LudumData

All this talk about “From scratch” vs Game Engines

[raw] LD18LD #18 - Enemies as Weapons - 2010 17. Aug 2010 · 22:39 UTC · 2 days before compo start by psnake · 0

I agree with the anything goes rule. This enables us to focus on our game design ideas, rather than painfully trying to open a window and making your draw_pixel() function to work with any pixel format (this is kind of a bad example since it’s kinda of trivial, but still reflects my point). But still I have this urge to learn assembly this week and do everything really from scratch 😀 (maybe in the next LD). But this time I’ll use Python + Pygame (nothing else to be kind of midway from low to high level :P).

In my opinion Python has the best productivity and Pygame is really a wrapper to SDL (with some extras) that isn’t a nightmare to cross compile across different OSs. Still both don’t give the edge that Unity, Unreal, etc give so I don’t feel so dirty 😛

Damn, now this assembly idea is stuck in my head. I still want something multi-platform… Oh, I know, NES assembly xD (with my rom running on an emulator)

Comments

18. Aug 2010 · 00:25 UTC
If you make an NES rom for an LD you deserve at least 2 internets.
18. Aug 2010 · 02:06 UTC
See, this is where I think frameworks (but not full engines!) are a good way to go. This way, you don’t have to mess with the most basic of issues, like window initialization, texture loading, and drawing a square. On the other hand, you still have to program all of the actual game stuff yourself, and you can’t use handy dandy AI or other goodies from highly developed engines.
18. Aug 2010 · 02:53 UTC
I agree with Entar. I hate entries that use Game Engines. I think they give you too much. I also think starting from ‘scratch’ is not wise, as you reinvent the wheel each time. A framework is the solution, and I wish everyone would stop using the GameMakers and such to create their entries.
sf17k
18. Aug 2010 · 04:00 UTC
I don’t understand what “from scratch” means. You can code in C++ but then you’d be using a C++ compiler that took a lot of people a long time to make. That’s not “from scratch”, you’re still using a tool that you didn’t make (without which things would be much more difficult indeed).
VillemVillemVillem
18. Aug 2010 · 04:45 UTC
While I haven’t seriously tried it, I think using GameMaker and the like directs & constrains your thought. I think making games with tools that focus on simplifying the matters that you are likely to use for an “average game” as imagined by the makers of GameMaker, results in less innovation.
The Jahn
18. Aug 2010 · 07:20 UTC
Everyone who posted in here has good points. But it all boils down to the taste of the individual, really. Nothing really gives an unfair advantage, since that same advantage is also available to the other members of the competition if they choose to use it.
18. Aug 2010 · 12:07 UTC
I am a “from scratch” guy myself (except for one drunken flirtation with OGRE) but I think that the compo should be open to anything, as long as it’s open and available enough that other people can recompile and modify your entry for education purposes (i’ve learned a lot this way).
SonnyBone
18. Aug 2010 · 15:41 UTC
I see LD as a way to come up with an idea based on a theme and slap together a working prototype as quickly as possible. I don’t see it as a BUILD YOUR ENGINE FROM SCRATCH compo. I’m all for game engines, as your creativity and interpretation of theme can shine through the troubles of coding an engine from scratch. Let’s focus on the fun, not the technical. This is about ideas, not coding prowess.
Parthon
29. Apr 2011 · 02:34 UTC
I lost an entire compo to trying to get java and jogl to work and failing. I wish I had a working engine before going into that competition so that I could have actually got started on gameplay. As it was, I had to drop back to c++/sdl and didn’t have time to finish.