Question

密切关注转换反馈的this教程，但是在python中。给出一个简单的变换反馈顶点着色器，

#version 400 core
in float inValue;
out float outValue;

void main()
{
  outValue = 5*inValue+6;
}

和周围的python代码来调用它，

...

varyings = ctypes.create_string_buffer('outValue')
varyings_pp = POINTER(POINTER(c_char))(ctypes.cast(varyings, POINTER(c_char)))
gl.glTransformFeedbackVaryings(program, 1, varyings_pp, gl.GL_INTERLEAVED_ATTRIBS);

gl.glLinkProgram(program)
gl.glUseProgram(program)

vao = gl.glGenVertexArrays(1)
gl.glBindVertexArray(vao)

data = np.array([1.,2.,3.,4.,5.], dtype=np.float32)
vbo = gl.glGenBuffers(1)
gl.glBindBuffer(gl.GL_ARRAY_BUFFER, vbo)
gl.glBufferData(gl.GL_ARRAY_BUFFER, data.nbytes, data, gl.GL_STATIC_DRAW)

inputAttrib = gl.glGetAttribLocation(program, "inValue")
gl.glEnableVertexAttribArray(inputAttrib)
gl.glVertexAttribPointer(inputAttrib, 1, gl.GL_FLOAT, gl.GL_FALSE, 0, 0)

data = np.array([-1.,-2.,-3.,-4.,-5.], dtype=np.float32)
tbo = gl.glGenBuffers(1)
gl.glBindBuffer(gl.GL_TRANSFORM_FEEDBACK_BUFFER, tbo)
gl.glBufferData(gl.GL_TRANSFORM_FEEDBACK_BUFFER, data.nbytes, data, gl.GL_STATIC_READ)

gl.glEnable(gl.GL_RASTERIZER_DISCARD)

gl.glBindBufferBase(gl.GL_TRANSFORM_FEEDBACK_BUFFER, 0, tbo)

gl.glBeginTransformFeedback(gl.GL_POINTS)
gl.glDrawArrays(gl.GL_POINTS, 0, 3)
gl.glEndTransformFeedback()

gl.glFlush()

r_data = np.empty_like(data)
gl.glGetBufferSubData(gl.GL_TRANSFORM_FEEDBACK_BUFFER, 0, data.nbytes, r_data)
print r_data

我得到输出[6. 6. 6. -4. -5.]。因此，似乎inValue始终为0.但是，如果我回读GL_ARRAY_BUFFER，我会得到输入数据。我是否错误地将缓冲区绑定到顶点数组？在任何情况下都不会引发任何glError，并且着色器没有编译错误（出于审美目的而省略了错误检查代码）。

Answer 1

解决方案很简单，因动态打字而容易错过。这一行

gl.glVertexAttribPointer(inputAttrib, 1, gl.GL_FLOAT, gl.GL_FALSE, 0, 0)

应该阅读

gl.glVertexAttribPointer(inputAttrib, 1, gl.GL_FLOAT, gl.GL_FALSE, 0, ctypes.c_void_p(0))

Answer 2

请使用以下代码。

r_data = (c_float * data.size)()
gl.glGetBufferSubData(GL_TRANSFORM_FEEDBACK_BUFFER, 0, data.nbytes, byref(r_data))
r_data = np.frombuffer(r_data, dtype=np.float32)
print(r_data)

简单的Python + OpenGL转换反馈

2 个答案: