite much to do!


EDIT: upon looking at this, I figured it's worth mentioning- I simplified the at a glance unit info, to have room for other stuff (like battle comparaison) that you more likely need. Now there's a UnitInfo pane that you can open, analygous to the R button in the GBA games. At the moment, this also shows a characters Growths. I don't think I'm going to keep that, but if I ever add in skills or similar things, that's where you'll configure them.

[sygpt]

This looks fantastic! If you don't mind me asking, are you using a library for the ascii graphics (like ncurses) or something you made yourself? I don't know for sure, but I think it would be possible for you use Noteye (http://www.roguetemple.com/z/noteye.php) to add in graphics and a lot of other great stuff (custom controls, tilesets, and so on) without too much difficulty.

[dspencer]

Yeah, I'm just using curses in a terminal; that looks like I could probably hook that up easily. I'll investigate when I have time.

[dspencer]

a couple of disparate improvements:
~ updated pathing logic, so you can move through fog of war without being blocked by enemies. Of course, if you try to do that, you'll hit them and lose your turn.
~ Now you can have theives on your team. While they can't steal yet (more on this later in the post) they do see further in fog of war. Theives, pegasus knights, and wyvern riders now all have scouting uses that make them quite useful.
~ Started a discussion reddit re: game balance. This has led me to decide I'm going to need some alpha testers. If anyone is interested in trying my game and helping me balance it, I'd love your help! PM me.

I've been doing some advance design work. In doing research of damage / exp formulas, on various fire emblem, I saw a number of things about Skills in later games. This seemed like a good idea for something I could do to differentiate my game, or different classes in my game. However, in discussing with a friend what constitutes a skill, I realized: everything could be a "skill". The difference between a troubadour, and a priest, largely, is that the troubadour has the "Rides a horse" skill. Class changes allow players to manifest this in the older games - as your guy gets better, you can use an item on them to boost their stats and improve them.

Rather than class change, I'm going to award skills or enhancements. I'm looking at a number of different systems, but at this point I like this one the best: At level 5, 15, and 25, a character gets an enhancement point. Independently, you'll earn enhancements as you progress through the game. You can apply an enhancement to a character by using one of their enhancement points. Consequences of this: You've got a decently strong cavalier, but you want to make him into a tank. Give him an Armor enhancement! At the cost of gaining vulnerability to a few weapons (hammer, and a couple of others) and one point of movement, he'll get an instant 1~3 defense and a better defense growth further on. That pegasus knight you have that has great skill and speed, but not great other stats? Buff her with an Inner Light Enhancement (extends LOS to that of a thief), and get an awesome scouting unit for fog of war maps.

Beyond skills already in fire emblem like greatshield, piercing, stealing, or boosted critical range; and basic skills like adding mounts, adding weapon types, boosting stats, there are tons of ways that new skills could be added with interested tactical uses. Zone of Control from wesnoth could be a skill (surrounding spaces stop enemy movements that move into them). Blinking - allowing units to zip past a wall of enemy forces. Laying traps, or having area of effect attacks. Regeneration, Passive healing of nearby troops, lifesteal attacks, area of effect buffs and debuffs. Etc. While these would all be useful, you'd be limited by the random selection of skills you got, which would hopefully have an impact on how you tackle an individual map.

TL;DR - get skill tokens sometimes. Characters get skill points every 10 levels, starting at 5. Spend tokens and skill points to give the represented skill to that character. Everyone (except the AI) rejoices.

I'm pretty excited about it.

[Miguelito]

I have to say the idea of this project thrills me to no small degree.

I see you're looking for artists to change to a non-ASCII style. In that case make sure there's not too many crazy permutations - y'know, things the Roguelike approach often demands, like a [Element] [Strength] [Weapon type] of [Adjective] [Enchantment], with each part having like 5 modifiers. Because artists will probably go mad having to draw 655 sprites for each weapon category.

Please develop this some more, because a Roguelike strategy game would be a sight to behold.

[dspencer]

Updates:

~ Added in skills. Right now, skills can be used to
1) Modify values (Farsight, which is a default skill on thieves, increases LOS by 2, and SwordWielding adds Swords to a unit's usable weapons).
2) Spy on events (Vampirism spies on hits, and heals the unit, and FireAura damages adjacent units at the end of the turn).
Not completed yet, but soon enough they will also be able to
3) Add activated abilities, available after moving a unit (for instance, a bard will be able to play a song that buffs a unit's power for a few turns).

~ Added in graphics! Right now it's just crappy programmer art but I'm going to get a better artist to work on it, in the near future. Right now I still have some legwork to support animations (the map is animated, but things like battles and unit movement, etc) but it's mostly there!

Behold!



My todo is (smaller tasks first):
~ finish drawing all the tiles (forts should not be represented by a green X, Monks should not be an M, etc), get some placeholder battle animations in.
~ Add some more skills. I have a lot of ideas for them, and I just need to implement them! Having a wide variety of skills, and a random selection to work with should make the game varied and interesting.
~ I need to limit units skill counts. Only let them add skills every 5th or 10th (or something) experience level.
Larger tasks:
~ Improve and modularize the AI. For each unit, the AI should examine all of the possible moves, and score each one based on a number of heuristics (damage dealt, damage taken, chance of kill, chance of death, distance from unit's lord, distance from center of enemy's cluster etc). Different AIs can use different values for each of these scores, defining different move characteristics. As you play the game, you will not be able to abuse the enemy's AI as much as you could in Fire Emblem, because you won't know for sure exactly what weights it's using. These weights could be randomized, or more likely I'll make a few presets for the enemies to use. One trick up my sleeve - enemies can (maybe) change AI in the middle of a fight! So, for the first 10 turns of a battle, the enemy might hold back, but after that turn they become extremely aggressive!
~Improve the level generator. Unfortunately, I'm not sure entirely what this means, specifically. I want to make the levels fun, and provide interesting tactical decisions (placement of forests, forts, and other elements like shops).  I'll need to spend some time diving into the generation algorithms to improve them, and add variety.
~ Improve level rewards / add goals /add alternate goals - Levels should give you a couple of different rewards: Items, Money, Units. Right now, they give you Units. Also, levels should have multiple possible goals to achieve, rather than every level be "kill everyone". I want to favor "kill boss" anyway, so that I can put challenge enemies guarding shops, or chests. Also, I want to limit the turns spent on a level - if you don't win in 15 or 20 rounds, then it's game over. Lastly, I want the fortune teller to tell you about hidden goals in a level (like, don't take anyone above exp level 10 to this level, for an extra bonus, or win in 10 rounds rather than 15 for extra money.

Woo!

Miguelito: thanks for the encouragement! I'm not planning on dropping this project. I do want people to play, however, and I think that having graphics will help encourage more people  to play. I don't need to have different graphics for every sword (maybe just a different icon, which is easier produce, if that). Additionally, I'm not worrying too much about having too many different modifiers for items; and the units which do have the modifiers will have just some minor differences to signify them, not additional sprites. Hopefully it'll work out! Worst case, I can switch back to console graphics 

[dspencer]

I've mostly just spent time getting animating working. I'm not convinced that I have my final scheme for interacting between the animator and the game, but I've got something that works; so it's likely going to stay as is until I have a graphical need for something better.

Ignore the fact that I got tired of fake animating and started just drawing scribbles... Also that I haven't even pretended to make the path display look good / animate a unit moving along a path.

[dspencer]

Added actually drawing the path as real animations, rather than just squares.

Major improvement - fixed the level generator. It's not perfect, but it makes simpler levels. A  bunch of paths are carved out of mountain ranges, and forests and forts are placed within.

The other major improvement is that now the level doesn't have to match the screen size. As you scroll around, the game will follow your cursor. This doesn't sound like much, but will let me support levels of all sizes without making it look bad!

Additionally, I cleaned up some sloppy routines that were causing the game to take a long time when doing certain actions. Stuff is pretty quick now.

Did some design work, and managed to limit the required number classes to 5, in a double RPS style.

Swordsman < Archer < Knight < Mage < Fighter < Swordsman < Archer

This generally makes sense, if I can come up with some justification for both swordsman and fighter to kill the mage.

I'm exploring a build tool that will allow me to finally make a distributable copy of the game. If you're interested in trying it, send me a message! Or if you've expressed interest in the past and I said "cool I'll let you know, then I'll assume you're still interested .

Another preview: you can see the new level design, a little bit, and that paths actually show where the unit is going.

[dspencer]

I updated to a better tileset (at least for the course of development), and added in AIs.

The AI can take multiple different strategies, by assigning weights to different aspects of a move. For each unit, the AI will find the movie that generates the highest score, given its own set of weights, and choose that action.

This lets me do things like: an enemy AI that moves as a squad, but slowly advances on players, attacking when it gets in distance. Or an AI for mages doesn't like to move, and doesn't like to take damage, but will value dealing damage and highly values scoring a kill.

Eventually, I'll be able to assign these individually to units. This will force the player to always be on their toes - they don't know what weights a unit has, so it might surprise them by ganging up on one target, or advancing in a different way than they're expecting.

There are still some rough edges, but this image shows a consequence of the new AI system:


Notice how when the monk goes in to attack the archer, he actually moves into melee to avoid a counterattack!

[OneMoreGo]

It's nice to see the graphics progressing 

[dspencer]

I've been out of town for almost a week, so I haven't had a chance to do too much major work / update the devlog. Oops!

Recently, I've been working on a lot of mostly minor AI and UI changes.
~ Giving the menus an actual background image.
~ Making the differentiation between player / enemy units clearer: because the source units aren't designed to support color tinting, I'm just adding an outline to units for the time being.
~ Re-enabled the attack/unit/comparison display, so you can see how much damage you're expected to do, how much HP they have, how much damage they're going to do back, etc.
~ Re-enabled game saves.
~ You can hold arrow keys to keep moving the cursor.
~ You can see rough growth rates (Low, Medium, High, and !!!) in the unit info inspector.

[dspencer]

Development has slowed down a bit, but it's still chugging!

Did a lot of pondering on design; units are going to be much less open ended than I had originally planned. I'm cribbing the advancement system from Seiken Densetsu 3, and giving each class a combination of stat bonuses and skills to determine what they do. To clarify: each of your units, at some level threshold, will be able to promote to one of two options; each of those options will also be able to promote. Each of these promotions aims to be useful, but in a slightly different way.

Let's look at one of these trees! The knight tree:
Your knight-type characters will start out as squires. They have decent defenses, but low speed; they wield spears. The squire can upgrade in two ways: Knight, or Knave. Knights of course, are great team players. Their defense continue to improve, and they've got the ability to lend support to nearby units in battle. The knave, on the other hand, focuses more on offense: they get a boost to their power, and the ability to wield axes, which allows them to take on opposing knights with ease. Remember, though - both of these options get another chance to promote! The knight can choose to again go the defensive supporter route, by becoming a Paladin (which can also take the role of healer) or an offensive support unit, by becoming a crusader. The knave, on the other hand, can promote to the more well rounded Dragoon (who can wield all weapon types, and thus can handle any threat!) or to the extremely negative... blackguard. Or something. Whatever his name ends up being, he actually attaches debuffs to baddies by hitting them.

The important thing to see here: each character has a role. They refine that role with each promotion. Each promotion gives them a choice between some of the following: teamwork and personal power, defense and offense, buff vs debuff. Possibly generality vs specific uses, but that doesn't seem clear which (good/bad) promotion should get which of those qualities; I'm already mixing that with weapon types, at least, so who knows.

While most characters will simply build on their skill sets when they promote, some will change fundamentally. For instance: your entertainer is a decent support unit; you won't always get the buff you're looking for, but you'll at least get something. When the entertainer decides to promote to Jester, however, they immediately lose any buffing ability; in exchange, they similar random debuffs. I haven't yet figured out the priest/monk characters, but going down the dark path might see you sacrificing your healing ability.



Design is great, but what actually got implemented?

1) Move unit definitions out of code and into configuration. This will make it easier to define new units in the future, and iterate on the ones I have now.
2) Skills can now provide new actions after you've moved a unit (in addition to "items" or "attack"). I'd been putting this off for a while, so it was good to get finished.
3) To make those actually useful, I've added in buffs. These are basically just skills, but they expire after a certain number of turns. As a proof of concept, entertainers can now perform for nearby allies, giving them a (small) boost in power for 5 turns.


I really need to get a packaged version of this out, so that I can get people playtesting it. The issue is every time I've tried, it hasn't worked, and I haven't felt motivated to actually fix it cause debugging it is hard. Wah. Hopefully I'll have some time to look at it tomorrow.

In addition to that, I want to actually implement character promotion, and finish the list of all the characters in the game. I have a gist of what I want, but there are a lot: I've planned out the knight tree (described above), the soldier (swords) tree (field marshal -> samurai or myrmidon/ mercenary -> freelancer/assassin), the fighter (axe) tree (warrior -> hero/champion, or barbarian -> berserker/viking) and the entertainer (support) tree (bard -> troubadour or dancer, and jester -> harlequin and madcap). In addition to these units, I'm thinking about a priest type caster tree (w/ light + dark magic), an elemental caster tree, an archer tree, a thief type tree, and a mounted unit tree.

At the end of the day, that's a lot of characters; but it's not too may different base characters (9). One of my goals is to make a tactics game, not a strategy game! I want there to be situations where wrinkles show up in your plans; where actively planning for contingencies is useful/required. I could use some more axe users in my team; but my cleric has terribly low HP. Should I promote my squire to a knight anyway just to ensure I have the possibility of having a healer in case he dies?

Oh crap, my best sword user is out of commission - better promote him to knave, so I can get a dragoon to cove that ground...

This has been quite a rambly post, but I hope someone finds it interesting.

[Rosstin]

Your slow but steady progress is beautiful. You show determination and potential as a dev!

[dspencer]

I've mostly completely finished debugging the packaging script - was running into a couple of different problems, the foremost being that I couldn't see what error was causing everything to fail! It's 99% complete and I'll have a build available for anyone who wants to play in the next day or two. 

[dspencer]

The mac build works. Success. I'll post a link sometime today. Do to me messing around with some other development, though, enemies don't currently spawn with weapons. Oh well.

Developments: continued to move unit data into configuration. It now takes about ~1 minute to add a new baddie: just specify their stats, some description, and what tile they use, and you're set.

sneak peak of some new enemies: farmers, cows, and flying eyes.

[dspencer]

I did a couple of clean up tasks. Weapons are not in configuration. Not exciting, but I think it's going to be good for iterating quickly when the time comes.

~ Units can dodge, based on their speed + con.
~ Weapons can define special on-hit abilities. Proof of concept: poisoned weapons. (Note: for the moment, poison is extremely nasty, able to wipe out a character in a couple of turns if you let it stack too much).
~ Refactored the terrain code. It's not properly stored as real objects. This will be useful in the future when units can place traps or I want to have spikes or fire, but for the time being it allows for....
~ Shops! The player still doesn't have money, so everything is free. Each shop has limited inventory, and one is generated per level. You better find it, too, because you'll need it now that..
~ Weapons break after a number of uses! Typically it's pretty high, but it'll be lower for some high-value weapons.

This should (finally) pave the way for level rewards. The three possible types I was thinking of were: special items, additional teammates, and money.

[juan]

I'm surprised more people haven't taken an interest in this. It looks like it could turn into something great and I'm enjoying seeing the steady progress that you're making. Keep it up!

[dspencer]

Thanks for the kind words.

Yesterday was pretty unproductive on my part, but there are some (small, but) substantial updates.
~ Added a bunch more units. Every planned unit tree (9 of them, total ) is present. I've gotta re-evaulate all the promotions at some point, but I'm starting to formalize my thoughts on them at this point. Hello, giant spreadsheet googledoc!
~ Fixed some misc small bugs. Example: Magic attacks now are reduced by resistance, rather than armor. Fixed a bug in the EXP formula that reduced the exp you got from the chapter boss, so now characters should level up a bit easier.
~ Added money - no more free items from shops!
~ Added HUD to display current money. It'll also display turn count and chapter objective, at some point.
~ Animated and displayed EXP gain. Right now it's ugly, but it shows you when you're levelling up, at least. You don't see the stats you gain, but that's in the near future.
~ At the end of every level you either gain a unit or some money. I want to add special items in as rewards, but I need to add more content first.
~ (mostly done) revamping the unit growth rates. Tried to explain it in a succinct way, but can't phrase it easily. Point: it should make characters generally balanced (more equal good + bad stats), while only tipping chars towards their class (no armor knights will have bad defense, for example, but also won't have awesome speed. Could have medium of one or both, though). This only applies to growths - how they wind up is still totally random.

Next steps, codewise:
~ Better reward logic
~ Thieves: they should be able to steal and unlock chests. This means adding a chest terrain type, and a couple of new skills.
~ I'm going to want to refactor the skill targeting code. This should allow me to generalize my targeting to allow:
  ~ bards only can target units who have used their current turn
  ~ clerics can only heal units who have taken damage.
~ I'd like that to support targeting open spaces, too, and other more advance logic options, but that might not happen yet.
~ Abstract chapter goals. Basically, these would just be test - at the end of the turn, is this complete? If it's the Main Goal, then the chapter is over. If it's a side goal, it's just added to the list of completed goals. At the end of the chapter, any goals that are completed apply their rewards to you.
~ Actually implement unit promotion. Each class is going to have a minimum level you can promote at, and you can promote units at any time between chapters.
Next Steps, Designwise:
~ Finish promotion plans.
~ Finish my list of items
~ Dream of some cool goals. Ideas: finish in X turns. Finish without killing enemy X. Kill the boss after killing all other enemies. Don't bring unit X. Survive for X turns. Protect neutral unit X (this one would require some refactoring, cause I'm assuming 2 teams right now, but should be doable...).
~ Think about Supports, and recruiting enemies mid fight. How do I communicate these? How do I choose who can recruit? How can I make it feel interesting, not random? If enemies are not standard units, but are enemy specific classes, how does this work?


That went on longer than I planned! Lots to do.