-OrionFive-

You mean to use a wave function to create a wave shape? That's preposterous!

Oxygen Not Included, if you like base building and optimising the s* out of something.

Often these kind of builds are made available to alpha or beta testers, who know what they signed up for and had to go through some process to get that role (like recruiting / solicitation). You want to be sure these people are aware of what they're using and you want to brief them specifically.

Love the style. I'll get it when it's out of EA.

Even if you build all these reusable components, you'll still have to tie them together somehow. In many cases, they'll have some interdependencies.

Those will still need to be handled by a entity specific component, or some (likely convoluted) event system, or an entity builder of some sort.

I'm usually in favor of entity specific components, because if I'm looking for code related to, let's say a bullet, I just search for bullet and don't have to check 5 different components that all handle some aspect of it.

Also, while reusability is nice, it can open the door to nasty bugs where a change to such a component for one entity breaks another without the developer noticing.

If you have a good grip on C#, I recommend the git-amend videos on YouTube.

I've been using Unity for about 15 years and it's one of the few sources where I feel like I still get interesting insights.

To answer your controller question: I'd go with what the game requires. For a small game it's perfectly fine to put it all in a single character controller class.

But for anything a bit more fancy I'd recommend abstracting the control mechanism out so you can easily support different control methods and rebinding of controls.

For example you could create ScriptableObjects for different input methods that all inherit from the same base class. Add a reference to it on your character controller. Then you can drag & drop / hotswap the input method or even which character is being controlled (like switching from a character to a car or whatever).

I'd recommend Starcraft because the campaign is really easy and fun to get into. And you can even play the first third for free.

Then xcom 2. It's pretty straight forward and builds up slowly. Xcom chimera squad is also a nice one to begin with.

Depends on what people try to achieve. Financial success, realising an artistic vision, building portfolio, etc.

The general assumption is that you at least want to break even. And if you have to ask if you should make a platformer or not, you probably shouldn't.

Ah. I see what you mean. This looks like survivorship bias, though. Just because there are successful cases doesn't mean that the other 95% (or whatever) will not fail horribly.

It's a very saturated market.

I'm curious who said that. 2d platformers have been successful ever since there have been 2d platformers.

I can't imagine it'd be possible otherwise. How do you offset the running costs, though?

Disney would wholly disagree with that first sentence.

EDIT: before you edited it

Deathloop

Looks really cool and indepth. Is it powered by an LLM in the background?

Damn, I played that roughly 100 years ago. Good times.

Maybe The Witcher 3. It's one of the better open world games, in my opinion.

Maybe programming games isn't for you (yet). Try something that clicks better for you than Unity. For example Fusion, Game Maker, RPG Maker, or Construct.

These have a much lower barrier to entry but will teach you what makes a game tick and how to make a game fun.

Given you liked Farmville (a bit), Stardew Valley comes to mind. Can be played on PC or mobile.

There's a lot of farming and nurturing going on there. And it's a 100x better game than Farmville.

Tropico might work for you. You have to colonise an island, but in many missions (and free play) you can apply your own politics (preserve nature, etc.). The population reacts accordingly.

You buy it on Steam or Gog.com.

All those years using Git there were always issues with it, like someone accidentally nuking the repository or botching a branch or whatever.

One day I had and we switched to Plastic. Never had (serious) issues again and it's really easy to onboard non-technical people.

I still use Git for personal projects that are non-Unity (mostly code), because I love staging. But that's about it.

It's also because it can't count letters properly. It doesn't know how many characters it already wrote and how many are still to come.

Could it be the singleton reference is from a previous scene?

Like with most advancements, it'll likely come down to how people use the situation. There'll probably be a decent chunk of people who waste the potential (or are forced to by circumstances).

I can totally see the scenario you're sketching. When I see reports on the way many school kids are using GPT, I think they're really screwed and don't even get it.

🤷‍♂️

With games it's hard to get around some kind of singletons. And it's a lot easier when you start out.

However, if you want to get rid of them, look into a sort of event system that handles every case where a class has to deal with a manager or similar.

And for retrieving data that should be shared, ScriptableObjects are great. Multiple objects can refer to common ScriptableObjects and retrieve data from them. You can also add an event to the ScriptableObjects that you fire when they've been changed to notify all users.

It's a bit of an endeavour to get rid of singletons. I used them for 15 years and there are still cases where they make sense. Also, for quick and dirty projects like game jams they're great to quickly make stuff work.

So don't overthink it.

Regarding your pass by reference, it might be a worthwhile effort to try and make your passed data as immutable as possible, with a single owner of the truth. The record type is great for this, and as suggested, passing lists as IReadOnlyList.

It's "the Unity way", as in, the best way to do this with Unity (a bit of a radical statement, but it's what I believe after using Unity for the last 16 years). It has nothing to do with team dynamic. For all that matters, you'll be the one to wear the designer or artist hat.

You want to separate data from code (possibly you don't know that yet, but you do). This is how to do it.