I have been holding out on writing a new post until the 2.0 patch was finished, but it has been too long, so here is a status update.


The good news is that you can play the 2.0 alpha already. Right click on the game in your steam library, click on properties and then channels and select "save_breaker". Do note though that there are A LOT of breaking changes, so you might want to back up your saves and start over (in the future there will be an upgrade guide and the game will try to upgrade things automatically as well as possible). Also be aware that there might be some progression breaking bugs in the new campaign.
You can find a list of some of the new changes on the community wiki: https://turingcomplete.wiki/wiki/Save_breaker_changes


Anyway, you might be wondering if the next version is Turing Complete 2.0, where did Turing Complete 1.0 go? The answer is that with a little bit more polish the current version would have reached 1.0. But instead I went on a side mission which grew and grew in scope to the point where the new version is a total rewrite of the game. In retrospect, I should have spent another week polishing the current version of the game and exited early access before setting off on this epic rewrite. However the current version is good enough (sitting at "overwhelmingly positive" score), which is more important than version numbers at the end of the day.


So why did the scope balloon so much? There are a few of different factors:

1. My outlook
Some developers release their game, their sales spike on launch day, then taper off and they subsequently abandon the game a few months later. That wont happen with Turing Complete. Turing Complete will be relevant as long as computer science is relevant (forever) and therefore it will not be abandoned. This is by the way the norm outside of games (windows, photoshop etc weren't just abandoned). From this perspective, a delay in releases is insignificant compared to laying a solid foundation for the game to stand upon for decades to come.

2. Discovering what the game should be
The idea behind the game was originally just a simple exploration of how CPUs work in a puzzle game. But over time the philosophy morphed and now I think of it as a computer learning / creation / exploration platform. This is why Turing Complete 2.0 supports user created campaigns, much larger circuits, more realistic campaign and having a powerful assembler framework is part of this. In any case, each time the philosophy changed, a lot of things had to redone to fit. Luckily, the current philosophy hasn't changed in a while and has now solidified to the point where I can't imagine it changing again.

3. Minimizing save breaking patches
Since this is a save breaking patch anyway, I want to take this opportunity to include all needed breaking changes in this patch. This way I don't need to break player saves more times than necessary. An example of changes that could have waited if they weren't save breaking is redoing the component shapes (some were too small, others use non standard symbols etc). Another example is updating the campaign to be more realistic and going further into modern CPU optimizations. In fact, I really hope that this will be the last time I need to break player circuits.

4. Making a compiler
I have made interpreters before and did not find it difficult. I was surprised that people treat it like it is a big deal and even write whole books about it and so on. So I assumed that people were probably also exaggerating the difficulty of building a compiler. But no, building a compiler is actually extremely difficult, especially the back end and optimizations. This alone took me over a year, but the compiler is essential for allowing custom levels, fast simulation, fast compilation (not done yet) and certain features that will come in Turing Complete 3.0.

5. Expanding the campaign
I have spent the last couple of months building levels about modern CPU concepts, like superscalar, pipelining and speculative execution etc. This requires deleting, adding and rebuilding levels over and over again, for each issue or improvement opportunity I find. This is a lot more work than people think.


So what is left before release?
First, finishing up the new campaign levels. This has already taken about 3 months and it is a bit unpredictable when exactly it will be done (since I keep discovering something that makes me redo the levels). However I think we are getting close.

Then I need to fix a lot of bugs, get things translated (which can be done at the same time) and make new solution videos.

The next update will likely be when Turing Complete 2.0 is done.