[java] Would Anyone be intersted
Author
Hey I know that java works in reverse byte order and it might be hard for some of you to load a 3D model for use in your game, I am currently developing a Converter for well converting a md2 file into a Java readable PFC (My format) file, If anyone would be intersted reply to my post and Ill post the converter and how to read the data as soon as Iam done, Iam 85% complete right now and by tonight at latest everything should be done. So if you are intersted please reply if not I wont bother posting it.
I don't want to discourage you, but the jME team has done a lot of progress with this, loading models from different formats to a XML description file, or the file format used by JME itself. I am mentioning this because I think you might be interested by their work.
BTW, md2 models are supported by them.
Son Of Cain
BTW, md2 models are supported by them.
Son Of Cain
I wanted to give you a hint, so that you could certainly support more formats, and increase interest in your project.
I have a question, though: how much independent of an engine's code can you be? I mean, if you load animation information, can you do it at such an abstract level that I might be able to use your information within my engine of choice?
Son of Cain
I have a question, though: how much independent of an engine's code can you be? I mean, if you load animation information, can you do it at such an abstract level that I might be able to use your information within my engine of choice?
Son of Cain
Yes, an "engine independent model loading framework" would be nice... the way Java3D handles this kinda sucks, and now that we have those lovely OpenGL bindings...
Basically, this piece of code would allow standard loading routines for popular file formats, outputting some ByteBuffers or a XML file, or whatever you please... I'm not an experienced developer myself, but I'm sure people here could help you gather the requirements for this.
That would not be reiventing the wheel, I think - because you'll allow people writting games in their own "engines" to use 3D models in an easy way.
Just my two cents, anyway...
Son Of Cain
Basically, this piece of code would allow standard loading routines for popular file formats, outputting some ByteBuffers or a XML file, or whatever you please... I'm not an experienced developer myself, but I'm sure people here could help you gather the requirements for this.
That would not be reiventing the wheel, I think - because you'll allow people writting games in their own "engines" to use 3D models in an easy way.
Just my two cents, anyway...
Son Of Cain
When you create a loader, you are making 2 things ; decode the file and convert it to your engine. Decoding the file is very very easy for all formats i know. The difficult part is converting it to your engine.
Therefore, I think an engine neutral loader would be just useless.
Therefore, I think an engine neutral loader would be just useless.
Quote:Original post by Niwak
When you create a loader, you are making 2 things ; decode the file and convert it to your engine. Decoding the file is very very easy for all formats i know. The difficult part is converting it to your engine.
Therefore, I think an engine neutral loader would be just useless.
That's because most people think of an internal format then try to find a good model format to use that would be easy to convert. If you create a universal loader with proper interfaces for people to implement, they would be more likly to use it and have their format more closely aligned to the model libraries format. Besides interfaces allow the flexibility for you to have your own format, but use another persons library.
I was a bit quick to say that it would be useless. It won't ;
- it will create a knowledge base for implementing our own loaders,
- it will provide a usefull specification for creating extendable tools,
- it will give a good starting point for new game library or simple 3d applications,
- ...
I was answering from the point of view my project is giving me (a java 3d engine). For me, loading models is a simple task. Interfacing would cost me ;
- the overload of a communication between the loader and my engine (data transfering / duplicating / not using the perfectly fitted loading method),
- the overload of creating an adapter layer that would sit between the loaders library and my engine,
- the burden of learning a new API.
and all this just to load simple files that are directly loaded in just a few lines of code. Therefore in the process of creating my game engine, I don't think it would be interesting.
I think the difficult part for the type of library you are describing is designing a set of interface that are efficient (problem of data transfer, file channels,...), complete (there is a huge load of different way of defining shading of geometry across formats) and easy to implements.
Anyway, I hope that your attempt will succeed and that it will proove me wrong.
- it will create a knowledge base for implementing our own loaders,
- it will provide a usefull specification for creating extendable tools,
- it will give a good starting point for new game library or simple 3d applications,
- ...
I was answering from the point of view my project is giving me (a java 3d engine). For me, loading models is a simple task. Interfacing would cost me ;
- the overload of a communication between the loader and my engine (data transfering / duplicating / not using the perfectly fitted loading method),
- the overload of creating an adapter layer that would sit between the loaders library and my engine,
- the burden of learning a new API.
and all this just to load simple files that are directly loaded in just a few lines of code. Therefore in the process of creating my game engine, I don't think it would be interesting.
I think the difficult part for the type of library you are describing is designing a set of interface that are efficient (problem of data transfer, file channels,...), complete (there is a huge load of different way of defining shading of geometry across formats) and easy to implements.
Anyway, I hope that your attempt will succeed and that it will proove me wrong.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
