Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Megamarc, what motivates you to make an engine?
#1
I figure this belongs in this board Smile

Megamarc, I'm somewhat intrigued that you put so much work into this engine for free but have never made a game with it. I imagine if I made it, the first thing I'd do is try to make a game with it, but then again, I'm not the sort of person who has choosen to put the time in to learn what it takes to create a whole engine (even though I would like to have the time to learn this stuff).  All of the work that goes into making this cross compatible and having all the language bindings is a lot of work and would consume a lot of your time. 

I assume you are confident a game can be made in it with your testing, but even if there are unforseen problems, I know as soon as someone runs into a problem trying to actually make a game with the engine, you do everything to fix it.  So in the end it doesn't matter if you yourself don't make a game with it, you'll make sure other people can.

So - are you consumed with the desire to make a game with it, like I am? Or is it enough for you that it exists and other people would use it?  And why is that enough?

(BTW I had to focus fully on something that will bring in money, and I haven't got an idea yet that Tilengine would be appropriate for, which would also bring in money... hopefully I get to do something with it someday soon)
Reply
#2
Hi dude!

That’s an interesting question, I’m glad you asked it. My first motivation wasn’t to create an engine at all, I just wanted to program some raster effects, to play a bit with them. I had just learned about raster effects in classic systems, I was amazed about that trick and the possibilities it had, I finally understood how many cool effects worked. So I started looking for a free engine, assuming that somebody else had created one with raster effects support. But such engine didn’t exist. There’s the Retro Engine, a closed-source, commercial engine developed by Christian Whitehead (https://christianwhitehead.com/?page_id=23) used to create all the official Sonic remakes and the recent Sonic Mania. But that’s not available for public usage.

I have big background in graphics programming, so I started creating my own “tool” to experiment with raster effects, that was the seed of Tilengine. I really hadn’t planned to release to the public, but as I was adding features and streamlining the API, I realized that it was a perfectly valid and unique product that could be used to power actual retro-style graphics. It proved to be a superb tool for learning: web hosting, web development, community publishing and management, open source licensing, cross-platform development, programming languages, cross-language bindings, documentation systems… With it I learned a lot of new things, many of them that I could use in my daily job.

Why I haven’t created a game with it? Easy: I’m not a game designer, artist, I don’t have any skills or experience designing a game. I’m a software engineer with a passion for graphics. But of course I’d really love to be part of a team creating a game with Tilengine. My role would be lead programmer, or at very least technical consultant providing very close support for Tilengine. But designing and leading development of a game is totally out of my reach.

Until now I haven’t received any serious offer to be part of such project. Nearly all requests have been of people requesting “x” feature because they have a great idea, but that missing “x” feature is a major blocker. As soon as I implemented and released that “x” feature, they vanished and their great idea was abandoned. Such “x” feature are: support for different editions of Linux, OS X, C# language, expanding the input scheme, releasing the source code… Now when someone complains about something that’s missing but that “would be great”, I only implement it if I think that adding it would make Tilengine a better product and that doesn’t conflict with my vision of what Tilengine should be. If not, I ask for a work-in progress of their game, which usually doesn’t yet exist. People doesn’t have problem asking for things when they’re free and they don’t have to put any effort on their side.

That said, some nice people have contributed to Tilengine besides myself: providing new language bindings, cleaning some code, giving useful ideas and suggestions, testing… that’s still a very small community, but I really appreciate every contribution I got!

How about your projects?
Reply
#3
You know, that makes sense when you put it like that - you're a graphics programming guy that took an interest in retro graphics and started to make an experiment.  So you came in from a different angle than I would have.

That's a fair way to filter out the worthwhile requests. Ideas are a dime a dozen and come and go, but rarely people actually even get as far as a prototype, and rarer still a finished game, heheh.

Well, I'm coming up on the release of "Super Jump 'n Bump", my re-imagining of "Jump 'n Bump", I think I mentioned my free remake already, that was a 1:1 remake that works on modern PCs with gamepads and 8 players.  But this next one is a paid version and has multiple levels, a "battle rounds" system (rather than just infinitely kill each other) settings to control everything in game, multiple characters with customisable colours, and a built in level editor that has gamepad support. It comes with 4 themes (hope to make it 6) and 9 levels (hope to make it 12) and is easily moddable to add your own themes, levels and characters.

Hoping to get Steam Workshop support in before release...

Here's the page:

https://store.steampowered.com/app/10124...mp_n_Bump/
Reply
#4
You've got that game running using Tilengine? That's pretty kick-ass. It looks like a solid concept, I look forward to seeing it in action. Might I recommend that you post some videos of Super Jump n Bump running? Not only would they serve as good advertisement, but they would also be a great way to promote Tilengine itself.

And to Megamarc, we are all grateful for your hobby project. It's a fun engine, and well worth using. I approve of your approach to handling feature requests. You've already made the engine open-source. Developers who are really serious about using it can always make their own changes and feature improvements, if they really want to. You should never feel obligated to kow-tow over every tiny request. You've already done so very much.

I've actually been looking into the possibility of compiling Tilengine for mobile platforms. As near as I can tell, it should be possible, even relatively easy. I just need to roll up my sleeves and get it done.
Reply
#5
@Domarius good work! Hope to se it finished and released. What tool(s) are you using to build it?

@Richard Kain: Domarius isn't using Tilengine to build its super Jump'n Bump. I really appreciate your kind words, and I'm glad you like my work. About porting to mobile platforms, yes, definitely it should be possible. I would closely support anyone willing to do it. Source code is cross-platform plain C with non-mandatory dependencies on SDL2 (windowing) and libpng (asset loader). These two libraries are universally available, and could also be replaced if they weren't available on the target platforms. At some point I had a build for Android using the NDK v10 (now obsolete), but I couldn't manage to integrate it with SDL2 and build the samples. I abandoned this build long time ago. Someone with experience in mobile development tools should do the porting, the NDK (I think it's now Adroid Studio) and Xcode for Mac are two beasts that require experienced people, not myself. Same goes with emscripten, the C to javascript compiler/webassembly that could bring Tilengine to web browsers. SDL2 and libpng are already available, but someone with experience should lead the effort.

Developers wanted to port it:
  • Android: NDK/Android Studio
  • iOS: Xcode
  • Web: emscripten
Anyone?
Reply
#6
Whooops, it crossed my mind to make it clear that Super Jump 'n Bump wasn't Tilengine, but in my excitement I forgot that bit Smile And yes Richard, a full fledged trailer is in the works, so a video will be there soon.

It's made using https://godotengine.org/ which I can't sing it's praises enough.  I've tried a lot of engines / development environments, including Unity and UE4, and if you want to make a retro 2D game, Godot is hands down the best balance of ease of development and flexibility.  I can work on the same project in Windows or Linux (or Mac, if I had one) with no problems, it's free and you can mod in your own libraries if you want, but most won't need to - right away I get a true 2D pixel engine, 1 click deployment to many platforms (including web and mobile) and this works - I built a mac version, emailed it to a mac owner, and they had no trouble playing it, even though I couldn't test it!  And it supports any gamepad I've thrown at it, so supporting gamepads has been simple.  And it uses GScript which is kind of like Python or LUA, you can just add variables or functions to objects and any code can use those properties or functions if they're present, without worrying about what object type it is.  It's made game development a lot less hassle.

But of course, Tilengine is the only thing that can do raster effects Smile
Reply
#7
Oh yeah, I've played around with Godot before. It's fantastic. Definitely one of the stronger choices for pixel work. I've actually considered trying to get Tilengine embedded into Godot, but haven't had the time to pursue it yet. I'm trying to begin work on a title of my own with some design concepts I've been playing around with, so I'm limited in how much I can devote to experiments. Also, there isn't nearly as much urgency for incorporating Tilengine into Godot, as Godot already has a good 2D implementation for pixel-style games.
Reply
#8
That's very cool, I hope someone is able to do that. My biggest concern is always multiplatform deployment. I got a Tilengine test project compiling on Windows and Linux, but I haven't set up a Mac (or Hackintosh, or VM) to do Mac compiling, and Mac is the second biggest audience to Windows (not Linux like I thought), so for the 2D games I want to do so far (the profitable ones anyway), there isn't a push to move away from Godot - I really want to attempt cross compiling that way for my own experience but I can't justify the time on it just yet. But part of me knows that I should really get that experience under my belt as a programmer. I'll see how things go in the future after some income happens from my games (haven't sold one yet). Seriously if I had unlimited time I would be prioritizing Tilengine compiling on Mac just so I know how much total work is involved making a game in Tilengine and where to rate it amongst my choices when picking a framework for my next game.

By going the route I am, I'm missing out not only on Tilengine's raster effects, but also MOD music, because the only way to get that these days is to use the FMOD library, which only Unity uses. Some other engines use SDL's MOD playback, like Love2D, but it does not obey the internal looping of the MOD track, which defeats the whole point of using them. Not only do I like the sound of all the available old MOD music at The MOD Archive, and the fact that they're only kilobytes in size for some amazing music, but I also feel at home making music as MOD files using OpenMPT. I really like being able to make an ambient track that goes on for several minutes and it still will never break the size of a few kilobytes (or a few hundred kilobytes, depending on your samples) rather than 1mb per minute of MP3/OGG. It breaks my heart, but I've had to convert MOD files into OGG files for my Godot game... Some people may say it matters less in this day and age, but I have always been a minimalist in every aspect of life Smile Which is another reason I'm still drawn to the idea of making my own game framework around Tilengine and FMOD.
Reply
#9
MOD Trackers are great, but they aren't the only solution. The advantage they provide is being able to use small but uncompressed audio samples as "instruments." But you could probably pull off something similar in Godot with its built-in audio system, you just would have to roll the solution yourself instead of using pre-defined MOD tracks. That would mean a custom-built music writing environment. But it would also mean greater flexibility in playback and performance.

A personal pet project that I want to cook up involves a similar approach, but instead of music, using it for character voices. Imagine having hours worth of spoken dialog that only takes up a few small megabytes. Even the more complex human languages never need more than about 40-50 audio samples for basic speech.
Reply
#10
Well if I could justify the time I'd use GDNative and implement FMOD support directly into Godot, no need to re-invent the wheel! FMOD already gives you intimate control over the MOD file playback.

That's a pretty cool project. My only concern is how bland it would sound.  As good as text-to-speech has gotten these days, it still sounds fake and uninteresting.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)