having tried to make my own engine yeah i don’t recommend it if your goal is to make a game.
there are people that can do it but they’re few and far between and they’re freaks.
if your goal is to make an engine, then fine, but if you want to make a game, implementing uh textboxes yourself isn’t the way.
The unity user who has lost trust for unity and looking for somewhere else to go. I guess it’s specifically godot vs anything else rather than godot vs unity, but yeah…
i’m only doing 2d stuff and i’m loving godot but i always disliked unity so ymmv
there’s definitely a learning curve (you have to learn how godot wants you to do things, at least so can make informed decisions about how to structure your project) but the manual is much more competent than i expected.
of course i’m comparing godot to a bunch of frameworks i’ve tried in the past so the fact that it has out-of-the-box solutions for common things in games makes it way ahead of what i was using doing before.
i’m surprisingly into the workflow – it matches how i want to make a game:
the great thing about this flow is it pretty seamlessly transitions you from prototype solution to formalized working solution. you can test things and play around with stuff with minimal effort, and then go right from that testing to the more formalized solution.
that said – there are definitely bugs in godot and rn i trust 3.5 a lot more than i trust 4 because at least most of the issues are already known in 3.5 and a lot more games have shipped w/ it.
ugh, i’ve written and re-written this a bunch of times now and i’m still not really confident i’ve made myself clear but here you go:
pros:
cons:
all these problems are going to be made particularly acute by the influx of people leaving unity (especially porting), and i hope a resolution can be reached wherein the community has greater control over the direction of the engine, and a greater capacity for contribution, especially in the small-but-meaningful areas that make daily use smooth and friendly.
i hope i don’t sound too negative. i think godot is great to use, i just worry about its management and future. obviously i’m a little on-edge because i very recently had to contend with the possibility that the tool i’d been using for nearly a decade was going to be unceremoniously pulled out from under my feet, so i’m on the lookout for less contentious, more stable environments to work in.
anyway i hope this at least goes some way toward answering your question. ^^’
Dont really have the energy to write an effort post about godot but I want to point out a couple things that haven’t been mentioned
First one negative of being run by FOSS people is that tendency to excitedly future proof features without really thinking about the present. Godot 4 uses webgl2.0 which is very cool but not stable or production ready at all and not even supported in the mono version. So their solution is use gd3 for web I guess 
. This happens a lot with these types of projects but godot4 has been particularly rough with it, breaking other important features like plugin and language binding compatibility from 3, though the community has been bridging the gap
Second my own experience with godot. It just works out of the box in a way that unity never did for me. Pixel perfect resolution scaling in unity is an ordeal. Godot has a project setting for it.
Third the ways that it doesnt work, you can just fix. I dont think this can be overstated. Googling issues with unity and seeing a bug report from a year ago that hadnt been addressed was way too common. In godot you can just compile from source and deal with it yourself, and in general being able to move performance critical code into the engine is extremely nice to have.
Overall I am really enjoying godot and while I know that honeymoon phase will eventually end and Ill come away with all the worst little issues with the engine, it already feels like its been better for me than unity ever was.
I do find this really strange because like in theory they could have proprietary engine modules available with proof of sdk license and they even say as much here:
Their main reasoning seems to just be that it would be too much work but if some of the core devs are already working on porting anyway it becomes hard to believe its not just a money thing… I guess we’ll see if they change their tune with this big influx of donations and grants as a result of the unity fiasco
this looks…interesting GitHub - ezEngine/ezEngine: An open source C++ game engine.
will defs check this one out, seems chill. if you give it a go pls post, I’d love to hear your perspective.
had a student mention that they were using unreal 3 and i thought that might be a compelling way to go as well. i haven’t looked at how flexible it is but between that and source i should look more into abandoned engines – especially cos some of them still use bsp 
slowly grinding away at a godot tutorial on moving things, colliding things, etc. unfortunatly the game the tutorial is based around sucks. i feel frustrated because i want one of my models walking around, and thats probably a ways away as i figure out how the asset pipeline works.
slowly figuring out 3d characters in godot, which is mainly trying and failing to import animations and having it do stuff. I know more than I did yesterday. I think soon I’ll have something walking around
will discord let me hotlink? lets see
ah yeah anyway, after a month of fiddling I have a walking character controller with mouse aim, and today I figured out how to attach objects to bones (like guns) and then give the tail some physics with a 3rd party jiggle bone script. this is almost fun to work on.
are you using 3 or 4? i’m using 4.2.x atm as part pf a series of little projects (working as a pair fot the first time) and finding myself getting tripped up wayyyy more than when i was using 3. just lots of little things like texture filtering and addon management and ide integration; setup being more complicated than i remember and settings not where i expect them to be. i think it’s just a matter of eliminating another layer of (godot-specific) muscle memory, but it’s still frustrating.
I’m on 4.2. I am not doing anything particularly fancy but I’m approaching the point where I have to script bullet collisions and I don’t want to, because combat is where things get hella janky. There’s no more…IK support? It’s there but it’s depreciated. I’m almost at the point of knowing where everything is or how to find it.
i’m way too busy rn to be pursuing this project seriously but i’ve been keeping track of engines to look into at some point:
stride: big fancy unity/unreal type engine w C# support. idk about how nice it is to actually use tho. kinda doesn’t seem like, game-oriented as much as sim-oriented. which seems not good.
coppercube: quaint source 1 lookin engine with a focus on accessibility over flexibility (at least as far as i can tell).
armory 3d: the blender game engine. idk if i wanna make games in python cos like, i want my games to run fast personally but still. integrating the engine into the modelling tools kinda seems like a no-brainer to me.
leadwerks and ultra engine: both made (and presumably used) by leadwerks. neither of them seem like… particularly user friendly. but w/e they exist. oh and they’re both premium, so.
gdevelop: heavily focused on 2d and no-code development, neither of which i’m really interested in. but it recently integrated real 3d, and i know you can code explicitly if you want.
if construct wasn’t 2D-oriented i’d be interested in that too, despite its monetisation scheme.
o3de: formerly lumberyard lole. ehhhhh seems suspect but i might as well try it at some point. i got a buddy who even worked on lumberyard so maybe he can gimme some tips.
ok this is where we get into the really dumb stuff:
titan engine by esenthel: “The Most Powerful Game Engine in the World.” lolllll this one looks goofy but i mean why not. it’s got real 2010 lookin UI so it’s cool in my book.
neoaxis: my notes say “seems sus” but i can’t remember why. looks pretty normal from here.
and finally, cafu: lmao. graphics peaked in 2010 and this engine knows it. hasn’t been updated since uh… 2019? honestly if i could get this to run nicely i’d be into it but given the single test i did with it i’m not confident. still, it’s a funny thing to include.
as you can maybe tell, my search criteria has shifted a little; i’m basically looking for the (probably non-existent) tool i can use until i die, regardless of philosophical foibles. i don’t mind learning new tools, but i’d prefer to like… make things.
anyway if anyone has any experience w and of the above (first or second-hand) i’d love to hear it.
literally moments after posting this i discovered the existence of castle engine so i guess add that to the list too lmao. it uses pascal!
Blender used to have a game engine integrated into the program but it got taken out at some point. actually…uh UPBGE exists as an offshoot of it
looking at some of these engines reminds me of when I was just getting started and didnt know anything, heck I didn’t even have an idea for what 3d game to make, i just downloaded things and poked around.
the engine I tried first was I think from a book called “3D game programming for teens” and it was incomprehensible, but came with the Torque engine, which is still around
thanks for posting this, i’ve been having fun poking around in it - still trying to get a sense of how usable it is but it’s been nice to find a 3d engine that’s both stripped down and jankily generous in what it gives you right out of the box (incl a little tile based draw-a-room editor!)
When are you gonna tell em that you’ve returned to the abyss dogs my dear
10 posts were split to a new topic: big OOP
