solenum / exengine Goto Github PK

I see you code at line 98 of iqm.c

the ex_frame_t_pose = NULL, not initialize before using in function ex_model_update_matrices .

above it is the first question.
the other one is skeleton may be the joint coordinate system of each bone? right?

Unfortunately all I'm getting is a black screen, it works fine in forward rendering mode and the imgui stuff still renders, not sure why this is happening so figured I'd let you know.

Would it be possible to have more examples whose code we can view (like that game with a ball on YouTube that you made, or the examples in your website)? What about documentation?

Right now, this engine seems to be a great idea but if you don't entirely know how to use it it's a bit pointless...

Looking forward to it

https://github.com/exezin/exengine/blob/7edbadecf833822138e7f4b2e20346b8c46e5728/src/exengine/render/shader.c#L4
https://github.com/exezin/exengine/blob/7edbadecf833822138e7f4b2e20346b8c46e5728/src/exengine/render/shader.c#L5

These two lines generate a compile warning

warning: missing braces around initializer [-Wmissing-braces]

I see that you check if the index of the values in the following lines:
https://github.com/exezin/exengine/blob/7edbadecf833822138e7f4b2e20346b8c46e5728/src/exengine/render/shader.c#L21
https://github.com/exezin/exengine/blob/7edbadecf833822138e7f4b2e20346b8c46e5728/src/exengine/render/shader.c#L26

Therefore I don't know if this will break the logic you have here. But I propose the following solution (it seemed equal on my system but maybe I am missing the change)

GLint ex_uniform_map[256][256] = {{0}};
GLint ex_uniform_locations[256][256] = {{0}};

I'm submitting this as an issue so you can double check if the shader logic is still intact with this change.

Curious if dirlight and ssao that were moved to src/exengine/render/legacy are coming back? Seems from the commit history a refactoring effort might have stalled, but hopefully not forever...

Excellent project and code quality, btw. Finding it very nice to read and use, so thanks for all the effort :)

On a fresh build approximately 10 minutes before submitting this, the game executable fails to load data/conf.ini and data/shaders/text.glsl

This is on a relatively recent (last updated ~48 hours ago) build of Arch with GCC version 11.2.0 and SDL version 2.0.20-2. On the off chance this matters, the CPU is a Ryzen 5 2500u and the program was launched using the internal GPU for rendering

If I missed any relevant information, do let me know.

Pros:
single header lib in C
Cons:
lacks some features (?)

sudo apt-get install libopenal-dev

Hey exezin, I used to join your IRC channel in past but nowadays nearly every gamer is on Discord. Did you ever consider to open a Discord server for your engine? Discord makes it way easier to share code snippets with Markdown and syntax highlighting, share images via drag&drop etc.

Could ExEngine be back-ported to OpenGL ES 2.0 / WebGL (v1) to support older browsers and devices? If so, would it involve re-writing the shaders, or are there specific OpenGL ES 3.0 / WebGL2 calls being made by the engine?

I see your some of functions, the return type defined bool, however you alway use 0 or 1 as the return value.

maybe you should include #include <stdbool.h> , instead of 0 or 1 with true or false.

above all, just my personal advice.

last but not the least, I like you project.

A command prompt window opens, and then becomes unresponsive immediately.

These are (stupidly) not cached currently, causing varied performance on certain hardware. Especially the web port.

file: scene.c line:36, 37
memset(s->coll_tree->region.min, 0, sizeof(vec3));
memset(s->coll_tree->region.max, 0, sizeof(vec3));

file octree.c line:15, 16
memset(o->region.min, 0.0f, sizeof(vec3));
memset(o->region.max, 1.0f, sizeof(vec3));

I think default we can set to 0.0 both sides.

Where can I find and modify player properties?

Hi, I recently try to load sponza scene as my test scene.

The scene file consists of two types of media, one is obj mesh file, the other one are images.

however, right now, the iqm tool chain not parser the mtl or usemtl to recognize material texture as the resource id later on.

so in my eyes, I should add some features that support mtl parser for obj format files. because only under this way, we probably gain texture id use you iqm loader.