next »

[Aloft]

MEGASTRUCTURE


Megastructure is a project which I have been pursuing in the time around my professional work, and other game development projects (most notably Hazelnut Bastille, which you can check out linked from my signature).

It can be described as an environmental exploration game with narrative elements; most of the story is told through the user's experience of the spaces, although there is also the mechanic of advancing the story through found documents. The nature of the structure they inhabit is kept purposefully vague, but is elaborated on as the story progresses. It is of vast but unknown dimensions and scope, consisting of many layers of accretional industrial systems layed upon one another in an organic fashion over a vast number of years.

Gameplay can be loosely described by the label of "walking simulation", although there do exist platforming elements.

A major design goal is the creation of a gameplay experience which is heavily psychoactive; the player is intended to experience the environment in a quickly induced trance-like state. This is acheived principally through music, which will consist of heavily dissonant experimental "drone" tracks, layered with binaural background tones. The environments themselves tend toward the monochromatic (principally through lighting), and have a quality of paradoxical detail and "sameness" simaltaneously, which is a kind of mild sensory deprivation. There is also a tendency for the player to experience long stretches of time in transit through spaces of great scale, rather than engaged in action-oriented tasks.

All of this presentation combines to bring our player into a state of mind which is in sync with the soul of the environment.

It is a somewhat unconventional approach to game design, which is normally focused on providing players with many alternating tasks to keep them engaged, but our intention is for the game to be experienced in a very different state than most games today. An important predecessor in this design approach is the indie title "NaissanceE", by Limasse5, which achieves a similar result through a heavily sensory-deprived monocrhomatic world devoid of (obvious) textures.

Other major art influences include the work of Tsutomu Nihei (Known for "Blame!"), H.R. Giger (inspired the artystle of "Alien"), Beksinki, and German Expressionism and Cyberpunk design in general.


Environments are constructed in a heavily modular-oriented way, allowing for rapid construction of the spaces primarily within engine. Development is taking place within the Unreal Engine 4.

Prototypes currently exist, although I am going through several stages of workflow studies before I will start on a production-simulating gameplay environment for public distribution.

I look forward to bringing you all more information as work progresses on this project, and it eventually shifts from side work to a full production schedule later on!

[Sustrato]

Hey Aloft, looks really cool! Do you think the whole game will have the same tone/color scheme? I understand you want that "sameness" but I'm wondering if there will be different levels or areas with their own standards.

Dunno if you've seen this, but there's another dev working on a very similar game, inspired by Blame!:

http://bac9.tumblr.com/

Both your projects look great, hopefully you'll push each other to even greater heights.

[sidbarnhoorn]

Sounds like an awesome idea. Looks good! :-)

[Aloft]

Hey Aloft, looks really cool! Do you think the whole game will have the same tone/color scheme? I understand you want that "sameness" but I'm wondering if there will be different levels or areas with their own standards.

Dunno if you've seen this, but there's another dev working on a very similar game, inspired by Blame!:

http://bac9.tumblr.com/

Both your projects look great, hopefully you'll push each other to even greater heights.

Yeah, I have seen that one. I hope they continue to update that one!

I am sort of acutely aware of any project with similar goals to this one, since this kind of design has only been tried a handful of other times, and I am always curious to see how others approach the design problems.

Mavros Sedeno, maker of NaissacneE, produced a work which was along similar lines, and also influenced by a number of Nihei works, titled "Inferences": http://www.mavrosarts.com/game.html



There is also Hiversaires, a point and click game which seemed to seek to recreate the style of Nihei as it appeared in graphic novel form: https://www.tigsource.com/2013/04/04/preview-hiversaires/



There was a prototype someone produced titled "Abandon":



There is a small handful of others out there too.



 

[Aloft]

Here is a look at a font which was produced for the project; the pictographs shown belong to a language which the player may slowly uncover over time to get a deeper understanding of the world they inhabit. It will probably be partially optional to progress, however observation will be rewarded.

The language has a simple grammar and syntax, and builds complex concepts by combining simple concept radicals, partially like the Hanzi characters of Mandarin do in some cases.

The aesthetic is created by combing strokes and forms from multiple languages, among them, Arabic, Kanji and hiragana characters, Hanzi Chinese characters, and Tibetan. There is also a substantial amount of 80's era cyberpunk influence in there as well.

The characters are intended to appear alien, but also intelligible, and designed around economy of strokes.

[Sustrato]

That's some very cool typography! Unfortunately it's difficult, nigh impossible, to tell apart the characters at the smallest resolution. Also the characters for the number five (I'm assuming) are messed up, they're a mix of three and six. I'll check out those other projects at one day, some of them look pretty tite!

[Pixelologist]

I am hungry for gifs of this. Next time don't start a devlog if you don't have gameplay footage already.

[Aloft]

I am hungry for gifs of this. Next time don't start a devlog if you don't have gameplay footage already.

Yeah this project is very much in pre-production atm. I could find some gifs to hold you over though!

There are a few hundred environmental assets, and a fullscale level built, but I am still in the process of streamlining the development pipeline, working out conventions in the modular design, and experimenting to find pitfalls in my process.

Right now, I am still working on study levels.

With medium-to-large projects like this, you sort of have to be very careful that you don't commit to a process which is going to be a ball-and-chain for you, and I very much noticed a few practices with that level that I think I should avoid from now on. For one, I think I will have far less unique assets, and more assets which are more universal. I also noticed weaknesses for how the assets joined based on how I measured the grid- I am going to be shifting from base 10 to base 2 measurements as a result (as is standard).

As opposed to my other project, this is one which I would say is going to happen on the timeline of 2-3 years, so sit tight I guess!

Here is a concrete example (pun not intended by welcome):



See the seam on the lower part of the pipes? Most of the pipe textures I have are seamless, and thus wouldn't have one of those; but a couple are not. Normally if you wanted to do something like that, you would just move the pipe so the seam was hidden from most logical viewing angles. A logical aid to that would be to put the pivot point in the middle of the pipe, to allow for easier rotation on the main axis for that purpose. But I didn't notice the pivot location I picked for the pipes was sub-optimal until I ran into the need to move it on that axis. Little things like that seem minor, but make too many errors like that and it adds up to a huge amount of time! I spent about 20 days assembling that level, and truth be told, If I had used only slightly better methods, it might have taken 10!

[Aloft]

That's some very cool typography! Unfortunately it's difficult, nigh impossible, to tell apart the characters at the smallest resolution. Also the characters for the number five (I'm assuming) are messed up, they're a mix of three and six. I'll check out those other projects at one day, some of them look pretty tite!

Hehe, the smallest ones are for low-res computer interface, and would of course look much larger on screen (only pixelated).

Yeah, looks like I have the wrong digit there for number 4, good catch lol!

[Oats]

this looks interesting, how much of the langauge have you worked out?

[Aloft]

The structure and syntax, and tentative values for the pictographs. The idea is to have as few basic concepts as possible.

Some of those radicals include "being / existence", "fluid", "And", "Or", "Not", "Harmony", "Part", etc.

The idea is that any concept which is not a compound idea, or a negation of another idea ("not" is getting a lot of mileage) is given its own radical, but every other concept is a "word" built of simpler radicals.

The language must be intelligible, and meanings must be discernable from context. A lot of the characters will have to be flat-out defined to the player at various places in the story, to give them some foundation, but there will probably be many which the player is expected to figure out on their own.

Not every player will be interested in exploring the world and its language that deeply, and so only a cursory understanding of it will be needed to complete the game, but for those who are interested, it will present several levels of challenge. I am considering adding a built-in tool to the game interface for allowing players to log their understood meanings for the various radicals, which will spit out those guesses when they encounter a place where the characters appear. 

The language has not been finalized yet, partially because I did it early to get a feel for the aesthetic of its appearance in the world. It is one of the things that is high priority, but lower priority than a few tasks which are still being worked through, such as building a larger library of module assets, and reworking some of the existing ones to reflect principles learned from earlier level designs in this project.
 

[Oats]

interesting, how long is an average playthrough going to take then?

[Aloft]

Shooting for 15-20 hours for a casual completion, but players who take the time to read all documents, and take notes on instances of the pictographs in the environment in order to work them out may take considerably longer.

[acatalept]

Just saw Chris Priestman's writeup on this in Killscreen -- I'm a sucker for monochromatic megastructures ;)

Looking outstanding, can't wait to see this in motion (and in higher resolution).  Keep it up!

[Aloft]

Just saw Chris Priestman's writeup on this in Killscreen -- I'm a sucker for monochromatic megastructures

Looking outstanding, can't wait to see this in motion (and in higher resolution).  Keep it up!

Yeah I was glad to see it too! Work is a bit slow on this one, because we are really pushing Hazelnut Bastille for the moment, but I really hope to kick this into high gear again afterward! It is probably the project that best addresses my personal skillset and interest in games and art.

Sing Me To Sleep looks incredible too btw... I am a sucker for expressionism and existentialism in games as well. NaissanceE was probably the first game I am aware of to really do it right in a big way, and I really want to see ideas like this explored in the future. 

[Aloft]

[Aloft]

Ok, Added a few things, and some ambient occlusion to dissolve the interior corners and edges; this is probably a pretty good rendering profile to use for the first public demo stage:

[acatalept]

I love the amount of texture, grit, decay evident especially in that last image... really captures a sense abandonment, disuse, as of ancient ruins of the remote future, as is done so well in Nihei's megastructures.

Will this be all static lighting?  Not sure what your target platforms are, but I've found even the commonly dismissed Light Propagation Volume GI lighting to work quite well in UE4 for certain situations (though I haven't tested it in recent builds), as long as you're aware of its weaknesses and work around them:





E.g., a huge moving fan blade or other moving geometry slowly occluding major light sources, washing the environments with subtle bounce light and shadows - that dynamic light/shadow interplay, even when subtle, goes a long way to playing tricks with the player's perceptions, giving their imagination more fodder to work with...

Also looking forward to seeing some of these environments in motion, set to some fitting "heavily dissonant experimental 'drone' tracks, layered with binaural background tones" as you mentioned in your first post   Wish I could just "thumbs up" your progress without polluting your thread, but I'll be watching this space.  Keep it up!

[Aloft]

I love the amount of texture, grit, decay evident especially in that last image... really captures a sense abandonment, disuse, as of ancient ruins of the remote future, as is done so well in Nihei's megastructures.

Will this be all static lighting?  Not sure what your target platforms are, but I've found even the commonly dismissed Light Propagation Volume GI lighting to work quite well in UE4 for certain situations (though I haven't tested it in recent builds), as long as you're aware of its weaknesses and work around them:





E.g., a huge moving fan blade or other moving geometry slowly occluding major light sources, washing the environments with subtle bounce light and shadows - that dynamic light/shadow interplay, even when subtle, goes a long way to playing tricks with the player's perceptions, giving their imagination more fodder to work with...

Also looking forward to seeing some of these environments in motion, set to some fitting "heavily dissonant experimental 'drone' tracks, layered with binaural background tones" as you mentioned in your first post   Wish I could just "thumbs up" your progress without polluting your thread, but I'll be watching this space.  Keep it up!

Definitely not polluting this thread, hehe. I pretty much live for stuff like this. Pretty psyched to see where "Memories of a Broken Dimension" and similar work go. Your work reminds me a bit of NaissanceE in that the textures are there, but you don't notice them until you catch the right angle to see a rough normal map. The emphasis with yours is on form and scale.

I plan to use a mixture of both static and dynamic lighting in UE4. Dynamic is costly, so has to be used very sparingly, but for stuff in motion, or something like a broken, flickering light on a script, it works well.

Music wise... I think Elaine Radigue is a good approximation of what a lot of it may be like:






https://www.youtube.com/watch?v=6YHTUxaUMxs


The idea there isn't about conveying simple horror, but something more profound... enormity. Enormity isn't horrific by nature (unless we are talking about Joseph Conrad's Horror at the base of existence), but it tends to remind us how powerful and vast the universe as a whole is, and tends to bear down upon us how little we factor into its plans and workings. We happen to be here merely to bear witness to it for a moment...

You can find this thread saturating Nihei's work liberally as well... I think he is filtering the Japanese flavor of Buddhism through his art.

You can sort of see that spirit coming through in Modern Japanese music (as opposed to Japanese Romanticism that is built on western ideas):

https://www.youtube.com/watch?v=euHhTE0OS18