Recently we've been doing weekly updates on our insider forum, the prelude alpha is nearing completion and we've been keeping those waiting for its arrival informed on our progress and what challenges we've faced and overcome each week. It's been a very busy time for us but now we see the light at the end of the tunnel and we thought to do a more summary public update rather than the usual insider update.
We set a lot of extremely ambitious goals for ourselves with Sui Generis, we have to admit some of what we set out to do sounds a bit high-reaching. It's how we imagined future games when we were teenagers and that idea has stuck with us. You might find it a little concerning as a backer but there's quite a bit of what we wanted to do that we just couldn't be sure would really work. It made a lot of sense to us, we certainly thought we could do it but with no one having attempted it before we just couldn't be sure. There have always been moments of "Oh dear, what if this just causes too many problems?" There's also been a few times in development when we've imagined just how much easier our lives would be if we'd have gone for something simpler that's tried and tested!
Much of what we're trying to accomplish is far beyond the scope of the prelude and that's partly the idea of the prelude. It's an intermediate goal that is built from the same blocks but doesn't yet need to take advantage of everything they provide. In the end we've invested more in the prelude than we originally planned in an effort to do things right. Core to our game philosophy is that nothing is a static asset or effect, a predefined animation or scripted behaviour. We need to start with such things as placeholders while prototyping but in the end everything must be complex, mutable and reactive. If anything is possible then it should always be possible and be a natural result of the underlying simulation. We're not just talking about physics here but rather how everything has a meaning or purpose, how AI understands what things are and how they may relate to other things. Our event system and AI are aimed at providing an emergent story but we believe the most important and challenging aspect of this is the little things that drive how that story unfolds; the actions and objects that are instrumental to what actually happens and how what happens and happened previously can be perceived by AI. This has always been central to our design.
So far we've been very busy with overcoming the technical challenges of how to run a game where everything is dynamic, interacts and all motion is governed by physical forces; how everything can be described, how information can be accessed, interpreted and exchanged; how things behave in consistent and plausible ways rather than through simple schematic models; how AI can form opinions and make decisions that give them access to potentially unlimited options and roles, and do this in a lifelike and natural way rather than through a rigid set of unscrutable rules.
While on the surface the prelude is quite simple we have always remained true to our goals, it is built on these things and they are functional. It shows that we've already overcome the most serious technical challenges. Really we could not have hoped for better results. SG's physical world works beautifully and it can communicate, beneath the surface it is so much more than the typical series of player activated things, it is bursting with meaning and potential ready to be unlocked.
Now we're less concerned with technical issues and what could go wrong but rather with how to take advantage of everything we've done. Nagging concerns are being replaced by a flood of ideas about what can be done with what we've put in place. We still have a lot of work ahead of us, there's no mistaking it. We have developed models and methods, algorithms and data structures but ultimately it's the data itself that drives the world. We still have to design many of the things themselves, the activites or specific behaviours, the knowledge that governs them. This is an incremental process however where we can gradually expand options, understanding and possibilities.
We realise a lot of this may seem quite abstract but it's difficult to go into detail in a concise manner. Some people might also find knowing how things work breaks the illusion. If you are interested and have access to our insider forum, you will find more information about our AI systems, about how things in the world carry information and many other things besides. We plan to continue making frequent posts with detail of our progress and some of the game's inner workings.
On a more practical note, the prelude alpha is indeed very near completion. We've solved every major issue we're aware of, including those mentioned in our last insider Sunday update. This last week we've stopped to take a breath for the first time in a while and just look at what we're doing which is what inspired this update. It's very satisfying and exciting to see everything working smoothly. This week we'll be finalising some user interaction features and work on an actual release build. Performance on lower end systems is still potentially a concern at the moment (this is not a serious issue and will be solved completely) but we think it may be acceptable for a first alpha release.