Lostcast 189: Platform Percolator


  • LDG



  • Hi guys, long post coming. TL:DR version, I think you guys make good games but you’re missing major elements that keep them from sustaining player interest over time. I definitely don’t mean to speak for anyone other than myself here in the following post, though.

    Now, onto podcast feedback and my very long post. :)

    First, for indie games in development distribution, you might want to consider itch.io’s Refinery: https://itch.io/refinery

    It has some tools for creating your own version of early access and apparently they’re working on Steam integration.

    Second, for HAXEFlixel, I tried it out but it took me forever to set up a working dev environment on Windows. I kept trying to get it to work with Flashdevelop. It seems very cool, but as a relative novice dev, it was painful getting started. I did get it to work when I set it up with Sublime Text. I got the Hello World app working. That was my 4th attempt. In general, though, HAXE and HAXEFlixel does seem like tech with good possibilities.

    Now for my long analysis of games I like, why I like them, and what about LDG’s games has made me bounce off them.

    In looking at the games I’ve played many hours of (many of which have been massive successes in the market), they seem to have a few things in common. I’ll preface this by saying that one of the things that attracts me to a game is its setting and storyline. That’s my personal preference. Puzzle games, generic shooters, platformers, are all games that are just so-so to me and I typically don’t spend money on them.

    Here are a few examples of games I’ve played a ton of. Note I purchased some of these games before they became big hits, others after they had been on the market a few years. This is to point out that marketing wasn’t specifically what attracted me to these games.

    1-Darkest Dungeon: Single player dungeon crawler. Has procedurally generated maps but defined visual themes for each area. Has a roster of characters that you maintain and steadily upgrade over the course of the game. Individual character deaths are permanent, but there are always new, low-level characters available. Has a number of set-piece bosses that you fight tougher versions of. Enemies are roughly similar but a few new ones are introduced in later levels/harder dungeons. You can grind for “trinkets” (special gear). Mix of classes in parties, all of which with different abilities, means, and you must complete missions to get money to upgrade your hamlet/village, which is persistent. Hamlet buildings give you upgrades for your roster of characters, enabling them to tackle harder dungeons/levels. DD has some unique elements, e.g. the “stress” mechanic, but for the most part it’s a well-done, horror-themed fantasy RPG with semi-procedural levels and steady progression. Major success, around 800,000 copies sold (via Steamspy). I did not back the game but did purchase early access as soon as the game was available via Steam. Matt bounced off it, so I get that it’s not for everyone.

    2-Stardew Valley: It’s a Harvest Moon clone with a persistent farm that you manage, continually gaining new crops to grow, unlocking new tools and buildings, upgrading your tools, unlocking access to new areas, building relationships with the people in town, etc. Single player. Huge success, 1M+ copies sold, etc. You can customize your character’s look and gender and customize your house and overall farm layout (what you grow, where buildings go, etc).

    3-Caves of Qud: Massive procedurally generated RPG. Single player. The quests and a few areas are the same each time, and the locations are in the same places on the overworld map, but the individual maps themselves are procedurally generated. Tons of character customization options and abilities, tons of gear to get, crafting, etc. The game’s graphics are a step above ASCII, though, Also, although this is a roguelike, that’s not actually an aspect of the game that I believe serves it well with its massive world, steady character progression, gear to acquire. Although I really like it, I’d prefer if the game had checkpoints at the very least, if not outright saving. The game has sold around 10,000 copies (via Steamspy). Great game, made by 2 guys.

    4-Warframe: Playing this a ton lately. Sci-fi action third-person looter shooter, F2P MMORPG. You can play solo or with a group. Has procedural levels, tons of gear you level up and customize, tons of warframes (basically androids you operate remotely and can customize) to grind to get (you can get blueprints to build them via ingredients/materials you get from enemy/boss drops and completing missions), etc. If has an overarching storyline with “quest” missions and all other missions are repeatable variations on specific mission types (defense, survival, exterminate, etc.). The game seems to be successful. The studio started out as independent but is now (I believe) partially owned by It has more than 12 million owners on PC alone and is on Xbox and PS4.

    5-Defender’s Quest: Lars Doucet’s game. Played this a lot. Again, it has a story, funny characters, a semi-unique setting, a sense of progression with persistent characters that you get new gear for, repeatable missions, the tower-defense-style gameplay that he mentioned (I’m a fan of tower defense games), etc. Seems like it’s been pretty successful for Lars.

    So, from all of that, here are the key common elements that keep me coming back and playing each of these games (or at least having me put hundreds of hours into them).

    1-Bite-size (20-45 minute) repeatable missions that get progressively harder in new regions/areas. Stardew Valley is the exception, but you get random “quests” in the game and have some objectives (get to the bottom of the mine, bring this item to so and so, etc.).

    2-Interesting, strongly defined settings with lore and unique atmospheres. Even Stardew Valley has a detailed town and surrounding area with its own character/feel, history, and a range of inhabitants with personalities, quirks, etc. The other games have in-depth sci-fi or fantasy settings. Besides the basic gameplay for each, each game’s setting is actually one of the reasons that put me over the top and got me to buy the game (or spend $5 on in-game currency in the case of Warframe).

    3-Steady sense of progression and increase in character capabilities over many, many hours of gameplay. In each game you get better gear, increase character stats, find interesting items in the game, and all of this persists (except in Caves of Qud) regardless of success or failure. These games don’t really have game-ending conditions that force you to start from the beginning. Progress can stall a bit, and you can ultimately fail missions, but in general each time you play you’re making a little more progress each time.

    4-Lots of items/gear to find/obtain from missions (all games above). Gear can be improved and/or modified (e.g. Warframe, Darkest Dungeon).

    5-Unlockable regions, i.e. reason to go to new areas/levels and attempt harder levels/explore new content and find new stuff. Regions are often unlocked by completing in-game tasks/quests.

    6-Character customization. In each game you can customize your characters to some degree.

    As an aside, each game was made by 1 or 2 people or a small team save for Warframe. Darkest Dungeon, 5 core people. Stardew Valley, 1 guy who contracted with Chucklefish as publisher and for help with distribution. Caves of Qud, 2 guys.

    Warframe is the exception, but the developer and publisher, Digital Extremes, was an indie studio when they started and they had had a string of failures, got turned down by all outside publishers, etc. Now they’re very successful. (Edit: DE is obviously a large studio - the point here isn’t to say that a 2-man team could make a game like Warframe - it’d probably be tough to make a game like that with a team of 50 - it’s that DE came up with a unique setting, stuck with development despite difficulties, put out a minimum viable game, used a form of early access, and have largely been successful, and, as I noted, the game uses many of the elements noted here, which has attracted a loyal player base.)

    So if I had to diagnose some of the issues with past LDG games, at least as for why I bounced off them, I’d say that you guys tend to focus on games with gameplay that doesn’t scale over time and doesn’t have a unique, interesting setting/story. You guys have done 2 roguelikes and a small-scale, relatively generic fantasy-themed tactical RPG, at least as far as I’m aware of. None of these games have really strong, unique settings.

    In the games I listed above, the setting/story actually informs the gameplay and the design decisions the developers made. Without a setting/story, it’s easy to end up with games that don’t stand out or really capture the imagination. Now this does mean content. Not so much that a relatively small number of people can’t make it (e.g. Caves of Qud, 2 guys), but it does need content. These games do have longer development periods, as in ongoing for years. The majority of these games have also been early access works in progress.

    I’ve noticed that most of the game designs you guys have shown/talked about are more variations on games in existing genres without any kind of strong setting/theme/story. In most cases you could literally replace the graphics (note I like Matt’s graphics and art, by the way) with similar ones from a different, broad theme (e.g sci-fi, fantasy, horror) and the game itself wouldn’t change much mechanically or gameplay-wise. The mechanics are fairly divorced from the setting. I can’t say the same for the games I listed above, where those games’ systems are more related to the setting and each game’s respective storyline.

    With respect to story and lore, those games above don’t shove lore in the player’s face but rather make it discoverable and provide basic motivations for taking on missions/quests/tasks (short text intro explaining why the player is doing x, the mission’s goals, and why it matters). The Souls series (Dark Souls, Demons Souls, etc.) does a great job of implying story through the setting and making lore optional, but still having that lore inform the setting and game design (bonfires, souls, the grim setting and atmosphere, etc.).

    So, all of that said, and my goal here isn’t to bash LDG’s games - I think your games have fun gameplay, have good art, etc. - it’s to point out where they fall down for me. I feel like you guys are getting 2 out of 4 when it comes to key things that make many games successful. The 2 things you have so far are really good, but the missing 2 things lead to the games not rising to “hit” or “indie success” status in an already-crowded marketplace.

    For example, with AWL2 Soul Thief, the setting is a semi-medieval house. It’s cute, but nothing special or interesting. Raga is a great character, but that’s the one unique element in the setting/story. Next, there’s no sense of progression except through the small number of levels in the game. If you die, you’re dead, due to you guys going with permadeath. So ultimately if you suck at it, there’s no sense of in-game progression, no new gear to find, no stats to upgrade, no way to permanently unlock new areas, so there’s nothing to keep me coming back other than banging my head against the same levels again. If I had been forced to start over but kept some gear, or improved Raga’s stats, I would have kept playing.

    So, all of that said, have either of you come up with a world/setting and characters, or at least a strong, unique setting theme, before coming up with a game design? Maybe a different approach to game design is in order now.

    Note that Matt’s prototype is pretty cool. I haven’t had a chance to play the latest build, but it does have a lot of the elements I’ve noted above. I’m planning on checking out the latest build and providing feedback.

    Ultimately, as you both said on the podcast, it’s up to you to make the games you want to make, so you can safely ignore everything I said above if the idea of making a game with a story/world makes you want to jump out of a window.

    Anyway, I hope some of that is useful and provides some insight into why at least one player has had some issues with your games.



  • With regards to Haxe… I like following haxe.io – generally a good source of information about the goings-on of the larger haxe community, which includes interesting news about different libraries and platforms (gamedev related more often than not).

    Occasionally they’ll do reports on games made with haxe (organized, generally, by library/framework) after a game jam, like the ludum dare, ex: https://haxe.io/ld/34/ – the best part is, as part of the competition, you get access to the source for all those games!

    Interestingly, HaxeFlixel seems to always have the lions share of games there, which might speak to how good/easy it is to learn and use!


  • Jammer

    @Ceric awesome post to read there. I find your thoughts on those different games and what applies to AWL and such to be rather interesting. Stardew valley is the one we have in common, and I have to agree with you 100%. There’s a pretty clear path of progression, and mechanics to work with. It even has combat if you’re so inclined. There’s a number of story lines that have popped up i have yet to explore as well.

    Anyways, here’s my notes from this weeks episode:

    • Inheritance. JResig is a mixed bag, I really just recommend to anyone needing inheritance to just load up babel these days. In terms of inheritance depth, do you have just one base class, or do you have BaseA > Sprite > Entity > Raga type scenarios? I’ve definitely had this, and MelonJS still has it to some degree. More than being a language limitation, as a design pattern it’s rather painful.
    • If “Rogue-like” is played out as a term, perhaps we need to move onto: “Rogue-ish” ;)
    • Meeting needs with an app: I listen to an iOS/Mac development podcast called Under the Radar. Neither of the two devs have worked on games, but they often bring it up as a difficult sounding market. With regular apps though, they have talked about issues of knowing your market, how do you differ from the competition, stand out, how to monetize, when to release, etc. It has a lot of the same marketing/eye ball issues that games have. Though I agree with you guys on needing a vision and focus for what you want to build. Solve a problem with your game. What game is it you’re missing in your life?

  • LDG

    Great points @Ceric, thanks for taking the time. In short, it sounds like you’re interested in games that are built with the world in mind first. World-building is super cool and a lot of fun … when it comes to story and stuff, I think Geoff and I are interested in the game design being first, and informing the other elements. But who knows, nothing we’ve done so far has been great, so maybe our intents need tweaking.

    I think what would solve a lot of these problems, and what I’m really interested in, is nailing the meta game. Most of our games seem to have relatively satisfying core gameplay (throwing swords, casting spells, etc.), but they largely fail to retain players as you were saying. Part of me wants to shore up those meta game skills with a ridiculously simple micro game (think “click to win”) wrapped up in a smartly designed meta game.

    Anyway good stuff we certainly have a lot to learn!


  • LDG

    @agmcleod said in Lostcast 189: Platform Percolator:

    do you have BaseA > Sprite > Entity > Raga type scenarios?

    Yep, there’s a Class > View > Sprite > Player > Raga

    It’s somewhat limiting but having shared libs makes it easier.

    What game is it you’re missing in your life?

    That’s the thing, whoooo knows? If you ask somebody what’s your dream game? They’d end up saying IDK Call of Duty meets World of Warcrafat or something. Inventing is hard. AND we have to temper that with “OMG I have to make this game” or it’ll never get done.


  • Jammer

    Yep, having a realistic goal for the game you want to play is hard as well. I think i want something like starcitizen, but it just sounds insane to actually finish. Though I also think such a big game would basically be a life style to play. My next idea I hope to prototype either end of this year or early next is because a game I’d like to play doesn’t really seem exist. Similar to your metroid meets zelda game that you wish existed :)


  • Tiger Hat

    @richtaur said in Lostcast 189: Platform Percolator:

    That’s the thing, whoooo knows? If you ask somebody what’s your dream game? They’d end up saying IDK Call of Duty meets World of Warcrafat or something. Inventing is hard. AND we have to temper that with “OMG I have to make this game” or it’ll never get done.

    I want to see:

    A game that leans heavily on Lemmings - there have been similar titles, but none have been as perfectly designed.

    A game that continues on from Populous 2 - Populous 3 was just an RTS, Black and White were too prescribed, and Reprisal harked back to Populous 1.

    A well designed survival game in the key of Minecraft, systems upon systems, but without all the crappy monsters, magic and parallel worlds, more Mining, and more Crafting.

    A version of Soul Thief where I can see past the second level :)

    More Magic Carpet

    I’m toying around with:
    An isometric racing game - a modernised version of ‘Racing Constuction Set’ basically trackmania from a fixed camera.

    A prison architect style game where you manage a dog rescue/pound.

    A real time squad based strategy game where you manage a fire-crew.

    Some weird space-ship game that looks like it may perhaps scratch the Magic Carpet itch.


  • LDG

    @salmonmoose said in Lostcast 189: Platform Percolator:

    A version of Soul Thief where I can see past the second level :)

    What happens in the second level? Too hard?

    A prison architect style game where you manage a dog rescue/pound.

    I like that you’ve got multiple concepts going. This one especially sounds cool to me. That agonizing decision – do we have to put a dog down? Which one? Could have some really impactful gameplay.



  • @richtaur No problem. The podcast is always entertaining and I want to see you guys succeed in game dev.

    I hear what you’re saying in terms of interest your (and Geoff’s) interests in game design/development. You guys seem more interested in systems development.

    One reason I’m suggesting having more of a setting/theme/story/world in mind to start is not so much because I think story is important to games that interest me (or games in general) but rather that in the Lostcast episodes where you and Geoff discussed AWL2, you went back and forth on the overall design and in-game systems many times (beyond a couple of core ones, specifically that it had a top-down perspective and procedural levels).

    That seemed to lead to a lot of wasted time with experimenting with really different systems. My suggestion is that maybe if you and Geoff have more of a core setting or theme to work with to start that it would narrow down the number and type of systems you experiment with and can therefore reduce total development time to a more manageable amount.

    My sense is that this is what happened with your current prototype. You had a definite setting/theme in mind to start and built strong core systems (platformer, the game editor, the metagame “Steam-like” marketplace) around it and now when you and Geoff discuss its development your conversations seem more focused on building out those existing systems and experimenting with system elements rather than scrapping entire systems and creating new ones as was the case with AWL2’s development. For example, I honestly can’t see how you could realistically scrap the “Steam” store or the editor without completely altering the game. Those two things are core to the entire concept of being a game developer (at least in a fictional sense. :) ). To me, that’s a sign of strong systems that align with the game’s theme and setting and are outgrowths of that theme/setting.

    Process-wise, having a strong theme/setting in place seems to be a more reliable way of both narrowing down which systems to create and making a game that stands out from others in the marketplace than others I’ve seen.

    One other method is to create a wholly new type of gameplay, but that’s 1) very difficult and 2) difficult to consistently make fun and therefore not easy to repeat, process-wise. Of course iterating on small games with different gameplay mechanisms could be one way to experiment (and be fun), but that takes time and doesn’t produce work you can necessarily sell. Game jam games can be great and potentially lead to interesting, larger games you can sell, but in and of themselves they (generally) don’t pay the bills.

    As for my other comments, the biggest issue for me (which I think you identified), apart from all others, is a lack of strong metagame, and for me specifically a sense of long-term progression throughout many playthroughs. Basically a reason to come back and keep tackling existing content (or looking forward to new content, e.g. DLC). That alone could go a long way toward beefing up interest in LDG’s games.

    How you address that (obviously) depends on the game, but in AWL2, I would have wanted to see the following:

    1-Starting out the game as a monster you possessed (encourages possession of new monsters - after you posses one for the first time it shows up in the start menu as a monster you can start the game as)
    2-Selecting a wand you’ve found when starting the game (again, encourages seeking out wands/items, after you find a wand, it is part of your inventory when you start the game)
    3-Permanently acquiring a small number of upgrades (extra hearts, extra damage, extra armor)
    4-Larger dungeon maps, which would have enabled scaling Raga’s survivability upward without creating massive game imbalances. This could help with the perception of a lack of content/short game length.
    5-Save game mechanic like Caves of Qud (save file is erased when you die, but you can save mid-game as long as you haven’t died yet, but overall certain upgrades you get are permanent and you start each new game with them).
    6 - Wand/weapon upgrades (3 levels of power) that you can get by spending gold (1 possible mechanic) or by finding items of some kind dropped by enemies (or in chests).

    Basically you start out weak, 3 hearts, tail attack, possession mechanic, and from there even though you always start at the beginning of the game when you die, you permanently gain stat bonuses, new weapons you can improve, new enemies to possess, and are better able to survive each dungeon level.

    Lastly, a more interesting/innovative setting/theme/world (e.g. more magic and lizard abilities, as you once talked about on the podcast, or weirder levels) would have helped make the game more distinctive. Story is fairly optional (Warframe and even Caves of Qud really aren’t about the story - Warframe is about progression and loot and action-oriented gameplay and Caves of Qud is about exploring this really weird world and advancing your character as far as you can before dying).

    I think Binding of Isaac succeeded, in part at least, due to the game’s aesthetic. Not because it was that awesome but because it made the game stand out from other top-down dungeon crawlers. The setting/theme was just something that helped it catch peoples’ attention. One of those “strange attractors” you’ve talked about in the past.

    Anyway, hope some of that helps explain my thoughts/perspective a bit better.

    Also, great community on the forums here and good/interesting discussion. Hard to find online these days. :)



  • @richtaur Not the OP, but I lost steam after level 2 in AWL2. I kept dying, didn’t really have enough motivation to keep trying, and gave up. The game wasn’t too hard, as I believe I would have been able to beat it at some point, but for some reason investing more time in improving my skill level just didn’t seem fun. Didn’t feel like I was making real progress in the game or getting much better, skill-wise, each playthrough. Progress felt sine-wavy (one good run, one terrible run, two more terrible runs, one good run, etc.).


  • Tiger Hat

    @richtaur said in Lostcast 189: Platform Percolator:

    What happens in the second level? Too hard?

    Battle fatigue.

    @geoffb has watched me play - My playstyle is far more reliant on health as a resource - see my love of Doom. The AWL games take inspiration from elsewhere, and treat health more as a failure tally (you could swap health points for lives with almost no change to the game).

    This is not a failure of the game - it just doesn’t fit my playstyle - however, if you could make it fit my playstyle and the current playstyle (and other playstyles as well) then you’re going to hit more market segments.


  • Tiger Hat

    Have you guys looked at Defold. It may be closer to the way you like to develop as it seems to be in between Unity and that Haxe stack you’ve been talking about lately. Probably more suited to simple 2D games than Unity. And it has a good IDE. But scripting is in Lua, which is a very good language by the way.


  • Tiger Hat

    I’ve looked at it - but King are the devil and it scares me attaching myself to them via a platform.


  • Patron

    Don’t have anything remarkable to say other than thanks for the episode. Had a long drive, and this provided a nice hour of content for it.


  • LDG

    @TokyoDan said in Lostcast 189: Platform Percolator:

    Have you guys looked at Defold.

    Ah yeah, I heard about Defold when it first launched, but it dropped off my radar for some reason. Gotta say, it checks a lot of boxes for sure. Open source Free, cross-platform, strong portfolio (Hammerwatch!), and Lua is one of those rare languages I actually feel is worth learning. It’s just everywhere, especially in gaming. Good stuff.


  • LDG

    @salmonmoose said in Lostcast 189: Platform Percolator:

    I’ve looked at it - but King are the devil and it scares me attaching myself to them via a platform.

    Hah, yeah I get that. Though I think I’m at the point where I’ve known enough good people, good engineers especially, who were at this or that company with a bad reputation. King acquired the engine team, so it came from an external source. Anyway I getcha but this seems on the lesser side of evil ;)

    @dannagle said in Lostcast 189: Platform Percolator:

    Don’t have anything remarkable to say other than thanks for the episode. Had a long drive, and this provided a nice hour of content for it.

    Cool that’s great, nothing like a podcast for a long drive. Sometimes when we’re casting it feels like hanging out with a bunch of friends. Cheers!


  • Tiger Hat

    @richtaur said in Lostcast 189: Platform Percolator:

    Hah, yeah I get that. Though I think I’m at the point where I’ve known enough good people, good engineers especially, who were at this or that company with a bad reputation. King acquired the engine team, so it came from an external source. Anyway I getcha but this seems on the lesser side of evil ;)

    Yeah - it’s more I don’t want to promote their evil, even if they’re doing something good - it’s not like Lumberjack, where there is some actual evil going on - or Rovio’s upstart project which just read like them trying to steal people’s ideas.


  • LDG

    Defold even (currently) requires you to upload your projects to their servers. How convenient! ;)

    Although … Activision now wholly owns King, so … mistrust King and you mistrust gasp Blizzard! And I think we have some folks in this community who will stand by their blizz


  • Tiger Hat

    Having to store your project on Defold’s servers instantly turned me off on Defold but there is a way around that.


Log in to reply