[Livestream Request] General solutions, methodology, techniques.

[ste1nar]

Generally the format of a stream looks like this:
Here’s something you can do [shows something fancy] and here’s how you do it [starts setting it up]. Which is great and all, very useful, but as a non programmer, non designer etc I often find myself wanting to create a system but really not knowing how to accomplish that system.

For example, there was a “finite state machine” stream. In my case I was late, I didn’t know what was going on and it was in c++ so I kind of lost interest. Later I read this: State · Design Patterns Revisited · Game Programming Patterns So I got the idea behind the state machine and (kind of) its use. So the book describes a problem and then gives a methodology to fix that problem. And that’s kind of what I am after.

So I kind of want a stream about general problems, programming techniques or pitfalls one might fall into and give a general overview of how to solve it, without going too much into the [setting it up] details.

It’s really difficult to come up with examples. Am I even clear with what I want?
I really just want to expand my toolset but instead where one toolset may be to know all the blueprint nodes, this toolset is about… methodologies?

[Raildex]

You basically want Epic to teach you how to program/code stuff.

[ste1nar]

@Raildex_;719401

You basically want Epic to teach you how to program/code stuff.

Depends on what you mean. Like the C++ state machine they did? As I wrote - no, not to that detail.
Rather, give an overview to a problem and a, perhaps common, methodology to solve that problem. Like a state machine can be used for one problem maybe there are other methods to solve other problems - I don’t know. It’s difficult to give concrete examples because I don’t know if they exist.

[SaxonRah]

No, this is a valid want. Even for advanced developers something like this is useful, let me expand. I talk with students learning game development and many of them are finding it hard to transition from one learned thing to another. Lots of them were taught nothing about how to problem solve or how to develop anything concrete, some of them were only taught how to use tools. even worse a few of these are kids with bachelors in game design and computer science and they can’t even write a simple game or app from scratch or even know how to begin. ( maybe this is a problem with my areas education, which is totally a possibility and a sad reality )

The burden of learning is not only on the shoulders of the person learning how to develop games, either from school or elsewhere, but it is also on the teachers and providers of code bases. They should want to help students and users be able to understand their thinking mentality so we can get them to better use the tools provided and so that they can better help new and old students and users. We don’t have to go crazy or even provide working concrete examples. but what we should provide is a somewhat high level/mid level view of general and no so general things.

a moderated developer roundtable where they all chat about how they might solve problems given to them by moderator. would be interesting sure, would it be useful? im not certain.

“How do i solve x in domain y?” is this person ignorant to the engine design? does x even have a direct solution in domain y?

example, "how do i change the camera to orthographic from perspective? " this how to problem stems from user being ignorant to where something is in the editor domain. the problem has a direct solution in the domain. this is how most tutorials are set up. solve x in y. there are many tutorials like this because it’s simple to solve these types of problems. much like P problems the search space is small.

example 2, “how would i go about implementing an rpg stat system for players and non player characters” this doesn’t have a direct solution in our domain, but there are many pathways to solve the problem. here in lies the problem with said open ended questions.What’s the scope of the game, is it multiplayer vs single, both with have wildly different code bases, sure they will have similarities but there is allot more to think about and to do when it comes to multiplayer. is the game a predefined genre like an fps or is it a new game idea entirely? what matters in your rpg system will probably be wildly different to another rpg. much like NP problems the search space is huge.

in a way we are trying to ask for a solution to an NP problem in P. that’s pretty hard.

If a game engine is a tool chest that contains toolboxes of tools. Knowing what toolbox to look in for the tool you need requires quite abit of information. you need to understand what it is you’re building before you know what tool you need to start looking for in the toolboxes. problem is, you need to know whats in allot of those toolboxes to understand what you are capable of building. many times the tool chest doesn’t even have the toolbox you need so you end up having to make tools from other tools that were designed to do something else. or even worse switch tool chests.

Here is what i see allot of. students and users of engine x use engine x for along time. students and users get used to thinking about problems and solutions in that domain, specific to the domain. for example ue4 has pawn and player controller player state those are specific to the domain of ue4. if you solve problems only using the ue4 domain specific things, then you move to say unity, it feels like you start from square one again. I’ve been thinking about this problem for a long time. im not sure how to remedy it at all. I don’t see allot of schools teaching kids how to make things on their own but rather teaching to be a part of a team to do so.