As @TimvanScherpenzeel pointed out, cmgen should better explain that --size applies to the DFG LUT as well and that a cubemap and the LUT cannot be generated at the same time.

我发现examples/retinaface.cpp中，如果开启OMP加速的话似乎在检测到人脸时会发生内存泄漏，但我定位不了这个问题的具体原因。

值得注意的时，如果将qsort_descent_inplace函数中的OMP指令注释掉这个问题就会消失掉。

static void qsort_descent_inplace(std::vector<FaceObject>& faceobjects, int left, int right)
{
    int i = left;
    int j = right;
    float p = faceobjects[(left + right) / 2].prob;
    ...
    // #pragma omp parallel sections
    {
        // #pragma

It was happening in my own code, but it also 100% reproduces for me in the examples. Simply build the examples, run and hit the window close button.

Seems to be something relating to it still being inside renderFrame() during bgfx::shutdown().

<img width="1582" alt="Screenshot 2019-12-25 at 14 38 40" src="https://user-images.githubusercontent.com/621629/71447311-a7752680-2724-11ea-8a29-c2eae

The documentation for glfwSetKeyCallback at https://www.glfw.org/docs/latest/input_guide.html#input_key offers little guidance on what will cause a GLFW_REPEAT action, which can cause confusion such as in the forum post Key Callback Not Registering Every Key Press
.

I think we should consider modifyi

Great tutorial! It clears up a lot of my questions about various effects.

Minor complaint: Focal length is size of the lens. Focal distance is the distance between the camera and the focal point.

As documented in #5392 using the packages for tflite and tensorflow 2.1.0 the test as in the subject line segfaults. It has now been skipped in the testsuite but this needs to be fixed.

	dependencies[0].srcSubpass = VK_SUBPASS_EXTERNAL;
	dependencies[0].dstSubpass = 0;
	dependencies[0].srcStageMask = VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT;
	dependencies[0].dstStageMask = VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT;
	dependencies[0].srcAccessMask = VK_ACCESS_MEMORY_READ_BIT;
	dependencies[0].dstAccessMask = VK_ACCESS_COLOR_ATTACHMENT_READ_BIT | VK_ACCESS_COLOR_ATTACHME

We use PIX labels to organize the events in a single frame (e.g. Z PrePass, GBuffer, Alpha, etc). These can get quite deep/nested.

It would be useful if we could see a 'path' for the event currently selected, e.g.

Camera 1\3D Stage\GBuffer\TreeBark Material

It would be extra nice if clicking on any of these would take you to the beginning of that section.
(think folders in Windows Explor

• Operating System
• Config Options

已经按照https://www.yuque.com/mnn/cn/cvrt_windows的步骤 ，安装好了protobuf。
但是在进行编译模型转换工具步骤时报错如下：

步骤为：

cd tools/converter
mkdir build
cd build
cmake -G "Ninja" -DMNN_BUILD_SHARED_LIBS=OFF -DCMAKE_BUILD_TYPE=Release ..

想知道问题是什么，感谢~

It looks like it's only used for one error variant and a bit in the backends. It's not really needed in the API itself.

Currently, the glDrawRangeElements() WebGL 2 function is not implemented in Firefox (38) and causes "Not Implemented." assertion. It is also not yet implemented in Emscripten, but I have a patch that adds it.

Rela

As of 00e1f52cf95c16b0cea34c83372145532b4745ba the LWJGL codebase has a new directory layout. This has broken the included Eclipse project.

Release Type: Github
Version: master

Describe the bug
The documentation states "The index of the component to access. Use 0 for the alpha component, 1 for the red component, 2 for the green component, and 3 for the blue component".

The actual implementation uses:

switch (index)
{
	case 0: return R;
	case 1: return G;
	case 2: return B;
	case 3: return A;
}

It is unclear from the provided documentation what the difference is between a chunk and a subbuffer. Without examples it can be very difficult to identify when using one instead of the other may be necessary. We spent multiple days tracking down why an attempt to pass a subbuffer of vertices to a call to draw() was failing to compile. The error message was extremely verbose, but ultimately see

The Forge sounds pretty great but there does not seem to be much in the way of documentation.

I don't even see simple "How to compile" or build instructions. Which seems like a good place to start.

The description talks about it being like "Legos" where you can use the different parts of the system as needed. So does that mean it can be built in sections and integrated into other projects pi

The Khronos reference page contains description of parameters and description of the behavior for every function. Wouldn't it be possible to add doc comments for the functions, so IDE can show it? I understand the function description is a bit cumbersome, as one description is usually related to multiple functions with similar functionality, however parameters description would help a lot.
I am c

Also optionally, if we make backwards incompatible changes to the config file we should notify users somehow. (E.g. changing defaults).

EDITED SUMMARY
Make CMake complain if user is only using precompiled dependencies, but is compiling in 32-bit mode.

ORIGINAL ISSUE
I don't think BSF is intended to be used as a 32 bit library, so probably the easiest way to resolve this user confusion is to make cmake complain loudly if it's being configured to build a 32 bit binary.

https://github.com/Overv/VulkanTutorial/blob/56389adfef6d51493df7a50475fcf01c36f75c4d/code/09_shader_base.vert#L6

Excellent tutorial. It would be instructive to highlight the importance of the order the vertices are declared in, and that they correspond to the winding order used for front-faced culling. I had unknowingly swapped the last two vertices as I typed them in, which caused the triangl

I want to preemptively start this thread to survey for suggestions. A cursory search lead me to this promising repository https://github.com/enigo-rs/enigo

Since closing the window is a common point of failure, that will be the focus for the first pass of testing as I learn how to use the library.

Components for testing:

• bridge
• editor
• renderer
• settings
• wind

It would be really nice to have a more complete set of tutorials/samples to work with.

E.g. None of the samples use Geometry Instancing,Indirect Buffers.
From the Vulkan/Metal docs it is also not clear, because each API handles this slightly differently.

Best,
Marc

I wish doxygen documents is hosted online somewhere, it's kinda annoying just to clone and make my own copy to read documentations.
Anyway, this framework really helps me to make my own game engine.

Thanks.