Games You Played Today ver.1.22474487139...

King of Posters
1905

Yeah, the comparison to arcade games is a pretty interesting one, but that’s more where I’d slot stuff like Spelunky or Slay the Spire (or recently I have been very into Peglin, which is in a highly content-starved early access but has good bones). The formula we like to call roguelike (as if any of these games are literally anything like rogue!) is imo a good one because especially in a time of huge sprawling repetitive games there’s nothing quite like a game that’s just actually good to play and actually tuned to be played repeatedly. Like trying to 1cc an arcade cabinet, it’s about getting fun out of a thing you already have rather than constant novelty or grindy compulsion mechanics, and I can get behind that.

The roguelite thing… I don’t know, do players like it actually? Rogue Legacy 2 is an awfully good game but has neither the “ok one more game” fun of Spelunky or the “I am going to understand this machine if it kills me” brain engagement of Spire. My first few runs were crushingly impossible and I got totally owned on like the third screen. By the time I felt like I knew how to play the game it was also just an easier game. What if it was just tuned well to begin with?

At the end of the day, I mean the whole thing works ok, and it’s very well crafted, but it’s not clear that the big selling point helps out with much more than marketing.

The introduction in RL2 of unlocks to skip dungeon sections during a run kind of demonstrates a lot of my issues with this game in comparison to the genre it helped build: if you need to let players skip something for the game to not be boring, was it really such a good idea to begin with?

6 Likes
1906

Did you know there are 755 active MUDs out there? Not that you can necessarily just get a full player account on all of them, but anyway I thought I should check some out. ; ) Valhalla MUD was the one that kind of clicked with me and I already rattled on about that one a little while back; here are the others I tried so far and bounced off of–and I should note I’m mostly trying the virtually dead ones because I’m antisocial or at any rate just don’t feel like being obligated to RP while I’m bumbling around trying to check stuff out : D:

(Using mudstats.com and MUD Database Search at Top Mud Sites to look for largely abandoned, free, large-scale fantasy MUDs without forced PVP, that don’t require an email address or funky hoops to jump through to join, and mostly with their own web client because those are kind of fun to explore too)

The Sea of Storms

“Wheel of Time”-themed MUD with heavy roleplay emphasis…apparently; nobody is online these days and so the requirement to contact a ranked player to promote you out of the newbie area is a steep one; I suppose you could go try on their Discord. The newbie area is rocky and wind-lashed but otherwise not that interesting; combat with a minor monster seems to start automatically for some reason, which is also not encouraging.

For character creation there’s a big emphasis on creating a “channeler,” whatever that is exactly; you can choose standard or “advanced” creation of such; “advanced” has you picking a lot of things and allocating points; it’s rather involved and I’m not sure what it meant but sounded fairly wizard-y.

Another sorta off-putting thing about the newbie area was you find a spot where you see different groups of the world’s peoples or something milling about in the distance, and you can move in most of the eight (!) directions: this takes you to spots to observe the racial group in some sort of natural habitat, an excuse to describe them to you. Felt kludgy.

MUME

Middle-Earth-themed MUD, fairly popular. After character creation I was plunged directly into a scripted encounter with Gandalf, which is sort of cool I guess but also a little uh dramatic, and anyway we ended up running from some wargs and I was being auto-towed behind Gandalf or something and next thing I knew I was dead. So, not a great start. And then I think I was sort of lot in the starting inn thing and I think it was this game where there’s some notion of an in-game-currency “rent” paid while you’re logged out, basically a way to let them delete old unused accounts I guess. So, probably won’t be back.

MUD2

Second MUD made by Richard Bartle, after his genre-defining “MUD” (now known as “MUD1”). There is a lot of text. A lot. It’s a bit much to take in.

On the current server I tried, you can’t interact with anything until you manually contact the admin to get a real character created, or something. I didn’t get that far.

LegendMUD

Long ago, UO and SWG guy Raph Koster worked on this MUD featuring many “exotic” settings around a sort of age-of-steam-themed world. Small current user base. You can pick your starting city, so I picked something in what I imagined was an “Arabian Nights” setting, and it was, but if there was something to guide me around the starting city, I totally missed it; I stroked a kitten and there was a puppy in the street too, but neither seemed inclined to show me interesting places to go. So I left.

Genesis

Fantasy-themed MUD. Used to play this back in the day in my buddy’s college computer lab; he was power-gaming it while I was trying to fight ants on an anthill–that was as far as I got. These days, Genesis still seems to be reasonably popular, and there’s a whole tutorial area that is very explicit about being a tutorial training park or something, which kind of weirded me out.

17 Likes
1907

Personally, I’d just say that it demonstrates that Rogue Legacy isn’t very good and that it doesn’t understand how to use its structure.

3 Likes
1908

My opinion on roguelites as a genre is that there’s nothing wrong with the basic concept, it’s just that most of them are designed for Maximum Content/Playtime or are just “these are how games are supposed to be” without really thinking about the why or how it affects the player. And that they would, almost across the board, be improved by being shorter, easier, and sloppier (more variance! why base so much of your shit on “randomness” and then have the actual random effects be so dull?!?).

20 Likes
1909

when will roguelites get their universal paperclips?

4 Likes
1910

Honestly the weirdest thing for me with Rogue Legacy is that the name and concept are stolen from Legacy board games, which do WAY more interesting things to the gameplay over time to keep a (usually pretty basic) board game fresh.

RL2 is, in comparison to the first one, very maximalist, which helps it land a lot better since, hey, if you want stuff, there’s lots of stuff in there. But it kind of also shows the paucity of imagination in most indie games designed for sales. Legacy board games became infamous for basically eating themselves—becoming basically unplayable at the end under the weight of rules changes, physically altering the board and pieces, etc.—whereas rogue legacy is just like “what are all the video game systems” and then it just tapes them together. If anything it feels more static than anything. Numbies go up but not much else changes.

It brings to mind what a lot of what I personally call the “steam indie” genre accomplishes in practice. The games can be better or worse to play, but they’re never surprising and rarely trying to do much more than copy each other

8 Likes
1911

I think most roguelites would benefit greatly from stepping back and analyzing WHERE randomization would be the most impactful. Everyone is very focused on randomizing levels, and I think that’s one of the worst effort-to-result ratios. It’s really hard to make randomized levels and in doing so you throw out all the careful design decisions you can make when you’re hand building them. And after all that effort, the “randomized” levels feel samey within 15 minutes.

Character variety is where a lot of roguelites should be spending more of their development time. The DOTAs and Fortnites of the world get a ton of millage out of a single map because of the character/gear and opponent variety.

13 Likes
1912

If I ever make a roguelite it’s going to end up more like the “randomizers” they use in speedrunning communities.

Making routing decisions on the fly based on what gear you currently have access to is very interesting and already has more play variety than most roguelites. “Guess I need to beat this dungeon without a sword” or “this enemy is way way harder without arrows” is spiky and interesting and good.

15 Likes
1913

As far as I can tell, nobody’s really made any advances on roguelike level design since Spelunky’s breakthrough, right? The insight that hand-made chunks can be overlaid with random elements and still provide enough systemic and experiential variety. But the pattern has become fairly obvious and the limits designed into the system are stifling more expressive outcomes.

No, wait – I’m forgetting about Unexplored’s cyclic generation, which worked towards a more coherent meta-room structure. Unfortunately that game plays more like an academic demonstration project.

This is something that’s gotten a lot more settled and interesting in the past few years, the communication to the player about hidden future information and the limited choices they have to influence it, rather than just react to it. Binding of Isaac and its keys and bombs was always better than its contemporaries but I think it took until Slay the Spire until the lesson and appropriate balance really sunk in.

And it reminds me again of the utter failure of the graphic design, Apple-adjacent games movement of the early 2010s that utterly failed in the minimal roguelikes it produced. Below, Tumbleseed, Overland all try to strip back and be ‘tasteful’ with it in ways that close down the possibility space. It’s just a terrible mix with their hyper-controlled style.

10 Likes
1915

Yep, basically.

Unexplored level generation is really cool and I wish it was connected to a better game haha. It also seems like it was a TON of work to implement.

5 Likes
1916

just to point out that Rogue Legacy unlocks the architect very early on, who lets you turn off level randomisation and turns the game into a tedious platformer while you hope the next character will have useful traits

8 Likes
1917

Is it a level they hand-made or is it just a seeded randomized level?

1918

diablo 2 is full of handmade chunks broken up by procgen, roguelites didn’t originate it. hell, i think Crawl even does that

imo the subgenre has introduced little that isn’t present in its influences beyond a tighter gameplay loop than a typical action rpg (let’s be honest, the action ones like hades take after diablo, not nethack) and nearly removing character customization to make permadeath less painful since you’re not watching hours of careful build planning and gear acquisition go poof

13 Likes
1919

Yeah, most traditional roguelikes mix hand-generation with algorithmic generation (I think one of few exceptions is Rogue itself). I didn’t mean to imply that was unique to roguelites.

I do think Spelunky’s flavor is distinct in a couple ways – it’s a harder problem to make something traversable and interesting in a platformer context vs. a top-down one and I think Derek’s hand-made designs are a lot “tighter” and well-considered than most of the games that come before it.

4 Likes
1920

Yeah, I mean Spelunky just uses Wang tiles, which have been around since the 60s and used in level generators for pretty much the whole history of games.

There absolutely have been advances in level design—there’s usually a few talks about it every year at Roguelike Celebration and the Short/Adams-edited PCG books have a couple good examples of things I’m pretty sure I haven’t even seen in practice, but yeah, the problem is they’re often just attached to games that aren’t good.

I know I complain about this a lot but a lot of it I think comes back to the fundamental tendency in game design to see games as collections of technical features, where more features is better and making the game good is about iterating on the features until all the features are the best features they can be.

In musical analysis, especially popularly, there’s this huge obsession with chord numbering and progressions (i.e., treating all music as fundamentally inversions of numbered chords and totally bypassing A. a century of art music and B. big unanswered questions like are all inversions always even audibly related to the root and how to resolve progressional ambiguities in things like suspensions, etc.). And the technical feature obsession in games, which tends to drive the Steam Indie game most of all, kind of feels like this to me. Without other elements of musicality like rhythm, voicing, orchestration, etc. chord progressions often mean very little to listeners (how much of “history’s greatest music” is really built on I-V-I?). But it appeals to a certain kind of theorist because tonal music is basically “solved” at this point. Any composition student can write a fair shot at Bach vocal counterpoint and it isn’t an outrageous amount of work for most even middling composers to put together at least a pastiche of even Debussy or Mendelssohn. Similarly, “roguelite mechanics” are in tons of games because they’re pretty easy to copy and from there it’s just a matter of balancing.

But what most of these titles miss (while they’re trying to find the exact combination of two existing games that makes a winner) tends to be, you know, all the other things that make up a game—like imagination (what elements that are not even actually present in the game make it compelling?), and experiential quality (what does it feel like to actually play it?) and craft (does it look/sound/work good?).

And without this stuff even cool mechanical ideas are just not going to shine. They’re not going to feel good because there’s nothing to hang them on. Like the romantic pastiche, they’re kind of just exercises.

7 Likes
1921

It’s contemporaneous with Spelunky, but Brogue has a much more dynamic approach involving generation rules with statistical parameter ranges, with a broad-strokes pass and a detail pass. It leads to a naturalistic cave setting with large variations in scale and organic connections between types of space. This doesn’t come at the expense of game design, on the contrary it makes the lighting, fire and gas systems play out more interestingly too.

Even the forts and treasure room puzzles (the most handmade-chunkish parts) are generated this way only with more specific rules and narrower ranges, so they give the impression that the builders adapted their design to the cave they were building in.

10 Likes
1922

We have to code-switch, right? Move between romantic, subjective language when describing themes and atmosphere and purpose and analytic language when deconstructing and planning.

And a lot of teams get lost because games can be inspired by both ends. I think it’s perfectly valid for a game’s driving purpose to be the exploration of specific mechanics, and search for the aesthetics that illustrate that action. Just as well, I think games can treat the mechanics to act in service of the aesthetics. Personally, I’m very aesthetically-driven but most of my time in teams is proposing systems to solve for earlier aesthetic choices.

In larger productions, I find this hard to talk about because it’s owned by so few people, because it’s known to be more dangerous to expose to committee. But the problem is that the subtleties of mood and values need to be shared and owned by the team so they can make the correct micro-decisions. One of the reasons I think small, human-sized teams are so much more productive is that they can all share and modify the aesthetic vision.

2 Likes
1923

Yeah, I feel Brogue reaches an ideal synthesis in part because it was a solo dev able to connect a difficult-to-explain technical system repeatedly back to aesthetic goals over the course of development. It would be challenging to iterate or add modules to a generation system like that as a team. (Brogue’s solo dev is uncommonly terrible at marketing though — the official website’s homepage is a wall of text with one tiny screenshot on it! — which is why it remains obscure despite in my opinion being the traditional roguelike GOAT.)

Also Brogue is in the traditional turn-based, top-down style, which I agree with Gate88 means that the procedural generation design problem is much easier. Noita is in my headcanon a spiritual successor to Brogue as a platformer roguelite (not that I ever heard Noita devs mention Brogue, but the gas/fire emphasis is obviously similar), and I respect Noita’s devs but also don’t particularly enjoy Noita: to me that proves that importing Brogue’s design philosophy to a platformer roguelite turns out not to be a solvable problem.

8 Likes
1924

turning off randomisation? maybe “stops rerolling the map when you restart” is clearer?

1 Like
1925

Jump on Catacomb Kids if you’d like to see more traditional roguelike interaction complexity married to tighter, Spelunky-esque platforming. Highly recommended.

In other Games I’ve Played Today, I can’t tell what Loot River is trying to accomplish. The top-down perspective and muddy implementation of Dead Cells’ polygons-to-spritesheets technique make the Souls-esque combat muddy, slow, and hard to read. The world is vague enough that it just comes off as post-gasp Souls-like, and the high-concept tile shifting just…is? There’s some amount of aggro-management play it allows but I don’t see it expanding into interesting combat scenarios, and it really limits the level design, which also has some notable poor choices, like split paths without one-way locks, which encourages far too much backtracking.

5 Likes