Irrlicht 2.0 - What's in store for the future!

[Nadro]

Yes, after a public release I will implement it in the shader-pipeline branch.

[Granyte]

Will we support the multi threaded rendering path on these drivers?

Because it kinda is a big part of dx12/vulkan

[Nadro]

In future definitely yes, because without it we'll lose too many benefits from Vulkan

[Granyte]

Alright I hope too ill naturaly devlop a dx12 driver to go with it and ill also add multithreaded rendering for dx11

[kormoran]

SDL2 - sure would be good. Simply no one volunteered to write it so far

ME! ME! My project uses SDL, so I gave a look to CIrrDeviceSDL.cpp and saw that can be ported to SDL2 rather easily.

So, as a collateral, I skittered down that path: at the moment I have converted everything but joystick discovery and textinput, as they are managed so different in SDL2 that some decision is needed about implementation (a minor change is needed to COpenGLDriver.cpp too, done already).
Right now the new IrrDeviceSDL compiles statically, but I haven't understood how to use the SDL device in examples. In short, I need help to complete the task...

P.S: irrext-devs@lists.sourceforge.org is broken... is there another address?

[CuteAlien]

Agh right, SDL2 was another thing we wanted for Irrlicht 1.9.
@kormoran: Someone in the forum also worked on this: http://irrlicht.sourceforge.net/forum/v ... =6&t=49877

Basically - I also want SDL2 (older versions are not even supported anymore), but it's easier for us to merge that after we have merged the ogl-es branch. Otherwise we have twice the work.

[kormoran]

@kormoran: Someone in the forum also worked on this: http://irrlicht.sourceforge.net/forum/v ... =6&t=49877

oh, ok, work already done

well, it was a good primer into irrlicht nonetheless.
However, I have a patch I made to get rid of some annoying compile time warnings, just need to know how to submit it.

[CuteAlien]

Always a good idea to check forum - lot's of patches floating around here.

You can submit patches either by posting them somewhere and mentioning it in the bugtracker forum. Or you can post them to the patches-tracker: https://sourceforge.net/p/irrlicht/patches/
Thought... be careful about removing warnings. Nearly all warnings which are still in Irrlicht are staying because they are real. Aka - stuff that the compiler correctly warns about. We try to get rid of compile warnings which would show up in user-code (still have some work there for 64-bit and file-functions, but rather tricky doing that correct, I tried it already once and gave up after a few hours...). But compiler warnings inside Irrlicht are for the most part stuff that really is open (like all the missing colors in switches or the reminders that certain loaders don't use yet all the information they have). In other words - nearly all warnings you see will not be trivial to remove.

[GameDude]

Will the next version of Irrlicht have support for the Vulkan API?

https://www.khronos.org/registry/vulkan/

[CuteAlien]

Next version (aka 1.9) won't have it. We only try to finish that one up somehow. After that for 2.0 - let's say it would be nice, but will probably take a long time again.

[Granyte]

Some of us still hope 2.0 gets DX11 so ... yeah vulkan not until irrlicht 2.1

[GameDude]

Ah I see.

[Nadro]

Vulkan support will appear in Irrlicht v2.0, however before that we need to add multithreading to IVideoDriver -> switch to C++11 (std::thread).

[Granyte]

So dx 11 will have to wait until vulkan is done to?

[Nadro]

No, I just think that after v1.9 we'll be able to prepare relative quick v1.10 release based on shader-pipeline, however we must solve following issues before that:
- optimize FVF stuff
- fix D3D11 stuff
- Windows Modern Apps support would be nice
After that we would focus at v2.0