Course of Temptation Dev Diary #81

Hello friends! The time has come for another dev diary.

First of all, I've just updated the early access build to let you see the changes made this week. I know it's just another day till public release, but even so!

If you've tried the new writing mini-career, then you've probably noticed that it's pretty good in terms of income. Well... there is such a thing as too good, so I've been working on tuning it a little bit to bring it more in line with other moneymaking methods. I'll be working on this for the rest of the day, I think — I also want to add a bit of a notoriety system so that, similar to other activities, you're starting small and building your way up to maximum income.

When I posted this month's early access, I said I wanted to add to it on the way to public release, and that's what I've done. I had it in mind to add the creation of illustrations and photographs (I have some ideas for those) and build on the creation aspect of this update, but then I was merging in some contributions and looked at Noonycurt's prototyped changes to clothing for the purposes of adding weather, and found myself thinking about that.

Weather, seasons, and seasonal events have been a major roadmap item since the very beginning, but they've never done well on those what-should-I-do-next polls. What is to be done with roadmap items that I consider necessary for the game's final design but don't win polls? The answer I've found is to, instead of devoting a whole roadmap update cycle to them, just start doing them piecemeal in monthly updates. So here we go.

(Forecast dialog is in progress and still needs a bit of work — kind of hoping Drakon takes a look at it! Me and CSS... we do not get along.)

So, this update will start to bring in some weather-related stuff. I'll say up front that if the idea of managing temperature and stuff doesn't appeal to you, then you can turn all that off under difficulty options. As with many aspects of the game, any friction here is meant to be light — weather stuff should nudge you towards interesting aspects of the game (like giving you an impetus to dress skimpier) rather than being a huge obstacle to overcome.

So if your outfit doesn't quite suit the temperature, it'll drain some needs a bit faster but it isn't really a huge thing. Also, forcing you to bundle up somewhat runs against the game's nature, so through most of the year things are quite temperate. The ultimate goal is that the player should still be able to wear basically whatever they want at least the majority of the time, but the outcomes that the weather system produces should be interesting for the player to engage with.

To design the weather system, I began by dividing the year into blocks. I decided on having two blocks per month: early Autumber, late Autumber, early Gourdtober, and so on. Each block then has an average low temperature, an average high temperature, and an average precipitation level. Through all of these we can derive the actual weather: clear, overcast, rainy, snowy.

The question then is how often to roll the weather. I considered doing it once per day. A day would be sunny, or rainy, or whatever, and that was that. It has some advantage in simplicity for the player, but it also means we can't create situations like suddenly being caught in the rain. Also, as long as we don't shift conditions drastically too often, it doesn't matter so much if weather is rolled more frequently.

So I decided on once per hour. Now, rerolling weather a lot has some obvious pitfalls. For example, if at any given time there's a 40% chance of precipitation, then you will frequently end up having rain at 4pm, clear weather at 5pm, more rain at 6pm, clear again at 7pm, and so on. It feels random and unnatural.

Instead, I decided to use a simplex noise algorithm. In game development, simplex noise is used for things that we want to feel natural and organic, such as terrain generation and, yes, weather. It involves a bit more continuity: conditions at 5pm are still pretty similar to conditions at 4pm, so if it was raining at 4pm, there are pretty good odds it'll still be raining at 5pm, but it may well clear up at 6pm and remain that way overnight. It's no longer just a coinflip every hour. It feels a bit more like real weather, and it also helps prevent things from changing too frequently for the player.

I allotted about three days for this, so I got as far as building the baseline simulation described as well as handling temperature effects. Other weather effects (like rain) do not yet actually do anything, but expanding this system will be a short-to-medium-term priority for updates, making sure that interesting events come of it. Longer term, seasonal events and changes will figure in.

For the next monthly update, it'll probably be a good idea to focus less on all new features and more on building up some roadmap items like this one that are currently in piecemeal progress, plus working on general polish and expanding event sets.

However, my goal in the immediate future will be a return to the Greek house update. When I left off, I was working on creating specific party activities, which is really the last truly major piece of that update. It'll need a bit more connective tissue as well, but after the next couple of weeks, I'm hoping the end will be in sight.

I'll be back next week with another progress report to let you know how I'm getting on, and you'll also hear from me tomorrow with the public release for the June monthly update and its final patch notes. But until then... happy Friday, and thank you as always for your support, it really does mean a lot!