Schrödinger's Levels - Legacy Devlog #27 

What happens in your example if the player chooses the left path, then backtracks and chooses the right path? From what you're saying, both would then load the up to down version and stick with it, creating the undesireable behavior you mentionned.

Typewriter and Cartographer are just so incredible.

Honestly I LOVE coming across a door that "doesn't open from this side." There's this rush of wonder, like ohhh I'm going to be coming back through here later but I won't know when. And then when you open a door and realize, oh, I remember this! It's magic. I feel like if you smooth that over, you wind up with something like the dungeons in Skyrim, where they all start feeling the same. You come to expect that easy flow where you clear everything out and, oh of course there's a path that conveniently loops back to the entrance. You also run into a problem where you can't make one puzzle more advanced than another because you don't know what order your player will encounter them in. So instead of having two puzzles in sequence that get more interesting and build upon knowledge gained from the previous ones, you have two basic puzzles. I think there's an impulse to smooth out a game's flow for a better experience, but the fun in games comes from a designed set of challenges. If you don't have just the right amount of grit in there, you risk losing form and challenge and fun. Now, if you used this tech to make sure the player always went the *wrong* way first, that would be absolutely hilarious.

I am once again amazed at how well-designed your editor tools are! How did you learn to make Editor UIs like that?

What an elegant solution to a problem that's become so routine we hardly notice it anymore! I hope to see other games implement similar measures! Also, I imagine this gets rather convoluted when rooms have 2+ entrances/exits

I've been here since the start of these dev logs, I've never commented but now I just want to take some time to apricate this mans genius at game design. He creates his own tools that look like teams are behind them, and he come up with solutions that have never been used before. Im truly convinced this man is a once a generation genius. As someone who want to be a game designer one day, It would be an honor just to meet you.

Wow, your UI for Cartgrapher is really nice. This is actually strange to see so how quality animations in the devlog

I subbed after Next-Gen Pixel Art, and it is crazy to see how fast time flies by. Your narrative and videos are of excellent quality, and I've learned much from your problem-solving skills. Thank you for the devlogs!

Lowkey best devlogs I've ever seen.

This somewhat reminds me of wave function collapse for level creation. Also for left and right entrances do you need a different variant or could you just mirror the level.

I feel like thia system would be really interesting to use in a long path of single rooms in order to reach a dead end, and when you get what you're supposed to do at the last room, the rooms going back change slightly as to make new puzzles or perhaps some kind of escape sequence, if you turned on the area's power system, for example. Great stuff.

Very cool idea, but it falls apart for players that have a similar playstyle to me: peek into every optional way and then decide where to go. In that case it would still feel linear but now you allways have to backtrack to you first peek, which might be very annoying

I feel like this very quickly falls apart if the player wants to make _informed_ decisions on where they go, they check both paths, thus setting the direction of the room, and then chooses one, missing out on the other one entirely because it's now completely irrelevant. So we're back to the original problem of irrelevant rooms.

The idea of changing a part depending on any fact is also a good one... you can open doors in other rooms, even though it's not intuitive. Like restoring power from several places to a central one, and other story wide puzzle arcs...

My first thought for the scenario at 1:23 would be just making one level for going down and one level for going up. And the player always gets the level for going down first (but mirrored). The advantage of this would be that you don't have to create 4 levels, but just 2. Though if the scenery is flowing into the other levels that wouldn't be a possibility.

i love your videos and as an aspiring game dev i reallyenjoy the fact you explain the actual concepts and how you did it

It's a neat idea but I'll have to see it in execution. From what I understand at the moment it sounds like making the player think they're making a choice without the choice actually mattering in the end by making both choices have the same outcome.

Once again amazed at the unique solution you find for the problem haha

pretty neat! Though I figure the inconsistencies of level design will make walkthroughs a bit more challenging to write. 'here's how you solve this room, unless you came in through this door, in which case, do this instead...'

What if a player enters a room and decides directly to check the other possibility, this will cause both side to be directed the same way. I think you may add a "quantum entanglement" (to stay with your analogy) to define a definite loop on your map: knowing the state of one room force an other one to be definite (entanglement). Otherwise, a player finding a puzzle difficult in a room may track back and cause two directed rooms to be oriented the same way, forcing the player to back track all the map to switch on the door that she/he didn't open at first. Otherwise it's a good idea. I personnaly prefer a well designed linear game than a "OK" open-world game but I'm sure you'll find out how to make this new level of abstraction interesting. Keep up the good work 💪

I'll be honest, if I were playing a game with two paths to get somewhere, there's no chance I'd leave the other path unexplored. I feel this is a bit of an over engineered solution to something that isn't even really a huge issue in the first place, cause players are either gonna meticulously explore every place in the map, or not care if they miss small areas like that. Not to mention you're greatly increasing the workload with the variation system, having to create multiple versions of every area, which you'll have to test and keep changes synchronized between if/when you find an error in one. Personally I'd just give the player a reason to want to check out everything. Maybe put money or items in areas. That's just my two cents. I think the system is really cool and could be put to good use, but for what you're describing, it seems like a complex solution to a minor issue, and introduces its own set of issues.

Cool concept. I enjoy these devlogs as they present interesting problems I never thought about before and also provide naturally fitting solutions.

I only just discovered ur Chanel and u do so much amazing stuff, thank you

I'm a software engineer making web applications and your Typewriter system is super inspiring for in-app onboarding. Makes so much sense to me to treat software as a game. UI screens can be accessed from different paths and might show different dialog etc.

Great concept! I develop levels procedurally with spanning tree to decide the routes deliberately to avoid the situation but you've met, solved, named and shared it!

This is a really interesting system. It seems like it would be pretty time-consuming creating multiple variants of many of the rooms though. I just want to remind you that it's okay to have players go the wrong way. Metroidvanias are built off of exploration and going the wrong way fairly often. It is a somewhat different situation with this game though, since you have to do the puzzles from the right direction to traverse the room, and it might not immediately be apparent that you're starting the puzzle from the wrong side. Anyways, very cool system and editor window👍

Absolutely can't wait to play this I'm glad I found your channel.

I love the little animations inbetween clips!

This seems like it could create an extremely interesting routing challenge for speedrunning, especially if some of these directed puzzles end up not being entirely symmetrical, e.g., if a room is faster to solve in one variant (maybe the entrance places you closer to key points) but is faster to *reach* in another variant. Very cool concept from many different angles!

Your channel just got recommended to me and I gotta say, this is really cool! Thank you RU-vid algorithm!

this is a very cool design idea, thanks for sharing your creative process!

in the example of 2 puzzles going down towards a goal room, you could make the secondary "now useless" puzzles first door open and reveal a new previously hidden door with a bonus prize

I miss the light rays that show you what anchor you're connected to, I think they made things much clearer. The colour indicators for certain things also seem too close to each other in shade and hue so people who aren't good at seeing colour variation would struggle to tell, for example, which trigger is attached to which thing because they are all shades of orange that aren't too different. Other than that I just binged the whole devlog series and this game is looking epic. You're going above and beyond in some really interesting ways, I can't wait to see the final product.

This is an amazing level design idea this could easily make the player try different things and increases the replayablility of the game.

The level of polish in your videos and tools is just incredible. They're by far the best devlogs I've ever come across, keep up the amazing work

dude you are doing AMAZING! Keep up the good work I love to see such a smart person making a great game, my mind explodes with all the stuff you are able to do cant wait for the game to come out. :)

A full game with this design philosophy could be ridiculously interesting for speedrun routing

I enjoy learning game design a lot, makes me appreciate the games I play a lot more

Even ignoring the amazing design decisions and the amount of care that you're putting into this game, the utilities that you're using to create it are simply fantastic. If I didn't see a million other people in these comments saying that it was a custom utility, I'd think that Cartographer was some ridiculously expensive Unity add-on made by a professional company.

This is actually such a brilliant solution to this problem, I don't think I've ever seen anything quite like this, though I guess it is also designed so the player doesn't notice :P

Amazing idea. I love how the world essentially designs and modifies itself around the player's choice of direction, it creates a unique experience for any replays

3:15 What if you walk into the room to the right, but decide you want to go the other way instead. You leave the room, enter the room on the left instead and finish the puzzle. At this point you would have to change the right room layout, destroying the continuity from last time the player entered the room, to avoid backtracking. It would not be an invisible system anymore.

Wow, I really like this Cartographer & Typewriter system you have in place, I'm actively working on a game engine, and this is really cool, this is very inspiring for what I might require in my future games.

I’ve just finished your 26 devlogs and man, what a journey you have done. I really recognize myself in your 1st video. I have been thinking about starting a RU-vid channel to devlogs for a while now and the discovery of your channel might be the trigger. Keep your amazing and inspiring work, and I hope I’ll have the courage to take the first step

Every new devlog I see makes me more excited to see this game completed! I love this little mechanic described here!

My mind is blown by the editor tools alone. Can't wait for the game.

This devlog series continues to be so god damn cool. Excited about the game but the tools you make along the way are so interesting

Fantastic solution to the problem!

As awesome as this is it seems like a scoping nightmare, especially for puzzles that go up or down or into multiple directions

Everytime I see another devlog from you I get inspiration to keep learing unity! Started from pretty much zero but have been learning the basics gradually over the last few weeks. Love the videos and can't wait to see the finish product, keep it up!

Using wave function collapse for levels is a genius idea!

This is really neat, thanks for sharing!

I adore your content. It keeps me hooked through sheer ingenuity. Just brilliant!!

Amazing demonstration once again, so curious to see how these play out in your game :)

You never stop blowing my mind!! this is genius and I love having this devlog because when I eventually play the game I'll appreciate all the effort that went into making it the best experience possible. Also, you make such high quality tools I'm sure they'll stick around even after astortion is released. Your next game will be standing in the shoulders of a giant with all these cool systems

Absolutely incredible

Super awesome! Love including ideas like this to keep things fresh.

Perfect name for the concept, I was wondering where you were going with the name but now it clicked hahah

i like this concept, i’ve seen it around sometimes, and i’ve always thought its an amazing idea. Good vid.

not only is this great for casual play and exploration, i can see this causing making routes for speedrunning to be way more fun as well more devs should take this approach

I spent the the day binging the entire Astortion devlog playlist and I am simply amazed by your work. Needless to say, I've subscribed and turned on notifs. Love the humour, and the educational value is simply icing on the cake. Can't wait to buy the game! :D

This is a truly amazing solution to this problem. Hats off to you!

Keep doing your work! You are one of the best gamedev youtubers! You are doing extremely good stuff, taking unusual topics and you have very good video quality! Good luck!

i love how your custom editors look in unity. I would love a video explaining a few advanced tricks you can do with custom editors

This game looks promising, I totally plan on getting it when you release it, keep up the good work

Clean video, clean game structure, clear code, clear UI, is there anything this man can't do??? Massive props, you're amazing

Your tool-programming abilities are god-tier.

Really nice video

Brilliant! However I feel entangled levels would be a better naming than Schrödinger levels.

You can make the example room at the very beginning bi directed by letting the player bring a green cube from the previous room.

Genius - you always continue to inspire me with your cutting-edge game design ideas

Man, I always love accidentally stumbling across these devlogs, helps teach me stuff about making my own games, even if I probably will never use these concepts.

this is gold for a game designer

This is brilliant!

Babe wake up, new aarthificial vid just dropped

This is so insane and genius! :)

Here’s a free level idea for you: You have to navigate around some abstract geometry by tethering to a blue anchor behind the walkable surfaces, all while dodging projectiles. If changeable difficulties ever get implemented, higher difficulties could add mass to the anchor or increase the speed of the projectiles.

I remember that character from a tutorial that you made. Nice job using it.

i’ve seen TWO episodes and i’m hooked 😅

Nice solution! I first thought you were going to use some form of entanglement, where entering one room sets that rooms and limits/sets the other rooms that lead to the same target. Also, creating multiple versions of the same room is probably best if you only have 2-3 entrances. I also considered making a level builder that flips the whole level on first observation if necessary, but that again is the old computation vs memory trade-off.

This is extremely clever idea, haven’t seen anyone do that before, its solves so many issues in Metroidvania games

What a very cool solution.

This is very similar to how I plan to make the levels in my game. Very well explained. :)

THIS GUY IS A GENIUS

I knew I was right about this being a Fucking Useful Map Tool™. Well done, yet again!

I like this idea of superpositioned levels.

"It's worth noting that there's no randomness here" oh but there is! The PLAYER is the random variable! :D

reminds me of fast traveling in open worlds bc you usually have to unlock them and they may or may not be near a mission/something it gives you a forced exploration or you unlocked one already while exploring/doing something else

For a puzzle game, a single correct path is not a problem if it is not obvious. In this case, the real mental work for the player would be to narrow down his options; from a world of seemingly infinite misleading possibilities to the understanding of the proper sequence. Another way to make a single path fascinating is by realizing the meaning of connections between rooms. For example, two rooms that don't make sense separately can suddenly make sense if you have to drop an object from the top room into the bottom room; or break the ceiling of the bottom room to allow an air current to reach the top room.

this is the best game dev concept I heard in a long time

This is some high IQ thinking. I really loved how the simplied version (the colorful one) made the explanation much easier to understand. Everything was really straight forward and I enjoyed the video. Great one!!

pure genius, really.

Oh my god. How do you manage to make such *amazing* content? This channel is a gold mine of great stuff. Keep up the good work!

Man, this looks really nice! Your tools are amazing, it really inspires me to make and improve my own.

You are an absolute genious

yay, astortion devlog again!!

Woah! What a cool idea!

I saw the title and thought about the Wave Function Collapse algorithm, it's an algorithm you could definitely use here too, and it will solve some of the edge-cases (but not all of them)

I have added Astortion to my steam wishlist

Love the system and application, but I'm mildly concerned you're losing a bit of something in the process. Aside from the obvious issue of the player backtracking despite your system, which you can fix by linking the decision-making of the room generation, one of the core reasons people enjoy metroid-vania style play is the feeling of exploration, of hitting a dead-end, recognizing it and then coming back later with a solution. I think constantly generating the correct path for the player basically puts them on rails and cuts out the experience of exploration. I'd advocate for a middle-ground where certain puzzle rooms and progression points have a fixed directionality regardless, so the feeling of exploration remains, but you can remove a bit of the typical inconvenience by reducing how frequently the player needs to backtrack.