I am wondering how many people would be interested in creating a game engine project using the Irrlicht.
The engine should:
OPEN SOURCE
Be very moddable preferably with existing tools (Radiant, IrrEdit etc...)
Be not only documented code but also PDF's to show people how to use the egine without knowing the code.
Have server/client setup
Have a console
Some thoughts:
Perhaps its best not to change any IrrLicht code so that the engine is upgradable to a new IrrLicht version
My main concern would be a FPS but I see a lot of other genres in this forum wich might be able to co-exist
At this moment I'm putting a lot of my time into learning how the IrrLicht engine works so this would be a great learning proces for me. But offcourse not only for me.
Idd like to know if there are more people that either have the same idea or would be interested in doing a project like this. I have searched these forums a few times if there were any similar projects but I couldnt really find anything interesting. And I hope that the core of the IrrLicht team would give a blessing if there are enough people interested
I guess they wait for the first bashing round. After that one can show some interest without being bashed, too
But honestly, this post is far too short and lacks the most important details or even concepts. So what would one be interested in - doing just something with Irrlicht is the stuff we are all already concerned with.
hybrid wrote:I guess they wait for the first bashing round. After that one can show some interest without being bashed, too
Well, let's get it out of the way then.
The big question is why a potential user would choose to use your engine rather than just using (for example) Source SDK?
To catch up to where Source SDK is already, you're talking a minimum 10 man years (~20,000 hours) of work. And that's work of the quality that would be produced by Valve employees, not just any J. Random H4X0rZ.
And the problem with using J. Random H4X0rZ is that he'll get bored with engine development, and at some point will drift off to create a game with it instead, leaving your engine to stagnate. I've seen that happen among paid developers, let along volunteers.
It'll be a fun and valuable learning experience for you to try this, but mostly what you'll learn is how to identify what's practical and what's not.
I think enough people (most likely not IrrLicht programmers like us ) would be interested in using an open source game engine for their game. Perhaps even some of us just want to make a game and dont want to go through all the steps of making a game engine on top of a Graphics Rendering Engine.
ATM its like a 100 people going over the same proces to create a game. Perhaps working together would save time and their games might even run more efficient.
And I dont have the illusion to be able to compete with engines like HL2
xhrit is working on a game engine at the moment, it's based on his game Zero Point 3D, an almost-complete cross-platform space shooter. It uses Irrlicht for 3D, ODE for physics, LUA for scripting, (iirc) OpenAL for audio, OpenSteer for pathfinding, not sure about network. The license is zlib style like Irrlicht, if you include ZP3 it's been under development for uh.. I dunno 2 years, longer perhaps?
I'm sure he's still looking for early adopters- users, tool writers, co-authors, etc. Try sending him a PM or hunt him down on IRC (#irrlicht on freenode.net)
Hi, I think right now. We should stay focused to have a good graphic engine.
A game engine would be good. But if all IRRLicht developpers would go in that direction, we'll loose focus and the engine would not get updated as fast. Unless you can find lots and lots of dedicated programmer willing to give their time to create that engine.
The good thing, it that some users have worked hard to make some wrapper between the graphical engine and other types of engine (sound, network, physic, AI). That it very good. We also have "basic" thing for gravity and collision.
The only things that could be missing: (Some ideas I've had)
Grouped tutorials (sound, network, physics, and AI at the same place). I think, this as started a little in the WIKI. Also, something that could be interesting is that some Irrlicht users that got a full game from Irrlicht could be interviewed and gave us some of the method they used to create their game (engines they used and how)
We should have more examples. It would be great that we could have some more examples (For me, I'll need more examples to use shaders, AI). Could have folders like, beginner, intermediate, advanced...
Also a basic AI for characters could be good (like the basic collision/gravity we have). This seem to be the hardest part to do. If we would need more, we could try to implement a wrapper to an AI engine.
I know that the effort was made on the animation system that needed to be reworked. We would have to wait.
Also all new graphical engines depends a lot of shader technology. If we could do something to ease the use of them for newbies (pre-made shader library that can be easily re-used and expanded?)
What I'd like the most on using shader is "blend" some layer with different material attributes in the same "material" definition. (Like using a Solid layer, with a Alpha map layer, a Lightmap layer, bump map/Normal map layer, etc.). If this could be defined easily in the engine that would be great.
Right now, I'm able to create a complete level in IrrEdit and walk thru it. But having ennemies there, right now for me look like impossible. Having some simple pre-made things done could help newbies like me.
If you look at most of the demo including characters that were made in IRRlicht, thats the area that lack the most. (AI)
One advantage would be that the bigger the project the more interesting it would be to request new features. Maybe implement them themselves if needed. It takes people away from doing everything individually.
100 people x implementing gravity instead of maybe a few times (just to use an example)
Well, I was thinking about doing some engine like stuff (mostly adding IrrNewt, IrrKlang and a Entity system working with callbacks...) but well, even thats to much work for me so i dropped it... I would say that a good idea would be to develop small extensions to irrlicht, that might come handy some time.. (Entity class etc...)
I was thinking about doing some Rendermonkey like shader editor, that would be very easy to use (template projects, C++ export etc...)
I was making my own irrlicht game framework called IrrGameShell some years ago, using: Irrlicht, Newton, Lua, Audiere, Dusty Engine..., but i think is a good idea to make a kind of "oficial" or "most used" irrlicht game framework distribution.
Obeleh wrote:
One advantage would be that the bigger the project the more interesting it would be to request new features...
...
I think enough people (most likely not IrrLicht programmers like us ) would be interested in using an open source game engine for their game. Perhaps even some of us just want to make a game and dont want to go through all the steps of making a game engine on top of a Graphics Rendering Engine.
ATM its like a 100 people going over the same proces to create a game. Perhaps working together would save time and their games might even run more efficient.
Yes, adding some extensions make sense. (Like the Klasker tree node and the Realistic water node by Elvman)
But, we'd also need some tutorials and examples to use theses extension. Theses examples should be simple and minimal. Basicaly show exactly what the extension is doing. Like the Irrlicht examples.
Thegamemaker:
I was thinking about doing some Rendermonkey like shader editor, that would be very easy to use (template projects, C++ export etc...)
Can you give me more details about this? What is Rendermonkey?
This is a neat idea. But don't most people using the Irrlicht engine make a game engine for themselves? While using the Irrlicht engine as the graphics engine?
I was thinking about doing some Rendermonkey like shader editor, that would be very easy to use (template projects, C++ export etc...)
This will be great as a plugin for irrEdit, it would be a waste to have it as a standalone app when irrEdit already has most of the needed functionality (and more).
ShadowMapping for Irrlicht!: Get it here
Need help? Come on the IRC!: #irrlicht on irc://irc.freenode.net
I think having an official unofficial Irrlicht game engine would be such a wonderful idea!
Why? These days, Irrlicht is mostly used by hobbyists to create small (and sometimes large) games. To create an actual 'game' engine, and not just a rendering engine, would save many developers the time of integrating and bug-fixing all of the stuff related to a game, including networking, physics, sound, input, etc.
And to have a 'official' game engine side-project would only unify all of the people using Irrlicht in games. With more people comes better quality, almost always.
