Discussion To "backward compatibility" or not to "backward compatibility"!!

[Mike]

Hahahahaha....I'm totally the wrong person to try and tell how to make games! And yeah.... hate to break it to you, but I ALWAYS make games I want to play - and so should you. If you don't, there will be no passion in them, they will be lackluster, lacking "something" because you can't put in the sparkle that makes them great. If you don't love playing your own game, why would anyone else? Playing your own content and loving it, is how you polish the feel of the game - sure you can add in suggestions, but you always make things YOU want to make and play - THIS is gamedev 101. I stand by my record.

Same goes for GMS2. I'm making a tool I'd want to use. If you don't, then you end up with a room editor like GM7. Mark never used it, so it was severely lacking in usability and features. Putting in stuff you think others might use is no way to create anything. I'll happily take suggestions from folk for things we haven't thought of, but just because you suggest it doesn't mean we agree. If thousands want it and we don't, then sure... it'll get added. But it's very easy to be blindsided by a loud few in forums, even though in reality, it's only those few who want a feature.

as i said.... you may not agree, but we do listen and evaluate each item.

 

[SilverSurfer1221]

Yeah, I have to disagree with the comment about NOT making games you want to play. The old game, American McGee's Alice, comes to mind. American McGee had always wanted to make that game. He even had to shop around for a studio that would let him do it, if I remember correctly. And when it was released, it was a hit! It was unusual and it was fun. I still enjoy this game from time to time.

I also have to disagree with the idea of getting rid of GML. I am a non-programmer who wants to create games. I am an artist first. I really don't want to learn C#, C++, or some other full-featured programming language. I would rather script or use a language like GML that is geared toward what I want to do ... even if it is limited compared to something like C++. But, then again, I own a company and make my living with my art. I don't need programming skills per se (apart from GML).

 

[GMWolf]

Getting rid of GMLnis definitely the wrong way to go. I like GML. Its far from perfect, but I like it.

I think it would be far better to either add to GML, though that may lead to random syntax syndrome, or, far better, add a second language, like C# or JavaScript.
The second language would not need to interface with the engine. If it could simply create methods that could be used called through GML, that's ebouh for me. (Similar to how DLLs work, but all worked on in IDE, and running with less overhead).

Adding another language would preserve backwards compatibility, but allow more advanced users to use a perhaps, more complex language, But be able to use classes, and existing libraries.
This would be a great way to add new data structures and algorithms, that are less than ideal to add using GML. (GML code suffers from array spaghetti, when building data structures.)

 

[mMcFab]

There's no way I'd want to get rid of GML. Maybe expand on it, but it's so nice to use in comparison to other languages it would seem silly to get rid of it now - it's one of the reasons I stay anyway. Expanding on it further would be really nice (and some things seemed to be hinted at for in future - though no promises), and we already got some nice extras too (array literals, better macros etc.). Heck, even if I'm not all for 100% backward compatibilty (though I can understand it), removing/replacing a core feature like this would make the transition awful and EVERYONE would have to relearn the tool. That can take a really long time.

If I need something GML can't do, I just write an extension, the tools are there (obviously, not everyone can, but that's what the marketplace is for).
I'm actually pretty sure you can create nearly anything with pure GML anyway though. Sometimes it can be slow (particularly when dealing with arrays), but there's usually a way to write a native extension if you are desperate.

I suppose one thing I at least keep forgetting is that GML isn't necessarily finished. GMS2 may have overhauled the IDE and we got some really good updates for GML, but that doesn't mean it's done. Heck, shaders were added more than a year after Studio was released - who knows what could happen!

I've never made a game I wouldn't want to play either. If I don't like it, I stop. If I have no passion for it, there's no motivation either. I probably also wouldn't want to promote a game I didn't like even if I had somehow finished it.

I can accept some things won't change. There's always that little bit of hope for some things, but if the devs have their reasons there's not much I can do anyway. Like I said, they're probably much wiser than me! I'll also try to be a little less frustrating in future - after all, they've got enough to deal with as it is.