Forbidden Technique and Electra have undergone a massive journey. Starting back in late 2016, it’s close to 2 years since the beginning of the project and so much has been accomplished.
I’ve already talked about exchange variety and the ramifications of multi meaningful segments within monologue exchanges both from player or from ai. So that’s the big big update still that will take a long time to out do, if ever.
The newest change is the integration of Context modes and functional Condition settings per root chunk (conversation starts).
This allows for the game to property organize and expose options for the player to initiate conversations without dumping dozens or even hundreds or even thousands of options at once.
One feature I do want to investigate is still the brain dump/ exploratory conversation start within VR. This would allow for a metaphorical ocean of conversation starting point nodes that the player could see the meanings of (summaries). This would allow for such incredibly high levels of natural divergence across all players (given cannon game application) that it’s a frightening and chaotic thing to chase after/work on.
Of course decentralizing the order of chunk selection this broadly makes the impact of chunks almost completely unpredictable if it’s not already. That said predictability is not really a good goal, it’s just convenient for developing theories and trying to conceptualize emergent behavior of players.
There is also a relatability/ common experience continuum/timeline in which at a certain point, the choosing of certain topics causes the experience of talking to the same person so different that it could be hard for people to talk about characters with enough overlap to feel they are talking about the same person.
Now I’m getting caught in consequentialism for limiting player choice. BUT I HAVE TO! Damn, okay fine I’ll circle back once the new version is updated and working in the game again, and add an experimental mode, code name Brain Dump, or Thought Wanderer, or Free Thought. Anyways a mode that can be turned on in the testing version of the game that enables access to all chunks with some options of what conditions to ignore, maybe time or just simply uncapping the number of root chunks that can be viewed at once.
At the moment there are 3 categories of chunks: Seek (curious about the character), Reflect (expressing your own feelings), Sway (attempting to get the character to do something).
Each of these categories currently can have 3 chunks. So at anytime depending on how I deal with stacking and OnGoing conversation chunks there would be about 9 max conversation routes/path starts to choose from. Each of which branch indefinitely as created, and with internal complexity of course too.
Anyways unlocking from 9 chunks to realistically maybe like 40 – 100 chunks floating around in 360 spherical layout around the players head in VR does interest me, even if the lack of control and predictability scares me.
For the 99% of people who this is nonsense to, but are interested be sure to check out my youtube channel where I record podcasts that are much more comprehensible and conversational rather than note based. (BadPlanGames).
Anyways, I gotta let alone Free Thought mode, and focus on the actual purpose here which is sharing enthusiasm about the newest version of Electra which will be the operating version compatible with Forbidden Technique for the foreseeable future of the demo mode.
So two forces wrestle for influence over Electra, awesome and important features, and then a stoic call for minimalism because if this tool is capable of transforming thought and language formation it needs to be accessible to as many people as possible.
But simplifying and cutting is what iteration is for right?
So what’s new?
Context modes! Now there is a way to organize all your chunks temporarily during writing and editing based on dynamic properties. It dramatically reduces the amount of headspace (RAM) you have to dedicate to remembering what your plan and focus of a set of dialog is about. It also helps you fill out all potential dynamics the player may have with the ai character.
One context variable is BOND. This represents how much the character likes the player in a platonic or friendship-minded way. If you set that variable as an active context mode then you can layout all your chunks that have BOND as a condition and then you can see what conversation chunks pop up when the character hates the player (BOND = 3) neutral feelings (BOND = 5) or very positive (BOND = 7) and of course everything in between.
It allows organization of two completely exclusive relationship dynamics. Of course anything exclusionary has a negative impact on the relationship between a playthrough and percentage of content available at a time.
That said maybe you would be able to rollercoaster a bit and experience extreme highs and lows of a relationship. It’s really pretty impossible to say.
So layouts should greatly organize the accumulation of hundreds or maybe thousands of conversation starting chunks.
This blog post is extremely disjointed, but I still feel like posting it. Sorry for the mess, but you should be very excited about the new version of Electra and try it out for yourself!