Next Gen ZC. (Otherwise known as ZC3.0)

[Gleeok]

I understand that you might inherit flaws, but consider that by the time you make something wholly new -- another ten years -- most of those flaws could have been fixed, and the entire programme improved upon. You also get the advantage of an extant community of coders, and users; which you would need to establish. This, coming from a man who waited nine years to start using ZC 2.5, thinking it a hopeless, dead project, for the last six of them, until RC1. I avoided developing any project under 2.50 due to instability, and perpetual internal changes, for nine years.
.

I just have to point out the fact that 90% of the time it took for 2.5 to be finished was getting the new code to work with the old code, not because adding new features is hard or takes 9 years. There was so many bugs at one time that the ZC bug forums were split into many subforums.


Also, I did take a look at "TilePattern" and co. It looks like animations have to be exactly 4 frames and there is a static frame counter in there as well, meaning all animations are the same duration even though a TilePattern is virtual (More red flags.. This actually doesn't even make sense). I was also pondering, it seems, about how a "Tileset" was a Bitmap with extra tile information. I was also correct about how there is no true low level data design in the rendering pipeline: Everything is object-oriented and treated as though only one ever exists, then shipped to SDL. (Unless I just happened to miss a file or two) :/

If you talk with him maybe just ask him about that.

[ctrl-alt-delete]

@Gleeok , if you hate C++, why are we possibly programming it from scratch using C++?

[Samer]

Well, C++ is the most cross-compatible.

C# only works on .NET platforms; Windows
Java springs up a bunch of memory leaks, plus you are required to have a JVM (Java Virtual Machine)

It's for the greater good. I can work with C# if need be.

[Zim]

I don't know about building another version of Classic from the ground up.
http://youtu.be/N3J2PZzuA0w
Here is a video of some functions I coded for Zelda Classic.
Its just more scrolling functionality, demonstrating warp return points and the like. (Also looping world map for Epoch, [you might want to skip from 1:50-2:20] which is fully functional and cannot land on solid combos even. I am aware that the enemy death graphic is a small boy.)
My PC is a little outdated and not very powerful, so please don't hassle me about the glitches/bugs.
The overall speed and everything of the engine is just fine like this.
Since my game doesn't use one single built in combo flag or anything like that (aside from Solidity), a build of ZC with all those things factored out of it would make a great game engine.
This code I wrote could actually replicate Chrono Trigger to the T, but it plays more like Zelda and looks like Chrono Trigger due to the fact that I haven't wrote a battle engine (yet) and probably won't.

Let me know if you're interested in having this file for yourself. It does what is described that 3.0 will be able to do with 2.5. The difference being that I overwrote combo flags to get it done instead of it being hard-coded into ZC.

I will only send the file via PM per request in an effort to keep it off the database pages so people don't see it and start using it thinking there will be updates for them like its 3.0 alpha build or something.

[Chris Miller]

Plain old C might be the way to go.

Take that with a grain of salt, as my programming abilities are rivaled by my ability to model Speedos.

[Gleeok]

@Gleeok , if you hate C++, why are we possibly programming it from scratch using C++?

lol. I guess nobody got that joke.

I suppose it's just something you have to just accumulate over the years of working on c++ to understand. c++ is one of the few languages that lets you do both low-level and very high-level programming, and everything in between.

Let me put it this way: Essentially c++ is like a hardware store; you can use a tractor and a chain-saw to make an end-table, or you can use a lathe and some hand tools. From the computers perspective [when I said "typical c++"] it sees programs as using bulldozers and wood-chippers.

So when I said I hate c++ I mean to say it's so easy to completely fuck up even something that could be considered trivial. I know this because I've done it over and over again. If anyone ever tells me otherwise I will show you someone that has either never tried to solve an actual problem with c++, or someone that who doesn't understand any problems.

tl;dr, No, c++ is fine. >_>

[ctrl-alt-delete]

I think we should make ZC 2.whateveropensourcefinaledition. Then we should completely leave Zelda and Zelda-like engines behind focusing on FFC.

[Samer]

I think we should make ZC 2.whateveropensourcefinaledition. Then we should completely leave Zelda and Zelda-like engines behind focusing on FFC.

I like this idea, OOP has been the mainstream since 2001. Data oriented programming is difficult for me, I could just objectify the data and that's that. Let's do FFC before Nintendo does.

[ctrl-alt-delete]

Is starting from scratch really starting from scratch? Is ANY of the existing codebase usable?

[Samer]

I don't think anyone with any sense would make something from scratch, unless you are extremely ambitious and never give up.

I'm not a kid anymore so if it takes too much effort I could care less.
I like ZC, which I stumbled across AGN about 2003 but never formally joined.

ZC, in earnest, is a joke of a game engine in my opinion.
It's good for beginners though who like the tediousness.

Programming is easier than scripting for me.
I only script for cross compatibility, fix some outdated applications, resource files (mainly pictures). Which are normally done in Assembly or XML. The only thing I can find scripting good for in a game engine is for AI, for which the base structure of enemy and NPC objects do not change. The only thing that changes is how they move, how they act, react. Events can be scripted as well, yet I would prefer a more real time situation. The best AI structure I have ever come across is Radiant AI which is used in Skyrim. It is so complex it is as if the NPC/Enemy is alive.

Why OOP is Important
Instead of having "Enemy" as one object, which it is insinuated, you could have a class for every enemy through inheritance. You can further increase it's capacity with encapsulation and polymorphism which are OOP basics.
Enemy Class Examples: Moblin, Goriya, Darknut, etc.
Then you can go further by making even more subclasses for specifics, and once the evolution has finnished, scripting is required.

Normally scripting is only needed for phase-out, which all patch work has ceased for a particular version (servicing/No Development/No Evolution)