next »

[crusty]

Caravan
website

Caravan is a 2D roguelike RPG with monster collecting, text-based encounters, and a turn-based squad battle system.







Recruit, buy, and enslave all sorts of characters







Journey through a procedurally generated tower and interact with the world via text-based encounters







Test your squad in turn-based battles







Fun beach hangs

[HoffmanIV]

This looks pretty darn cool! Are the elemental type labels just placeholder sprites for now? They look a touch too close to what the Gen III-V games used.

[crusty]

This looks pretty darn cool! Are the elemental type labels just placeholder sprites for now? They look a touch too close to what the Gen III-V games used.

Thanks!  Yeah those are just placeholders The whole UI is probably going to change

[DireLogomachist]

Turn-based pokemon battle roguelike! Sign me up!

[ActualDog]

Cool stuff, this is looking pretty slick. Excited to learn more about Westley Burgers and his quest...

[andyfromiowa]

Very intriguing concept.  You have a great imagination for character/creature design.  Will definitely be following. 

[Ashedragon]

Safe to say my interest is piqued. Looking forward to it see how it progresses!

[crusty]

Thanks for checking it out everyone and for the kind words!  It's really great to hear people are interested in the concept.

Excited to learn more about Westley Burgers and his quest...

Westley's Quest...

...

...

Sorry I zoned out there.

For my first dev post let's talk about... layers in Unity!  I am using Unity for this game, and more importantly I am currently almost exclusively using the Canvas UI system to position + display everything.  It made sense to me at the beginning because of the ease of scaling/pixel perfect/whatnot.

However, today I am implementing some real lighting for the first time and I've just run into an issue where my Light's culling mask setting does not seem to respect the layer of a Canvas object's child.

The Westley on the right is the child of a Canvas and he ain't gettin no lighting love!


Here's the scene hierarchy:


And here's the light (IMPORTANT PART IS THE CULLING MASK FIELD!):


Here's the right hand Westley and his parent Canvas (NOTE THE LAYER!):
 

And finally the left hand Westley and his parent gameobject (NOTE THE LAYER!):
 


So hopefully that makes my issue clear.  I need to be able to control which gameobjects inside a Canvas UI receive lighting.  But the Canvas object's layer seems to override all the child layer settings.  Maybe this is intended behavior?  Does anyone know what's the deal with this?  I looked around on the internet but couldn't find anything.  Is there a different way to cull objects for lighting besides layers?

For my specific use case, I can hack the desired functionality by swapping render materials at runtime... but I don't wanna do that!

[JamesRossi]

Oooh, a fellow Oregon Trail-like dev. What kind of gameplay time are you aiming for? Like 5 minute battles and maybe 1-2 hours for a full game experience or something longer than that?

[crusty]

Oooh, a fellow Oregon Trail-like dev. What kind of gameplay time are you aiming for? Like 5 minute battles and maybe 1-2 hours for a full game experience or something longer than that?

Haha hey man!  I never posted in your thread, but I like what I saw of apocaleague a lot

I'm still a ways off from knowing how long the battles will/should take.  I am hoping I can make quick battles work (1-5 mins), but it may be a difficult thing because I am asking for a lot sticking a turn-based battle system into a roguelike.

For example, if I want quick battles, but I don't want the player's squishy characters getting one-shot, then that means I'll probably have to weight combat math in favor of the player such that enemies will do a lot less damage relatively.  That way player character deaths (it's permadeath btw if that wasn't assumed by roguelike) will feel justified because the player made the choice to enter a battle with a low health unit.

I have a resource-based healing mechanic (heals consume FOOD) that will hopefully allow for this war of attrition type vision I have.  But honestly I don't know yet if it will work.  I want to reward brinksmanship with respect to healing and damage mitigation.  Battles are the most profitable encounter (on average) in terms of loot/upgrades, but they come at the inevitable cost of health.  Healing too often as a result of poor battle strategy will mean you have to forgo the more profitable battle choices once you run out of FOOD.  Healing too sparingly means you are not picking enough risky choices and FOOD will be wasted (after each encounter, the Caravan makes camp and eats whether they are damaged or not).

On top of that, I have built a battle system with mechanics which will (again hopefully) force a balanced "holy trinity"-style team composition (tank, healer, dps).  I don't yet know how my battle math will work out, but there will probably be a sweet spot (time-wise) that will be just long enough to still allow for that level of strategy.  I just hope that sweet spot isn't a 20 min tank-fest...  Otherwise I will have to make big changes.

As for one playthrough of the game, I am going for an FTL-ish length.  I have a TON of ideas for the metagame loop which will encourage replayability and new strategic choices so that even when facing certain defeat the player may yet have a reason to see the run through to the end -- and on the other side of the coin winning a playthrough shouldn't feel like the end of the experience.

[JamesRossi]

Haha hey man!  I never posted in your thread, but I like what I saw of apocaleague a lot

Thanks! I appreciate it!

I'm still a ways off from knowing how long the battles will/should take.  I am hoping I can make quick battles work (1-5 mins), but it may be a difficult thing because I am asking for a lot sticking a turn-based battle system into a roguelike.

For example, if I want quick battles, but I don't want the player's squishy characters getting one-shot, then that means I'll probably have to weight combat math in favor of the player such that enemies will do a lot less damage relatively.  That way player character deaths (it's permadeath btw if that wasn't assumed by roguelike) will feel justified because the player made the choice to enter a battle with a low health unit.

This is something I've been thinking on lately. I actually also may include a turn based combat system. I'm thinking of having it be more along the lines of early Might and Magic combat where your team takes a turn moving all their units, then the other team. Not sure if you plan on that or not as well. Due to the potential number of combatants with a baseball team (10-15), I was imagining something where you can never have more than 5 units, and often your are limited to even fewer. I was thinking perhaps this will work if it is yet another risk/reward scenario where players who enter combat are likely to gain stats. So you might have an All Star lineup that can win every battle, but the returns would diminish and your other players would suffer from a lack of leveling up. With the pokemon aspect here, you could run into a similar problem of larger numbers of potential units, and the struggle to prevent abuse by overpowerful units steamrolling everything.

1-5 minutes definitely feels like the right range for the minigames. I've been relating to them like the Hunting event in OT, where you sacrifice a resource of bullets to gain food, so I've been very active trying to follow that example. An alternative I have been contemplating to combat, (or in addition to but I am trying not to overscope) is having a series of less in depth minigames, closer to the complexity of Hunting. In my game for instance having something like Spyhunter on NES, using the bus. Play for a few minutes, determine and outcome based on performance, continue on. You have the Beach Party thing, and I actually like the idea of finding little moments of relaxation, although, while this is kind of a jokey suggestion, I wanted that Golem to kick the party into overdrive with Rock Bash, and just pull out a boombox and play some classic rock. Using special abilities at a Relaxation point for an additional bonus of some kind could be interesting?

I have a resource-based healing mechanic (heals consume FOOD) that will hopefully allow for this war of attrition type vision I have.  But honestly I don't know yet if it will work.  I want to reward brinksmanship with respect to healing and damage mitigation.  Battles are the most profitable encounter (on average) in terms of loot/upgrades, but they come at the inevitable cost of health.  Healing too often as a result of poor battle strategy will mean you have to forgo the more profitable battle choices once you run out of FOOD.  Healing too sparingly means you are not picking enough risky choices and FOOD will be wasted (after each encounter, the Caravan makes camp and eats whether they are damaged or not).

And I should have read the whole post before commenting because you do have battles be upgrade moments. That sounds like a good way to manage the health/food resource dynamic. Would it be like setting your rations? So you can dictate the speed of recovery? Would be interesting if Low Food options were good to preserve resources when units are fully healed but perhaps at the cost of a morale/some other factor.

On top of that, I have built a battle system with mechanics which will (again hopefully) force a balanced "holy trinity"-style team composition (tank, healer, dps).  I don't yet know how my battle math will work out, but there will probably be a sweet spot (time-wise) that will be just long enough to still allow for that level of strategy.  I just hope that sweet spot isn't a 20 min tank-fest...  Otherwise I will have to make big changes.

As for one playthrough of the game, I am going for an FTL-ish length.  I have a TON of ideas for the metagame loop which will encourage replayability and new strategic choices so that even when facing certain defeat the player may yet have a reason to see the run through to the end -- and on the other side of the coin winning a playthrough shouldn't feel like the end of the experience.

One thing I like on your concept is that your tower gives an excellent goal, while also allowing you to arbitrarily change game length. Short game, short tower! And there's also variation possibilities where you could theme the towers after certain aspects, to create inherently different experiences based on the same core concept.

All in all, this is looking really cool and I look forward to seeing this develop!

[crusty]

Thanks for the feedback James!  Good stuff here.

I was thinking perhaps this will work if it is yet another risk/reward scenario where players who enter combat are likely to gain stats. So you might have an All Star lineup that can win every battle, but the returns would diminish and your other players would suffer from a lack of leveling up. With the pokemon aspect here, you could run into a similar problem of larger numbers of potential units, and the struggle to prevent abuse by overpowerful units steamrolling everything.

This is something that happens in games with party rosters which are greater than the max battle slots all the time and it actually drives me crazy.  Especially if they suddenly stick you with a certain character for story reasons... Looking at you, Banner Saga.  I don't think that choosing whether to spread the wealth to all your units, or power up only a handful is an interesting decision for the player.  Even if you could design a game where that decision had real impact (most games do not) it would probably be difficult... but maybe really cool?

That being said, depending on the game, I don't think it's necessarily a trivial thing to avoid.  But I am going to try.  So, currently in Caravan, you can take 1-4 units into battle, and you have 6 unit slots total. Units don't have levels. They grow in strength only through the application of talent points (which specialize their role in battle) and stat points (which make them better at a role).  Winning in battle will award the same amount of talent points and stat points to all the members of your party.  This makes sure that all six units are on equal footing, even if they are a niche unit that doesn't often see battle.  However, once you allocate those points, you can't respec.  So if you have a unit that you spec entirely for out of battle utility, they are always gonna suck in battle.

You have the Beach Party thing, and I actually like the idea of finding little moments of relaxation, although, while this is kind of a jokey suggestion, I wanted that Golem to kick the party into overdrive with Rock Bash, and just pull out a boombox and play some classic rock. Using special abilities at a Relaxation point for an additional bonus of some kind could be interesting?

Would it be like setting your rations? So you can dictate the speed of recovery? Would be interesting if Low Food options were good to preserve resources when units are fully healed but perhaps at the cost of a morale/some other factor.

Haha the beach is just a campsite.  Specifically, a campsite when resting on an Ocean biome.

The main game loop is:
MOVE TO MAP TILE -> TEXT ENCOUNTER -> BATTLE/SHOP/REWARD -> CAMP -> REPEAT

When the day goes to night, your caravan sets up camp.  This is when default food consumption + healing happens.  A time to check out the latest rewards from the latest encounter and gear your team up before the next day's journey.  It's the screen where you allocate skills, talent points, and stat points.  I intentionally wanted it to be small and intimate to create a juxtaposition with the massive scale of the tower.

I have thought about the rationing system you mentioned (like the one in OT), but for the vision I currently have I feel it doesn't present an interesting choice to the player.  The logic would always be trivial: set rationing to the level which will recover >= the missing health of your most important unit.  But the actual player implementation of that logic would put a burden of knowledge on the player: to know the exact conversion rate between food + health.  I don't want that.  Additionally it puts an extra constraint on how I scale rewards for the game.  But like you said, if there were some kind of morale variable in play it could be a totally different story.  We'll see if I need a tool like that in the future.

To address the boombox sesh: camping gives access to special "camp encounters".  These are text-based encounters that can be engaged only while on the camping screen.  An example is the humanoid engineer skill tree.  You can get a talent point that will give a unit the expertise to create a mech "barricade" unit every 3 days.  So every 3 times you camp, you will have the a camp encounter pop up that (if engaged) will give you the option to put a barricade unit into your party.  Barricades have no speed (so they get no turns in battle) but they start with ~100 hit points and 3 charges of taunt.  So they are essentially cheap disposable tank units.  There may be other camp encounters which are always available but offer some kind of resource trade.  For instance, a utility talent point that grants the ability to scout nearby map tiles at the cost of food.  So will there be a boom box session?  Yes, now there will be.

 

[crusty]

OH mm.. mmm.. me?!

You look at m... m.. my devlog? THANK YOU. Here's an update:

I've been working on refactoring some battle related scrimp for the past 2 weeks!  It's a little bit tedious and unfortunately not too interesting to look at but I have opened up a whole new world of player feedback while targeting skills! I can now do things I could never do before to show the player all the relevant skill modifiers and effects of casting a skill. Here's two skills that require a bit of extra scripting that can displayed properly during user targeting now.

[Buckshot] - this skill does more damage and has a greater area of effect the closer it is to the caster!



[Grondr's Charge] - this skill moves the caster toward an area, gaining charges of TAUNT for each enemy unit in the area



Some things to NOTE:

A lot of the visuals are still janked up and/or placeholder.  Notice that the colored arcs under the unit's feet are very bright, what the heck.  That's because Unity is a bad boy.  I posted about it earlier.

What is TAUNT?  It restricts movement on the battlefield!  Units cannot move past other units with taunt.  Many attacks also have a ranged component (solid line vs. dotted line).  The ranged component can target past taunt, but the movement component cannot.  TAUNT controls who can attack who on the battlefield.  You can remove charges of taunt by attacking that unit.

Did you notice the 2.0 x modifier on Buckshot?  That is granted by an EFFECT owned by the caster.  The effect is called Steady Aim and grants double damage.  Upon receiving damage, this buff is lost.  It is hard to see because the placeholder visual is very dark, but above the caster's head the Steady Aim icon appears when targeting Buckshot.  However it does not show up when the same unit is casting Grondr's Charge.  This is because only the EFFECTs which mutate the skill will be shown.  There is a lot of information flying around in this game, so I am striving to always give proper feedback when passive EFFECTs come into play.  If more than one EFFECT mutates a skill, they will be shown in the same popup in a radial fashion (mousing over enemies can also pop up EFFECTs above their heads if they apply!)

[bacon]

This looks super interesting! I dig the mixing of contexts, like moving from the tower, to battle, to hangs. Can you say anything about the world? I'm trying to figure out if there is a guiding aesthetic or narrative behind it. So much of what made Pokemon and Oregon Trail appealing is their very distinct sense of place.

[crusty]

This looks super interesting! I dig the mixing of contexts, like moving from the tower, to battle, to hangs. Can you say anything about the world? I'm trying to figure out if there is a guiding aesthetic or narrative behind it. So much of what made Pokemon and Oregon Trail appealing is their very distinct sense of place.

Thanks!  Yeah here's the story such as it is so far:

     Caravan is set on a strange alien planet on the verge of catastrophe.  A deadly gas is slowly rising from the surface of the planet, destroying all life.  The presence of ancient, massive towers throughout the land seems to suggest that this sort of apocalypse may have happened before, and these hastily-constructed monoliths are all that remain to tell of ancient exoduses.
     The player’s journey begins at the base of one of these towers.  Each new game the player chooses a starting roster comprised of one humanoid to lead, one beast to pull, and one carriage to bear.  As the player guides his charges upwards to safety, the small party will grow to encompass all manner of alien denizen, fearsome beast, and fantastic contraption, for this tower holds many secrets -- its rickety form supported by magics both technological and arcane.  The caravan will journey through many lands: from primal jungles ruled by shamanistic eggmen to vast mountain ranges comprised of electronic hardware and patrolled by sentient clouds.  Though these exotic lands undoubtedly hold many treasures, the player must always beware, for in trying times the world becomes a dangerous place and he will be faced with many difficult trials.  The player must keep his caravan well-fed, safe from harm, and always on the move.

And here's a passage describing some ideal gameplay:

     The caravan is starving.  The player can see a node higher up the tower which guarantees a food bounty.  The player must path through an ocean node to reach this, but upon departure the caravan is halted by a popup revealing that one unit (the party's tank, a big rock dude named Truck Bignuts) had a hidden trait, hydrophobia.  Truck Bignuts refuses to pass over the water.  The party must decide to either go around the water node (potentially starving to death) or leave Bignuts behind.  But the enlightened player will realize there is a third option!  A unit on the team has a talent called vicious, which allows the unit to engage in one-on-one battles with his own party members during camp.  A talent which was previously undesirable turns out to be the player’s salvation.  The player decides to pit the vicious monster against Bignuts and takes him down, salvaging his corpse for meat.  With the surplus of food gained by killing Bignuts the team can afford to take a more risky path that leads to a map node with a guaranteed treasure bounty.

There is a great wealth of lore to be discovered, the plot of Caravan is kept intentionally bare bones.  Caravan is rightfully all about the journey, not the destination.

[Bricabrac]

I like both pokemons and text-based games. Following! :D

[crusty]

I like both pokemons and text-based games. Following! :D

:D

[Pixel Noise]

Oh wow - this is great. Really like the website too - very nice work! Super following 

[crusty]

Oh wow - this is great. Really like the website too - very nice work! Super following 

THANK YOU!

Also it's really cool to hear you liked the website!  I built that because I didn't have a great way to show off the game in job interviews, and I think I bit off a bit more than I could chew web-dev wise to be honest...  So I'm glad it worked for you.  I know there are quite a few browser compatibility issues with it that I just haven't gotten around to yet.

[crusty]

I finally built an editor window for all my CSV (comma separated value) files.  There's a lot of immutable data in this game: stats, skills, etc.  Editing it manually in the files is painful because of the many enum values that I would have to remember mappings for (e.g. EARTH=3, AIR=4, etc).  The editor replaces those fields with the appropriate dropdown list.

Another added benefit is alphabetization and automatic compilation to the master runtime database.