Has anyone developed using adaptive feedback design?

I know that modern casual games have been using A / B testing to ensure they get the maximum retention and income from their players, but has anyone used a similar system when building their game.

In theory a game can take many directions from inception to delivery and normally it's up to the development team to find the sweet spots along that route.

But what if you developed your game as an open Alpha and then A/B tested the new features recording data and feedback from users to steer the next phase of the project?

In theory the industry is mature enough now to have a set of available mechanics that can be used as A/B options. Unity with it's asset store could even be an ideal candidate for this type of development.

SideNote: Although it would be nice of the asset store had a free to try option for all it's items.



Now add in those funny skull caps that read the players brain waves and you could have a game development system that would allow developers to evolve their game towards it's optimal entertainment level.

Or we can just use lots of prototyping.

We would have adaptive design discussions where someone puts forward a game idea and then we could discuss design options on a thread, what do you think of the idea of adaptive feedback design?

 

The problem with this is that it's only good at detecting somewhat broad changes in the electrical waves, like emotions.
Unless it made major leaps since I last researched it, it was not capable of correctly understanding what the person wanted.

The orders for movements or actions, that are specifically meant not to be transmited at our muscles (you don't want to have to run in order to give the order to run), were not significantly different for themto be correctly interpreted.

It's been a few years since I last went into researches about our understanding and monitoring of brain waves though, so it's possible it changed enough to be able to do it.

 

I just don't know when they are going to start using iris detection and you double-blink to select things.

 

This sort of borders on user generated content, allowing the audience to define where the game goes and what it includes... if it's a matter of you deciding for them what the possible paths are, and testing them out, then I guess you're putting a barrier between you and them to retain control... but you could open it up in such a way as to let the audience have more of a say.

I think this will be happening more and more in future... game development as a service... not just 'heres the game I made because I'm awesome' but more 'what do you want me to put in the game?'..

 

What makes you say this?

You also mention the Asset Store, but it's worth noting that stuff in there for game mechanics isn't anywhere even remotely near the point where you can trivially swap one asset out for another for the purposes of testing.

(As an aside, though, I am currently developing a simple puzzle game where you can do exactly that with the rules at each of its phases, and with the properties of each of those rules.)

 

I kind of don't think it's a good idea. I mean, making games is like writing a book or drawing a picture. The only person you should be worried about pleasing is yourself. Now, I'm not saying to ignore people's feedback, but when you create something (like a game, book, picture, etc.) the main question is this: Are you happy with it? Because there will always be people who agree with you and like the game, even if it's a small amount of players. My main point is if you're not making the games that you want to create/play, then game development loses any point and ceases to be fun.

 

That's why I think Unity should develop standard game templates, think what it would be like if you build your game in a template as an FPS then while your prototyping you could port it into a turn based shooter to see how that would play.

Well there is writing the game you want, but you can't get away from the fact that to make it a career you need to make a game that people want to play and are willing to buy. So some features or aspects of your game you might be better off A/B testing, ideally it would not be the core but supporting or meta game features could benefit from this.

 

I'm having trouble imagining anything practical there that doesn't reduce game design to "fill in the blanks" type stuff, which I think would result in samey and boring results. You'd have an FPS and a TBS which you could transfer art between, mostly, because most of the rest would require significant rework.