Background

I clearly remember my journey into the world of Allegro. Up until 1999, I was using QuickBASIC with a few assembly language routines for key state detection, sprite plotting and a few other things. I'd tried C once but couldn't do anything except a boring console app, but I think I started to hit QuickBASIC's real-mode memory limits and thought it might be worth looking into C again. I found DJGPP and, with that (via the zip picker), Allegro. My first version of Allegro was version 3.11, and it had all my assembly language routines, and much, much more, already implemented. I wasn't sure whether to be happy or sad about this.

Nevertheless, I used it to make some fully-fledged games such as Line Wars and Rock 'n' Spin. I joined the Allegro mailing lists and was able to contribute some fixes and improvements of my own; there was something to do with off-by-one polygon rendering, and something else to do with colour maps, I seem to recall. While looking for somewhere to put my games, I came across the emerging Allegro Games Depot, which soon became Allegro.cc, and posted a rather embarrassing advert for Rock 'n' Spin in the main news column. (I was trying to sell it as Shareware at the time with no knowledge of marketing; no one bought it or probably even found it.) After a while, Matthew put the current forum discussions in the left-hand column, and I met more fun people like Sp0oN and Nicholle.

Thanks to Allegro, I even started writing a Portal-based 3D engine, complete with editor. I had tried Half-Life once on someone else's computer and knew I could create better input controls. My engine went on to be a very popular game. OK, maybe not. Actually I ran into serious performance problems, possibly as a result of my decision to build everything up from tetrahedra and treat every single tetrahedron face as a portal, with no optimisations for adjacent tetrahedra. I could probably do a better job these days, if such a style of engine were still relevant! (Actually I don't even know - maybe it still is?)

I notice Steve++ had quite a similar experience.

The point of this post

I just wanted to say I still occasionally find time to work on a game at home, and I absolutely use Allegro to do it. While Allegro 4 (and perhaps the software-rendered, 256-colour days in general) had a certain charm that I sometimes miss (who remembers the grabber?), I do think Allegro 5 is a good modern-day equivalent, in the sense that it makes it really easy to develop for traditional PC-shaped things without neglecting any minority groups (you know - the ones with X's in their names), at a low enough level that you can really be creative with the implementation and make your game look distinctive. (Case in point: I wonder how many of these quick game-making frameworks people talk about could have led to Braid.)

Allegro is great, and it continues to make progress. It may not be very well known, but they say it's all about marketing. When people come to the forums and see a thread about "Allegro's irrelevance", they think, "Oh, is that what people think? Maybe it's true." But now they'll see a thread about "Allegro's relevance" instead, and they'll think Allegro is worth talking about.

So, whoever said it's all about marketing but didn't post this thread (instead of me), you have my permission to kick yourself.

By the way, the fact that I can write my code in D these days (having now used Java a lot and realised just how much time C and C++ waste) makes Allegro even better.

Go Allegro!

I see what you did there

I think this is one of the fundamental weaknesses of forums as a form of communication. Chatter is what puts threads at the top of the page; and the relative strengths of threads is not taken into account.

When someone posts something controversial, it tends to start a long running argument which stays at the top of the forum for a long time, gaining a lot of publicity. This can happen even if there is only one person supporting one side of the argument against the rest of the forum memebers. That one person's voice is made unproportionally loud.

When someone posts something that most people agree on, it can be held up for a little while by people posting trivial things like "I agree", but ultimately it will gain less air-time than the controversial posts. And when someone posts a complex argument that not many people have the time or skills to respond to - not many people will respond, and so again the thread falls off the top of the forum.

(I suppose reddit is pretty popular because it solves the problem I'm describing - but reddit has different problems instead.)

Grabber and the built-in GUI are two reasons why I still use A4. Though I don't code very much anymore.

Well, I'll also join in with the chorus: Allegro , and certainly Allegro 5 are great libraries. Well fleshed out, clean, and easy to use for people who know their C. I, for one, am very grateful for Allegro.

Furinkan, as I said in the other thread, if plain C and mruby are fine with you, we could cooperate on such an RPG engine. And Chipmunk is ideal for good collision handling and physics for action based games.

You made me laugh >_<
If you want to talk about making an Allegro framework...

Everybody spout his random ideas brainstorm.

Over the next three years, provided I don't suffer an untimely fate and the Vogons don't decide to start building their hyperspace bypass through our solar system, I'll be making and adding FOUR commercial products to the roster of Allegro 5 games.

I did indeed try to build a game engine for Allegro 4 at one point. My thinking was that having a game engine would simplify the process of game making immensely. I got about half-way done when I realized that, even if I finished this engine, it would only cut down development time of my projects slightly.

Instead, what I do now is: When I develop something that's generic enough to benefit more than just the game I'm working on, like a new timing system or an extremely useful mathematical function, I stuff it into its own independent code file which I can then copy and paste into any project I work on. At present, I have five such cpp/h file pairs:

g-rand.cpp/h -> My own pseudo-random number generator
g-time.cpp/h -> My own timing routines including interpolation calcuations
g-gui3.cpp/h -> The third revision of my very basic (but functional) GUI
g-frag.cpp/h -> Fragment shaders
g-misc.cpp/h -> Everything else like math functions, string functions, etc.

I also had a g-idle2.cpp/h pair for awhile for doing accurate idling in Allegro 4, but since Allegro 5 doesn't have to worry about busy waits it's been rendered obsolete.

Women. Guns.

Wow, your experience sounds almost identical to mine. I was using QuickBASIC and started to run into it's limits. Then I found DJGPP and a book called "QBASIC TO C" which helped me convert over (even though the book got bad reviews, it helped me immensely). I started to write my own library for C with various VGA graphics functions etc until I found Allegro (I think Allegro 2). Like you, I was sort of saddened, because my little library was starting to take shape. But that went away fast when I started programming with Allegro.

Looking back at where Allegro started and now with some of the stuff we can do with Allegro 5, it's come a long ways thanks to the dedication of some gifted programmers.

I still program in C. Specifically, C99, although I recently learned about the C 2011 standard and may start using it. I am not a fan of C++, I think it's a lot of bloat.

P.S: I still love programming using Allegro 5 for the PC. I have had a friend bugging me to make games for Android for $$$. I enjoy programming for the PC, using Allegro. I am working on something right now with it.

The only thing that has been slowing me down with development is age I think. My enthusiasm isn't the same it used to be + various other life issues.

I'd personally turn it into a static library. And share the library between projects, so you don't have to maintain several different copies.

This thread is sure to become quite the lulzbucket here soon.

This forum is oddly like an internet coffee shop! But that's part of its charm, there isn't thousands of squabbling masses and we don't have to divide the forum into sections and assign moderators to keep the sheeple in check. Its a small, friendly community, which is why I've just hung around for the past 4-5 years. I even miss people when I don't see their posts for a while! (*cough* Mark O, Amarillion, verthex, type568, piccolo, et al. cough)

@beoran:
You'd have to give me a good reason to program in C. I don't think I'm bamf enough to program in straight C. I haven't done any coding at all on mine, I've just been white-boarding a bunch of algorithms, planning file format requirements, working out the layering system, researching libraries (which you have helped immensely with in the other thread), yada yada.

Furinkan: If you're not "bamf enough" for C, then I fear C++, which is far more complex, will be a disaster. And there you have one good reason to use C. Apart from the fact that you have to be a bit more explicit in C, and do a bit more manual work, it's a much simpler and straightforward language. Single or multiplayer isn't that much different I think, the engine I have could become useful to other people too as it develops.

@beoran:
What I mean is, I'm used to classes/Objects in higher languages like PHP/JS. I like the organization that a little OOP can give, and I like the terseness. I'm alright in C++. I'm just curious as to why I should only use a subset of C++ (that is: C), when the compiler eats most of the overhead anyway.