Unity 5 or Unreal Engine 4?

[Gender: Xenomorph]

Actually, I am already modding. I'm making a mod for Doom 2, a mod for Doom 3, a custom map for Star Wars Battlefront, a custom map for Quake 3, and a custom map for Jedi Academy. I didn't start all those at the same time, I would get interested in working on one, and either lose interest or get frustrated before going back to it. The reason I don't try to just plow through them is projects I work on won't look as good if I work on them when I'm feeling uninspired.

I don't usually have trouble with scripting. The biggest obstacle I tend to struggle with is getting custom sounds to work. This wasn't an issue in UE4, but Doom 3 is very fickle on what sounds it will let work. Even more so with Battlefront 2004; I couldn't get any custom sounds to work for my map in that game no matter what I did.

I started with WC3 custom maps/mods. That's a good start.

I'd say, avoid visual scripting like the plague. It's a trap. It's very unintuitive and takes ages to debug. Also it makes you dependent on visual scripting engines and you will never learn how to use and adapt with coding languages.

 

[Synthetic Smug]

I don't really think you'll be able to go wrong with either one, especially if you already have some modding experience and feel somewhat comfortable with scripting. I guess blueprints are supposed to be somewhat slow and under the hood and Unreal is something of a rabbit warren of legacy code if you end up having to optimize things but I've seen Unity tend to break small things in frustrating ways with their frequent updates.

 

[ZMOT]

I've seen Unity tend to break small things in frustrating ways with their frequent updates.

this is something I never understood, why update? you stick to one version for each project, and might update a separate fork if you run into an issue you can't work around. or you make a big jump after the game is shipped if you feel like it (since the working and finished game is already out there) for new features or better performance.
for the same reason they tell you to start with an LTS version where updates are slower and minimal at best. anything else is just making things more complicated than they need to be.

that argument always strikes me as "linux is unstable shit" after using either ass-old debian or bleeding edge arch.

It's a combination of me wanting me to mod the likes of Civilization IV: Colonization and Doom 3 but getting frustrated that I was hitting limitations with what those games allowed and not getting custom sounds to work in Doom 3 but in UE4. But yeah, I would like to have money too. One of my goals would be to sell my own version of Civ4Col where the Kings are more involved than just financially exploiting their colonies and only use their Royal Expeditionary Forces to pummel their colonies if they try to rebel. Seeing as the European powers did militarily assist and defend their colonies in real life.

isn't there an open source civ where you could add that?

I've heard the argument you're making applied to Quake 3 Arena maps and Battlefront maps. People told me to make basic and IMO boring generic Quake 3 maps, instead of following my ideas. While I kind of understand now because such levels are made to accommodate for the fact that AI bot generation isn't something you have much control over, trying to make a level with complex features helped me learn how to use them, and how to use them well. I feel it's best I try to work on what I'm motivated to work on, otherwise it feels like a boring slog rather than something I'm pumped to work through. Is this a false equivalence?

the reason people say that is because most people want to make their dream game fortnite clone in game maker without understanding how if/else works. in terms of maps that means basic (but not necessarily boring, ever put 10 people in a single cube with a bunch of boxes for deathmatch?). complex also doesn't necessarily mean better, there's a reason people still play de_dust in whatever form..

the point of that argument it's better in the long run when people start small and get the basics straight before moving on, either to prevent burn out or frustration if the learning curve is too steep (of course everybody learns differently, but that also means being able understand how to apply lessons and overcoming issues yourself, which most people simple can't, especially not in the beginning). takes a while to internalize KISS.

 

[Megaton Punch]

It's a combination of me wanting me to mod the likes of Civilization IV: Colonization and Doom 3 but getting frustrated that I was hitting limitations with what those games allowed and not getting custom sounds to work in Doom 3 but in UE4. But yeah, I would like to have money too. One of my goals would be to sell my own version of Civ4Col where the Kings are more involved than just financially exploiting their colonies and only use their Royal Expeditionary Forces to pummel their colonies if they try to rebel. Seeing as the European powers did militarily assist and defend their colonies in real life.

So basically you want to make a spiritual sequel of Civ 4: Colonization? That's probably doable as long as you're either willing to outsource the models or become a really good 3D artist. From a programming perspective, it shouldn't be very hard. In this specific instance, either engine should suit you fine. Games like Civ don't require amazing graphics or advanced stuff like raytracing. In fact, I'd argue excessive lighting would detract from that sort of game.

Sprites are a pain in the ass. I've made custom monsters and weapons for Doom 2 and know this. Models are tricky too, but I haven't invested the time necessary into learning how animating works to really complain about that yet. If I can get models I make to use the animations I download or the ones that came with the engine, then it's just a matter of making the skeleton and naming everything the same as the ThirdPersonCharacter.

I would highly recommend not using default assets except for testing purposes. Nothing makes a game feel cheaper. Even if you download free assets from the internet, that's still a much better route. Animations aren't as noticeable as models, but someone will notice and meme on your game and then you're "that guy who uses Unreal Engine default animations".

I've heard the argument you're making applied to Quake 3 Arena maps and Battlefront maps. People told me to make basic and IMO boring generic Quake 3 maps, instead of following my ideas. While I kind of understand now because such levels are made to accommodate for the fact that AI bot generation isn't something you have much control over, trying to make a level with complex features helped me learn how to use them, and how to use them well. I feel it's best I try to work on what I'm motivated to work on, otherwise it feels like a boring slog rather than something I'm pumped to work through. Is this a false equivalence?

It depends on your level of experience. When I first started making games, I had no idea how complicated something as simple as an inventory system or dialog box could be. You really do have to start at something incredibly simple and work your way up. Mapping isn't quite the same, because mapping ultimately isn't a mechanically complex thing. It's more of an art than a science. Video game logic is an absolute science and if you don't do it exactly correct, everything will break and you'll often have no idea why.

If you have experience with programming, especially OOP, then sure, start making a Civ clone. But do it iteratively. Make a game where you can move a settler around an empty map. Then implement cities. Then implement resources. And so forth. Don't try to build everything at the same time, because having 40 different partially-implemented systems will overwhelm you. It would overwhelm anyone. It's better to fully implement one system, then add on another.

In the game I'm making I'm using other people's free assets as placeholders until I learn how to make my own models that function. I have experience using Autodesk Softimage 7.5 when modelling for Star Wars Battlefront, but that mod tool I'm not allowed to make any money off of models I make with it. So now I am trying to learn how to use Blender.

Free assets are fine. Asset flips make tons of money all the time, as long as they're fun. Don't spread yourself too thin. Blender is a pain in the ass and you have to be a very specific type of person to enjoy using it. Every hour you spend wrestling with Blender is an hour you're not actually making progress on your game and you'll burn out.

 

[ZMOT]

Free assets are fine. Asset flips make tons of money all the time, as long as they're fun. Don't spread yourself too thin. Blender is a pain in the ass and you have to be a very specific type of person to enjoy using it. Every hour you spend wrestling with Blender is an hour you're not actually making progress on your game and you'll burn out.

to be fair, wrestling with blender might improve your skills with blender and modeling, which in return will profit the game. it doesn't really matter if it's blender or anything else, if you're stuck, work on something else (but not have too many "projects") or take a break to come back the next day.
there's also the "efficiency" argument, but imo unless you're a wagie or on a deadline it doesn't really matter, being stuck on something for days just means the learn effect is so much bigger (as in "I'll never forget that shit since I spend so much time on it") etc.

kinda related (the unity part can be ignored):

assets are also a moot point, even AAA studios buy assets and adapt them. what matters is the game is solid and the look fits together. people only ever notice the bad examples when devs failed to do it properly, otherwise how would people even now?
see (some short shilling but can be skipped):

as for assets:

https://kenney.nl/assets

should be more than enough for prototyping, or even a full game.