next »

[diegzumillo]

I decided to use the Design subforum because the important aspect of rapid prototyping isn't technical, it's related to design. That is my opinion but feel free to move it if necessary. Also, I searched the forum for this obvious topic and surprisingly found nothing. I mean I found countless pages of results but no dedicated topics I could find.

I am a seasoned Unity user but every time I see people on game maker spitting out fleshed out prototypes super fast in gamemaker it makes me wonder; is GM better for rapid prototyping? I gave the trial a shot but the trial isn't long enough to answer that question, there is a learning curve that eats most of the trial time. And it's not too cheap. I'm not in the US and $100 in my currency feeds my family of 3 for a couple of weeks. So it's not an impulse buy at all.

I also looked into Construct and it looks promising. Also not free and the trial is too limited to encourage a purchase. Then there's this thing I found called Novashell. Seems designed for rapid prototyping in mind and it's open source! Very intriguing! Very much dead too.

So the topic question is: what is your preferred rapid prototyping engine for 2D projects? List your tools of choice and make your arguments. For me and for posterity.

[nova++]

I've only messed with it a tiny bit but have you looked at Godot? Free and open source and quite actively developed. It has nice 2D support from what I've seen, way better than unity's hackjob. A while back I put together a basic platformer controller in less than an hour I think, with almost no prior experience.

[Foxwarrior]

I've amused my brother on a few afternoons by coming up with an idea for a game, prototyping it in Game Maker, playing for a bit, adding more features, and so on. This was long before I started learning Unity though... it takes me longer to build a little game in Unity now than it took me to build little games in Game Maker then, and I had less Game Maker experience at that time than I have Unity experience now, but I'm not sure how much of that is just Unity's significantly more difficult learning curve. I guess some things like being able to put quick 2 second drawings into your game in 2 seconds, and not having to remember deltatime in every single calculation, easy pixel-perfect hitboxes by default are genuine prototyping advantages of Game Maker.

[diegzumillo]

I've only messed with it a tiny bit but have you looked at Godot? Free and open source and quite actively developed. It has nice 2D support from what I've seen, way better than unity's hackjob. A while back I put together a basic platformer controller in less than an hour I think, with almost no prior experience.

I quite like Godot. With some practice I can probably use it just as fast as Unity for rapid prototyping.

I've amused my brother on a few afternoons by coming up with an idea for a game, prototyping it in Game Maker, playing for a bit, adding more features, and so on. This was long before I started learning Unity though... it takes me longer to build a little game in Unity now than it took me to build little games in Game Maker then, and I had less Game Maker experience at that time than I have Unity experience now, but I'm not sure how much of that is just Unity's significantly more difficult learning curve. I guess some things like being able to put quick 2 second drawings into your game in 2 seconds, and not having to remember deltatime in every single calculation, easy pixel-perfect hitboxes by default are genuine prototyping advantages of Game Maker.

That is the promise I keep hearing. I might need to bite the bullet and buy this thing one day.

It would be really cool to see an engine made with rapid prototyping in mind. Prioritizing that over performance and other things people generally want in an engine.

[diegzumillo]

Pico-8, and its fantasy console relatives, have a few of the features I would like to see in an ideal rapidproto engine. A integrated environment with everything you need. No need to manage files, it's all in there in the same place. It falls short of being ideal on the coding department. One of its design goals is to present a nice set of limitations to encourage creativity. But quick prototyping requires a faster way to get things working than implementing yourself a loop that goes through a list of actors, and then implementing the actor enum with all the variables you need even for the most basic task of positioning a sprite on screen etc.

[InfiniteStateMachine]

I still love haxeflixel

[diegzumillo]

It looks really cool!
It made me think a bit more about what makes the ideal rapid proto engine. It really depends on your general goal though. The ideal prototyping engine is the one that lands you closer to the end goal from the start, which means there is no ideal engine for rapid prototyping.

I actually just bought GM studio 2. I noticed they have a permanent license for a pretty decent price on Steam, specifically. Beats me why they hid that particular license there but anyway. GM studio seems ideal for traditional 2D games, which is what I'm doing now. But if I tried to use it when working on Apple and Worm it would never take off.

[Ordnas]

I made my 2d prototype with Unity, with the help of a 2D platformer asset pack called Corgi Engine, it has a lot of features already implemented, so I can start to prototype the level design immediately.

[pelle]

Godot is very nice to quickly throw some sprites into to prototype. But do not underestimate paper prototypes. Or just dragging around some graphics in Inkscape.

[droqen]

Hello, this is an old thread but I have a lot of opinions and I'm curious how they bounce off other people's thoughts.

I've been using Godot and loving it lately, but significantly more important than the engine you use is the approach you take, the specific ways in which you use the tools, the libraries you build for yourself, the goals you set, the goals you forget, ETC ETC ETC

For example, my latest project(s) was(/were) built around enabling a specific workflow of mine: making 8x8 tilesets and pixelart in GIMP, building small levels in Tiled that are, by default, 160x144 in resolution, and loading them into Godot. I've only really recently started to embrace this type of limitation: yes, it's possible for me to start building any sort of game, but if I set some constraints for my designs with the goal of constructing them faster and easier, turns out I can construct prototypes faster and easier.

I spent a long, long time making platformers, and I'd pretty much build them from scratch every time. I built up a lot of intuitive experience, and each time I built another platformer, it took less and less time to get the, say, collisions and movement feeling right. So that's part of it, too.

I'm not saying don't innovate, but you can save a lot of time if you actively self-select to work on new projects within a 'genre' (of your own definition) and work to support that choice. Doesn't matter what tool you use as long as it's flexible enough to let you design your own workflow.

[droqen]

Pico-8, and its fantasy console relatives, have a few of the features I would like to see in an ideal rapidproto engine. A integrated environment with everything you need. No need to manage files, it's all in there in the same place. It falls short of being ideal on the coding department. One of its design goals is to present a nice set of limitations to encourage creativity. But quick prototyping requires a faster way to get things working than implementing yourself a loop that goes through a list of actors, and then implementing the actor enum with all the variables you need even for the most basic task of positioning a sprite on screen etc.

Also my particular workflow is definitely inspired by Pico-8 feeling this way to me. Same with Bitsy. I adore the 8x8 sprite look and in particular the way that sprites cover up the background in Bitsy is something I wanted to keep. Lucky for me it also makes it easier to make objects out of tilesheets


e.g you can (just barely) see here that the player sprite, the golem, and those weird wind chimes at the top are rectangles with opaque background colour, it blocks out the background tile. i know it looks a bit bad and dumb, but it's a technical laziness that i really started to fall in love with as a stylistic choice

If you like the idea of the fantasy console but it's not built for you, build your own fantasy console in an engine you already like using the power of your mind (by which I mean, don't technically construct a fantasy console, just - as per my previous post - just decide what features you're going to re-use over and over again, and lean into that. Choose features that have flexibility, that you expect to be able to play with across multiple prototypes.)

Also, Godot does really well on the 'integrated environment' front. In particular I like that it feels a bit like "Unity with a built-in script editor." I never did get VS Code working flawlessly for Unity.

[diegzumillo]

Hi droqen!

I've been using Godot and loving it lately, but significantly more important than the engine you use is the approach you take, the specific ways in which you use the tools, the libraries you build for yourself, the goals you set, the goals you forget, ETC ETC ETC

That is a very good point.

What I said earlier "The ideal prototyping engine is the one that lands you closer to the end goal from the start, which means there is no ideal engine for rapid prototyping", although true, is not a useful statement. I think you brought something that I can use with your thoughts.

Modularity can be limiting on some ways but it's one way to ensure a more agile prototyping workflow. I could make something of a simplified "sub-engine" with unity, with a fixed resolution, fixed input etc, that is, remove most of the trivial decisions and stick to a specific framework, a set of limitations (I can see why you thought of pico-8 and fantasy consoles). That is not a bad idea at all. With those things fixed it's much easier to come up with modules for game logic that can be used repeatedly, like platforming controls, tile collisions, grid based logic etc, since the framework limits the possibilities.

[droqen]

yay, hi diegzumillo!

When it comes to fixed resolution, note I really only do this because it's a decision that I could see myself struggling with and I definitely plan to work in 2d, often without moving cameras. Peeking at the dev logs in your signature, fixed resolution wouldnt, for example, help at all if you were going the route of making more things like "Scifi flightsim".

Also, I spent (wasted) a ton of time trying to top-down this process, e.g build a fantasy sub engine from scratch: the perfect one. I'm shit at planning ahead so ymmv, but the process that I found worked for me was build a small project, converting small reusable pieces into easy modular bits as appropriate just by cleaning up the workflow involved with installing them into a new project. So if I made a level loader, I'd convert that into a piece of my library, and from then onwards commit to reusing that level loader in future projects, always aiming for as few lines of code as possible to use it in that new project (usually its reduceable to one simple line, but not always).

Sorry I did miss your quote and it's definitely correct the ideal here is to kind of help your tool and the end goal meet in the middle. Find a nice compromise if possible. Don't stretch either one so far it breaks, but do all you can because you profit in saved time & effort.

[droqen]

I will treat this like an engine and do some rapid prototyping with it.

It sounds like you're already kinda on this path! But you haven't posted anything else! What happened, if you don't mind my asking? In particular to that Mario clone!!

[diegzumillo]

I will treat this like an engine and do some rapid prototyping with it.

It sounds like you're already kinda on this path! But you haven't posted anything else! What happened, if you don't mind my asking? In particular to that Mario clone!!

(my quote is from the apple and worm devlog if anyone else's reading)

Yeah, it's still on my list! It has been several days since I even touched any game dev project, including that one, and I'm not sure why. I do have a lot on my plate personally so that could be it. But I have been craving an idea to get me passionate again but it's taking a while, because without side projects I get extremely bored. But without something that really sparks my interest there is no point in pursuing it as a hobby, you know.