在OpenGL中繪制多個2D紋理

Question

我正在使用Visual Studio 2013 Ultimate用C ++編寫自己的游戲庫。 我已經設法得到一些適用於1種紋理的基本圖形，但是當我添加另一種圖形時，它似乎會“覆蓋”以前的紋理。

例如， 如果我有紋理A和紋理B，則B會在應有A的位置繪制 。 我檢查過我沒有為每個紋理使用相同的紋理ID。

我看了一下OpenGL，嘗試繪制多個2d紋理，只有第1個出現 ，但這並不能解決我的問題。

我對OpenGL和較低級別的圖形編程非常陌生，有人可以向我解釋我的代碼有什么問題嗎？

---

這是我的紋理加載代碼：

Texture::Texture(const std::string& filePath)
{
    m_pixelData = PixelData(stbi_load(filePath.c_str(),
                                      &m_size.x,
                                      &m_size.y,
                                      &m_numberOfImageComponents,
                                      0 /* number of requested image components if non-zero */));

    // generate 1 texture name and store it in m_textureName
    glGenTextures(1, &m_textureID);

    // make this the active texture
    glBindTexture(GL_TEXTURE_2D, m_textureID);

    // specifies how the data to be uploaded is aligned
    glPixelStorei(GL_UNPACK_ALIGNMENT, 1);

    // set texture parameters (need to look up the details of what's going on here)
    // TODO work out what this does
    glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP);
    glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP);
    glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
    glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);

    // TODO work out what this does
    glTexEnvf(GL_TEXTURE_ENV, GL_TEXTURE_ENV_MODE, GL_MODULATE);

    GLenum imageFormat;

    // Work out image format
    // TODO add proper error handling
    switch(m_numberOfImageComponents)
    {
    case 4:
        imageFormat = GL_RGBA;
        break;
    case 3:
        imageFormat = GL_RGB;
        break;
    default:
        DDGL_LOG(ERROR, "No image formats with " + std::to_string(m_numberOfImageComponents) + "components known");
        throw;
    }

    // upload the texture to VRAM
    glTexImage2D(GL_TEXTURE_2D,
                0,
                imageFormat,
                m_size.x,
                m_size.y,
                0,
                imageFormat,
                GL_UNSIGNED_BYTE,
                m_pixelData.get());
}

---

這是我的“開始抽獎”代碼

void GraphicsAPIWrapper::startDraw()
{
    // TODO does this need to be called every frame?
    glMatrixMode(GL_PROJECTION); // select the matrix
    glLoadIdentity(); //reset the matrix

    glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
}

---

這是我的繪圖代碼

void GraphicsAPIWrapper::draw(std::shared_ptr<IDrawableGameObject> drawableObject)
{
        glPushMatrix();

        // convert pixel coordinates to decimals
        const Vector2F floatWindowSize          ((float) getWindowSize().x,                             (float) getWindowSize().y);
        const Vector2F floatObjectSize          ((float) drawableObject->getSize().x,                   (float) drawableObject->getSize().y);
        const Vector2F relativeObjectSize       (floatObjectSize.x / floatWindowSize.x,                 floatObjectSize.y / floatWindowSize.y);
        const Vector2F relativeObjectPosition   (drawableObject->getPosition().x / floatWindowSize.x,   drawableObject->getPosition().y / floatWindowSize.y);

        // transformations
        glTranslatef(relativeObjectPosition.x, relativeObjectPosition.y, 0.0f);
        glRotatef(drawableObject->getRotation(), 0.0f, 0.0f, 1.0f);

        // TODO should this be triangles or quads? I've been told triangles are generally higher performance
        // right now QUADS are simpler though
        glBegin(GL_QUADS);

        glBindTexture(GL_TEXTURE_2D, drawableObject->getTextureID());

        glTexCoord2f(0.0f, 1.0f); glVertex3f(-relativeObjectSize.x, -relativeObjectSize.y,  1.0f);
        glTexCoord2f(1.0f, 1.0f); glVertex3f(relativeObjectSize.x,  -relativeObjectSize.y,  1.0f);
        glTexCoord2f(1.0f, 0.0f); glVertex3f(relativeObjectSize.x,  relativeObjectSize.y,   1.0f);
        glTexCoord2f(0.0f, 0.0f); glVertex3f(-relativeObjectSize.x, relativeObjectSize.y,   1.0f);

        glEnd();

        glPopMatrix();
}

---

這是我的“結束抽獎”代碼

void GraphicsAPIWrapper::endDraw()
{
    glfwSwapBuffers(m_window->getWindow());
}

---

因此，非常明確地說，我得到的行為是一種紋理在各處繪制，而不是所需的不同紋理。

Answer 1

您正在glBegin/glEnd -Block內部調用glBindTexture() ，這是無效的，除了生成錯誤之外沒有任何作用。 因此，真正綁定的對象是在此之前的最后一個綁定-很有可能是在加載紋理時進行的綁定操作，因此最后加載的對象是為所有對象顯示的綁定操作...