BOUNTY for AmiDARK Engine source code release.

[Hans_]

what concerns linking against mesa rather than minigl, yes it should be possible, just it woulnt be just linking against other header i guess. remember also os4 minigl has been extended by hans, while mesa was not and certainly amiga(68k) minigl contains only a subset of both.

StormMESA should implement the full OpenGL 1.2 specification (or whatever the version was when it was created. The additions to MiniGL were for stuff that MiniGL missed.

i dont discuss that basic style coding may be easier for noobs (like me) though the sole availability of this option doesnt effectively make people program anything, like for instance amiblitz example teaches us. although it seems a good option i can count programs done in amiblitz on fingers of one hand.

AmiDARK's advantage is much more than BASIC style coding; it's a game engine. So, you don't have to deal with all of the low-level stuff like loading & rendering objects, etc. If you use OpenGL directly, then you would have to write all of that stuff yourself. That's a lot of work.

As to getting people to use it, the following would help: providing good tutorials, documentation, example code and a large library of free to use models, sounds, etc. There are no guarantees that many people will use it, but I think that it provides a good starting point for beginners. It makes it easy to create good looking & sounding results quickly, which can be fun and motivating.

now, while you would link amidark against storm mesa and wait for people to write software with it there are still numerous opengl opensource stuff that might have been ported over and simply linked directly against mesa headers, been there, done that. im just too dumb and the hardware s too limited to make it work really well.

MiniGL was originally created on the 68K, because StormMESA was on the slow side. From what I heard, it was significantly faster. If MiniGL 1.2 provides all needed features, then it might be better to use that over StormMESA.

Hans

[Hans_]

yes, thats what ive been told all along. few years ago i modified the gl examples to be compil and run-able against minigl and mesa side by side. a simple measure, because minigl inits differently to the standard. blame it on my inability, but funny enough it occurred minigl wasnt much if any faster than mesa. btw, im happy to be proven wrong.

Well, performance differences will depend on what you're rendering (and how). You might not notice much difference with simple examples like the gears demo, but a game with a full immersive 3D environment... Even so, any performance boost would depend on how the game rendered its graphics.

EDIT: IIRC, using manual locking in MiniGL can boost performance. This means that Warp3D is manually locked and unlocked, so that you minimise the number of times that this happens. Locking/unlocking Warp3D is rather expensive.

also the subset of features minigl provided in comparison with mesa didnt allow to compile anything i can think of. im pretty sure if sm3.1 was ogl1.2 spec then minigl was a rather narrow fraction of it. may be different if amidark can live without all that stuff so far, still im not sure if there is anyone left in the so called amiga scene, who would program 3d games or apps exclusively no matter what easy it was.

AFAIK, MiniGL's subset of the OGL spec was whatever Hyperion's game ports (Quake II, Freespace, etc.) needed.

Hans

[Hans_]

..so far seems my points still stand. minigl may be faster (marginally or not, we must rely on word of its programmers that happened to be proven wrong in the past, perhaps except of you, hans;)) but its not open and therefore it isnt extendable if need be. i think it rules it out for amiga (68k). neither way gunnar would need his w3d driver

MiniGL is open. The source code is available for both the OS4 version and the old 68K one. If you want the 68K version's source-code, then just download minigl.lha from aminet.

Regardless, yes, 3D drivers are needed. In theory you could bypass Warp3D, and build your HW driver directly into a modified version of MiniGL (so long as you shared the code as per the license).

Hans

[Hans_]

I guess its worth noting Hans's comments about the merits of this engine, and how simple it would make game creation, demo making etc. Im sure developers could find all kinds of use for the engine as they familiarise themselves with it.

Link to the mainpage of Amidark;

http://www.amidark-engine.com/

Link to a codesnippet to illustrate the power of its design;

http://www.amidark-engine.com/spip.php?article10

If you read Hans's comment, what requires 1000+ lines during normal developent, can be done with 2-3 lines in Amidark.

One point that I don't think I got across properly, is that a game engine like AmiDARK would give those who would like to create a game but are lacking in coding skills a chance to create their game. It lowers the barriers to entry.

Hans

[Hans_]

Don't be kidding. Those who are lacking the coding skills won't become a game designer or just a software engineer no matter if they chose Basic, Amos, Logo or any other simplified and easy to learn language.

Utter total BS. Nobody is born with coding skills. There was a time when I was lacking in coding skills too, but I've become very good at it.

I started using BASIC on the C64, but quickly wanted something better (performance sucked). Back then, I could write simple programs and would have liked to create games. However, I lacked the ability to do the bare essentials like load and display images, move sprites, play sound effects and music, etc. A "low end game engine," As you put it, would have been exactly what I needed to get started. It would have helped me a lot.

I remember a friend had AMOS, and with a few lines of code we could create stuff way more impressive than anything that I'd ever done before. That was awesome! It lowered the barrier to entry for creating games/demos. No, we never released anything, but it certainly helped me to learn how to write software. It was also fun.

What something like AmiDARK, Hollywood and AMOS provides is an easy way to get impressive results quickly. That helps get people started, and also helps boost the motivation needed to keep on working, learning, and improving. No, not everyone will go on to become a game designer, but it will provide those who are genuinely interested with a good starting point. Plus, creating stuff is fun, even if the end result isn't always release grade.

The dreaded attempts to create something usefull with Amos or SEUCK or any other low end game engine (adventure construction set) are history now.

Of course the first attempts that anyone makes at a new task are going to "suck." However, with practise people get better. The goal for a rookie is to learn and enjoy the process of getting better. You have no idea how many of the people who made "dreaded attempts with low end game engines" now work in the gaming industry.

Of course, if you tell people how much you hate their "dreaded attempts," then they're much less likely to try again and make something better. Way to go, Grinch!

AmiDark is proprietary, uncompareable and incompatible to all other. How to port C code to or back if you need cross-platform developement?

The bounty is to open-source it. It closely mirrors the DarkGDK API. You would have known this if you had bothered to look into it before spouting your opinion.

Hans