Lostcast 157: Learncast


  • LDG


  • Tiger Hat

    It’s not particularly fleshed out - but perhaps you could de-linearize AWL2 by wrapping a Cludo (Clue in the states) like hub.

    This also gives you a final boss pass (this would replace the accusation round at the end of the game).

    Perhaps you need 3 unknown tokens, and you get a token at the end of each level. You’ll need to hit at least 3 dungeons to finish, but it may be more.

    Like I said, not quite fleshed out - but I think its fairly unique.

    Thanks heaps for the shout-out too guys :)


  • LDG

    aw crap I just realized I didn’t add Poker Kingdoms to the show notes

    Vote it up on Greenlight!
    Download on Android!



  • Hey guys, here’s a totally unimportant detail, but since it came up in the podcast, I figured I’d post.

    In Zelda 1 you do need to buy the arrows (they’re 80 rupees, at least in the shop on the eastern coast of Hyrule) in order to actually make use of the bow. When you shoot an arrow, after you’ve bought them, each shot costs 1 rupee.

    Anyway, good episode as usual. :)


  • Jammer

    Geoff tried Phaser eh. No love for Melonjs? :(

    Nah just kidding, you’re quite right about phaser being well documented, and that it has a lot of examples. It’s something I want to improve with melonjs, just it’s hard to find time of course. I do feel like some aspects of melon are also less than intuitive or even tricky to grasp. Not entirely certain how to fix it yet, but potentially redesigning how the scene graph works is one idea.

    I’m not super big on how one codes with Phaser for some reason, but I agree it’s easy to pick up.


  • Tiger Hat

    Can I mount a defense for the blueprints/hostages from AWL1?

    Personally, I was a fan of those mechanics. It was a good way to feel like you accomplished something after every run, even if you didn’t quite make it all the way to the end.

    As far as the blueprints are concerned, I’m a fan of being able to roll with items that you know you like before the run even begins. For example, I know I won’t have to spend any time walking around at normal speed like a plebian–speed upgrades all the way, baby! I know this isn’t universal, but I happen to be the type of player that likes to plan a good build, achieve it, and then execute on said build. I like that my ability to do this increases with each blueprint I acquire and with each hostage I rescue.

    Also, maybe I haven’t rescued enough hostages yet, but I haven’t experienced any devaluing of the gold I find during a run. It seems like charms are always worth sinking money into as you go through the game (they are, right? I’m assuming their effects stack.), and I still have a ton of blueprints to purchase.

    Anyways, just my two cents. I guess the takeaway here is that I hope to still see some sort of overall progression mechanic (progress that sticks between runs) in AWL2.


  • Tiger Hat

    I like the idea of some kind of hub, I think since the last time, I still think you should double down on a more Zelda-like experience. Instead of the overworld though, I keep thinking of it being a mansion like the Clue board game. So the hub I guess would be the great hall or foyer or whatever the front is called. Then you could go to different rooms, kitchen, library, billiard room (? hehe clue), basement/wine cellar, garden (out back), dining hall, various bedrooms, etc.

    I also like the idea of being able to experience the content in any order (at least to some extent). I’m ashamed to admit in AWL 1 I’ve only ever killed the Zombie Warlord once and died shortly after. Granted I haven’t played that much in general, but you are competing for my time with the other 140+ games in my steam library.

    Also, silly side note, I thought that in Zelda 1 they put invisible walls on everything but the old man cave, because you can’t go alone… maybe I’m remembering it wrong.

    The over engineering and “familiar code stink” are the usual issues that I have as well. Phaser is nice, it’s fun to code in and pretty easy to use. It’s got a lot of built in game mechanics and a ton of bells and whistles. Almost to the point where it can be a little bloated. Probably one of the reasons Lazer (Phaser 3) is being rebuilt from the ground up. You can get a lot done quickly however when you just accept the way that it is and go with the flow (just like Unity). I haven’t worked on anything that I’d call large scale, so I am a little nervous about it falling apart somewhere down the road, but it’s open source javascript so hopefully that means it won’t be irreparable.

    I have looked at melon before, though can’t really tell you why I haven’t dug in. Prolly for the same reasons, the silly ones, getting hung up on the codebase for some stupid reason or another. I’m sure if I were to take the time like Geoff did with Phaser and just dig in, it’d be nice as well.

    The other issue is that I like solving engine related code puzzles, but at the end of the day, I made the wheel myself, it was fun, but I could have just used the other wheel that was already built and driven down the road already.


  • Tiger Hat

    @fallse7en said:

    Personally, I was a fan of those mechanics. It was a good way to feel like you accomplished something after every run, even if you didn’t quite make it all the way to the end.

    Mhm, I also enjoy taking my lvl70 Diablo 3 characters through massively under-powered areas and just obliterating them. I’ve heard many people say part of the RPG inspired leveling system they enjoy is coming back to parts of the game that were once challenging and just letting loose.


  • Tiger Hat

    @Warspawn said:

    I like the idea of some kind of hub, I think since the last time, I still think you should double down on a more Zelda-like experience. Instead of the overworld though, I keep thinking of it being a mansion like the Clue board game. So the hub I guess would be the great hall or foyer or whatever the front is called. Then you could go to different rooms, kitchen, library, billiard room (? hehe clue), basement/wine cellar, garden (out back), dining hall, various bedrooms, etc.

    To expand on this, maybe the first time you complete a room, you get a “room key” that you can then use to unlock another room? So you start the run with one “room key” which you can use to unlock whichever room you want to begin with, but you need to finish that room in order to get the next key so you can unlock another room. Dunno, though…that really gives the feeling that Raga is searching for something, which also implies that he doesn’t necessarily need to go through all the rooms to find it. Is that a good thing? I don’t know.

    Actually, come to think of it–do we know anything about what the story/plot for AWL2 will be?


  • Jammer

    @fallse7en said:

    As far as the blueprints are concerned, I’m a fan of being able to roll with items that you know you like before the run even begins. For example, I know I won’t have to spend any time walking around at normal speed like a plebian–speed upgrades all the way, baby!

    Maybe it could be cool just to allow trinket slots? Maybe you earn more trinkets as you go, and you can apply a certain number at the beginning of a run? Sounds fun to me :D

    I’ve always been a fan of having some way I want to play, and random drops aren’t always the best way of achieving that.


  • Tiger Hat

    @fallse7en said:

    Actually, come to think of it–do we know anything about what the story/plot for AWL2 will be?

    I don’t know if it was final or if I’m even remembering it right, but I think Raga opens a pandora’s box of curses or something and then has to go clean it up kinda Sorcerer’s Apprentice style. (Wizard is out wizarding, hurry up before he gets back).

    The keys are interesting, though hard to say if it’s Raga’s house too, then why are all the doors locked. Maybe more Zelda / Mega Man like @richtaur mentioned, go get some thing to beat some other thing.

    But then again, maybe some rooms are under lock and key.


  • Jammer

    @Warspawn said:

    So the hub I guess would be the great hall or foyer or whatever the front is called. Then you could go to different rooms, kitchen, library, billiard room

    Maybe the hub is Raga’s room in the center of the house? I like the idea of Raga being in a centralized location, having choices all around him- it’s inspiring. Maybe every new run, the rooms are all switched around. The first time you play, one track is to start at the library, which leads to the kitchen, which leads to the, ummm, garden? Yeah- and maybe the next time you load up, the cellar leads to the kitchen, which leads, umm err, somewhere else. Anyway, you get the idea. I envision short stints where maybe you go through a dungeon, but then you are given a choice to either go to the kitchen, or the library, and maybe you can see your path laid out before you so you can just start in someplace new and hopefully end somewhere familiar. Maybe difficulty can ramp up over by how many dungeons deep you are. But honestly this was complicated enough to type, let alone implement into a game, so yeah…

    Anyway I’m excited to see where this “overworld” concept ends up, it just sounds like it’s already headed in a really cool direction.


  • LDG

    @Ceric Thanks! I was right, eat it @richtaur!

    @agmcleod The documentation does help quite a bit. For this particular experiment, I wanted something that had everything I would need in order to package up the game for licensing. Flawless mobile support with regards to viewport sizing, audio, etc, etc. I’m not saying MelonJS doesn’t have that stuff, but I know that Phaser has been used for oodles of those kinds of games. Agreed on the Phaser style, though; it’s not my favorite. I much prefer something a little more ECS based.

    @fallse7en I definitely think there were some good parts of blueprints and hostages. It’s good to hear from your perspective, though. I tend to be overly critical of our work. Charms are a great gold sink, and the blueprints do keep you purchasing for a quite a number of runs. I think that it really bothers me that the eventuality is that gold does cease to matter. It’s a tough problem and, to be fair, it takes a lot of playing to reach the point where gold breaks.

    @Warspawn I really like your Clue examples. The hub is a big draw for me right now, and a large part of it boils down to non-linear progression. Regarding code puzzles: I’m the same way. It took a lot of willpower to sit down and say: “No, I’m going to use a tried and true solution here because my goal is shipping and low maintenance costs”.

    @Vox Good points about having your preferences. We’ll definitely try to keep that in mind with progression.

    Lots of awesome feedback here, everyone. As always, thanks for taking the time to post your thoughts!


  • Jammer

    @geoffb didnt know phaser has viewport scaling. It’s something we added in melonjs in the last year. I have to admit I haven’t used it very much myself. Might be worth considering with how ive changed the snowball game.


  • Patron

    Wow, so much stuff to comment on!

    • Guys, have you ever thought about doing a sort of Metroidvania-Roguelike (is that even a thing?)?
      You could have some fixed obstacles, something like, IDK, the fifth room you enter in the library is always going to be the “locked door room”. Then you could have an item specifically designed to overcome that in another dungeon… you know, standard Metroidvania stuff.
      The difference I see from a sort of “standard roguelike design” is that instead of having a bunch of fire-spitting wands, for example, you would have a fire wand designed to overcome obstacles in the ice-themed dungeon. So, a smaller selection of hand-curated items made with a specific puzzle in mind. Or perhaps just having themes in items in general, say, fire-based wands for the ice dungeon or something.
      You could also use the procedural generation itself as a puzzle in that Metroidvania aspect. In Castlevania: Harmony of Dissonance, the shops are only accesible under certain conditions. I feel that they did that in a really bad way tho, as those conditions are pretty arbitrary and unpredictable. However, being the smart designers you are, you could make things like, IDK, something like the Forest in AWL1, but make the entrance only accessible when certain items are equipped…
      Those are just some silly ideas… in general, I really like the idea of an open-world roguelike, tho. This sense of freedom and exploration… discovering which dungeon I’m supposed to go first, where a certain item is useful, going into a dungeon and having your ass kicked, only to find weapons in another dungeon which make it super easier to beat…
      I feel like that’s way more exciting than floor-based progression.

  • Patron

    @agmcleod said:

    @geoffb didnt know phaser has viewport scaling.

    Yeah, it sort of works…
    Just don’t expect it to be pixel-perfect with your pixel art, as it’s CSS-based, IIRC.

    If you want to make, say, a Low-rez Jam game, you’ll have to scale it yourself.


  • Jammer

    Ah, it just stretches the canvas? I was more wondering if it will zoom out from the game view depending on your settings.


  • LDG

    @Josue yeah love that approach. We’ll almost certainly have elements like that but it might be more along the lines of abilities, monster possessions, keys, stuff like that. Especially if we do nonlinear level progression we’ll probably have stuff like that to unlock certain things or pave your path. Sounds fun!


  • Patron

    @agmcleod said:

    Ah, it just stretches the canvas? I was more wondering if it will zoom out from the game view depending on your settings.

    It has various settings, actually.
    I think you can have it maintain the scale but change the viewport size to fit the screen, make it round the scale to the closest integer or have it scale the X & Y axis individually.


  • Jammer

    Ah okay, yeah that’s what I figured :)


Log in to reply