What is the target market of Irrlicht?

[dlangdev]

Just wondering what is the target market of Irrlicht? In terms of:

1) Hardware
2) Portability
3) Latest-and-greatest stuff.
4) User base (game users)
5) Developer base (programmers, artists, modelers).

Let's discuss.

Thanks.

[rogerborg]

Purely IMHO - this is not a policy statement.

1) Hardware

The hardware that most gamers actually have: last generation gamer cards, budget cards, laptop chipsets, on-board chipsets.


2) Portability

No strong feelings beyond Windows, Linux and MacOS. Other platforms are nice but niche. Porting to most platforms (exceptions: PalmOS/GarnetOS, Symbian, BREW) isn't a big deal, if you only have to do it once, since the interfaces with the OS and drivers are well defined.


3) Latest-and-greatest stuff.

Doesn't particularly interest me, since the resources required to do it justice make it prohibitively expensive to produce more than tech demos. That said, Irrlicht has come on a great deal recently, so the gap between it and bleeding edge engines isn't huge.


4) User base (game users)

Any gamers in any genre with the appropriate hardware; Irrlicht is general enough to be used as the 3d engine for any type of game.


5) Developer base (programmers, artists, modelers).

Aimed at developers who enjoy ease of use, rather than people who want to optimise every last stage of the pipeline. There are other engines for them. I can't speak to artists and modellers, but I imagine that being able to load models in the native format from a range of modelling tools is a plus, compared to the workflow-disrupting step of having to convert to a specific engine format first.

Again, this is strictly my personal view of Irrlicht. I'm not speaking for any of the other devs here, and I'd actually hope that they have different views. Allah loves wondrous diversity.

[hybrid]

It's a hardware-independent, portable, nicely looking, beloved, and working thing. For me.

[CuteAlien]

2) Portability

No strong feelings beyond Windows, Linux and MacOS. Other platforms are nice but niche. Porting to most platforms (exceptions: PalmOS/GarnetOS, Symbian, BREW) isn't a big deal, if you only have to do it once, since the interfaces with the OS and drivers are well defined.

There have been several project lately working on ports to OpenGL ES for the iphone and similar platforms. I don't know how much choice for 3d engines people on those platforms have so far and it's also not so much in my interest currently. Still I wonder if that could maybe be a rather big deal for Irrlicht.

[rogerborg]

Fair enough; OpenGL ES is a significant target. I should be more excited about it, but I work on mobiles in my day job, so it's a bit of a busman's holiday.

[dlangdev]

This is great, guys. All these comments will definitely influence how my games will be written later, as they will be more likely be going along the path where developers wish to go.

So yeah, please keep those comments coming. This is not the end of the conversation.

[jmpep]

Fair enough; OpenGL ES is a significant target. I should be more excited about it, but I work on mobiles in my day job, so it's a bit of a busman's holiday.

Well, actually I have Irrlicht 1.4.1 running on my Nokia N95 8Gb.

I was going to show a video, but I lack a proper camera to record the mobile screen with a decent quality. Just to give some figure, the Quake 3 map is shown at 19 fps.

I have done the port as my Final Project for my bachelor's degree, so I cannot release my code yet -I'll wait until I defend it to the University committee- but I will within a month or even less.

[hybrid]

The ogl-es branch is also usable, now. Includes also some of your suggestions already. Just the color scheme problem hasn't been addressed so far (as we're targeting for iPhone first )

[jmpep]

Yeah, the color format problem is quite tedious to solve. Anyway as I have already done it I think that you would be able to just use my code -which I think I already sent you, but I am not sure-.

Anyway, wouldn't it be necessary a iPhone OS device to be able to run Irrlicht on the iPhone? Apart from the OGLES 1.x support, I mean. For S60 I have had to create a new device.

[jontan6]

we're using it for visualization of property. so we're quite happy with compose scene with irredit and render in irredit with the built in collision

the many file format support is great! irredit is great! and irrlicht is the best

[Mirror]

Purely IMHO - this is not a policy statement.
I can't speak to artists and modellers, but I imagine that being able to load models in the native format from a range of modelling tools is a plus, compared to the workflow-disrupting step of having to convert to a specific engine format first.

i would be in favor of such a practice. having a binary optimized irrlicht mesh format and a tool that converts every other format to that, and irrlicht supporting only it's own format. that would help in many ways and standardize things.

...one people...
...one reich...
...one fuhrer....










...one mesh format...

[dlangdev]

I get the feeling alot of posters are making games for the handheld device. Is it somewhat true this observation does affect to where Irrlicht development is going?

What about other developers who aredeveloping on Vista PC platform, targetting high-end rendering? What is the current opinion on this?

[rogerborg]

Purely IMHO - this is not a policy statement.
I can't speak to artists and modellers, but I imagine that being able to load models in the native format from a range of modelling tools is a plus, compared to the workflow-disrupting step of having to convert to a specific engine format first.

i would be in favor of such a practice. having a binary optimized irrlicht mesh format and a tool that converts every other format to that, and irrlicht supporting only it's own format. that would help in many ways and standardize things.

...one people...
...one reich...
...one fuhrer....
...one mesh format...

ein Ring um sie alle!

You know, I can see a pragmatic case for focussing Irrlicht support on the Ogre format, at least informally. On the other hand, support for native formats is one of Irrlicht's strengths.

Wait... I can see the issue from both sides. Am I... am I turning into some kind of liberal? GODDAMN OBAMA!

[Mirror]

Purely IMHO - this is not a policy statement.
I can't speak to artists and modellers, but I imagine that being able to load models in the native format from a range of modelling tools is a plus, compared to the workflow-disrupting step of having to convert to a specific engine format first.

i would be in favor of such a practice. having a binary optimized irrlicht mesh format and a tool that converts every other format to that, and irrlicht supporting only it's own format. that would help in many ways and standardize things.

...one people...
...one reich...
...one fuhrer....
...one mesh format...

ein Ring um sie alle!

You know, I can see a pragmatic case for focussing Irrlicht support on the Ogre format, at least informally. On the other hand, support for native formats is one of Irrlicht's strengths.

Wait... I can see the issue from both sides. Am I... am I turning into some kind of liberal? GODDAMN OBAMA!

well since you like the idea of pluralism ( as in lots of mesh formats ) i'd say you're already THE liberal. change IS here.
p.s. did you notice that obama from osama, is just a letter of diff ?

[hybrid]

Ogre's binary format is not documented anywhere (besides by the Ogre code itself). Moreover, a native format should be tailored for exactly that engine. Anyway, ogre format support was significantly enhanced today