diff --git a/CMakeLists.txt b/CMakeLists.txt
index 29a9e9b..2dc4019 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -7,7 +7,6 @@ add_library(picoled
   include/picoled/SSD1306.h
 
   src/picoled.cpp
-  src/buffer.cpp
   src/2D.cpp
   src/SSD1306.cpp
 )
diff --git a/include/picoled.h b/include/picoled.h
index baf539e..13fbcb1 100644
--- a/include/picoled.h
+++ b/include/picoled.h
@@ -32,7 +32,7 @@ public:
 
   uint8_t grayscale() const
   {
-    uint16_t unscaled = red() + green() + blue();
+    uint16_t unscaled = static_cast<uint16_t>(r) + static_cast<uint16_t>(g) + static_cast<uint16_t>(b);
     return unscaled / 3;
   }
 
diff --git a/include/picoled/buffer.h b/include/picoled/buffer.h
index 4ad04ea..a5b74ab 100644
--- a/include/picoled/buffer.h
+++ b/include/picoled/buffer.h
@@ -6,23 +6,24 @@
 namespace picoled
 {
 
+// behavior when out of bounds
+enum class buffer_read_mode
+{
+  empty, // use default constructor for T
+  wrap,    // use modular arithmetic to put back in bounds
+  clamp    // pick from the corresponding edge
+};
+
 template <typename T>
 class buffer
 {
 public:
-  // behavior when out of bounds
-  enum class ReadMode
-  {
-    Default, // use default constructor for T
-    Wrap,    // use modular arithmetic to put back in bounds
-    Clamp    // pick from the corresponding edge
-  };
-
   buffer(size_t width, size_t height) :
     width(width),
     height(height),
     data(new T[width * height])
-  {}
+  {
+  }
   ~buffer()
   {
     delete[] data;
@@ -44,17 +45,22 @@ public:
     data[x + y * width] = pixel;
   }
 
-  T read(int x, int y, ReadMode mode = ReadMode::Default)
+  T read_uv(float x, float y, buffer_read_mode mode = buffer_read_mode::clamp) const
+  {
+    return read(x * width, y * height, mode);
+  }
+
+  T read(int x, int y, buffer_read_mode mode = buffer_read_mode::clamp) const
   {
     switch (mode)
     {
-    case ReadMode::Default:
+    case buffer_read_mode::empty:
       return read_default(x, y);
 
-    case ReadMode::Wrap:
+    case buffer_read_mode::wrap:
       return read_wrap(x, y);
 
-    case ReadMode::Clamp:
+    case buffer_read_mode::clamp:
       return read_clamp(x, y);
     }
 
@@ -74,13 +80,13 @@ private:
   size_t height;
   T* data;
 
-  T read_default(int x, int y)
+  T read_default(int x, int y) const
   {
     if (x < 0 || x >= width || y < 0 || y >= height) { return T(); }
     return data[x + y * width];
   }
 
-  T read_wrap(int x, int y)
+  T read_wrap(int x, int y) const
   {
     if (x < 0)
     {
@@ -103,7 +109,7 @@ private:
     return data[x + y * width];
   }
 
-  T read_clamp(int x, int y)
+  T read_clamp(int x, int y) const
   {
     if (x < 0) { x = 0; }
     else if (x >= width) { x = width - 1; }
diff --git a/include/picoled/gfx/2D.h b/include/picoled/gfx/2D.h
index fddf2dd..2e445fa 100644
--- a/include/picoled/gfx/2D.h
+++ b/include/picoled/gfx/2D.h
@@ -21,18 +21,53 @@ public:
   float operator^(const vec2& other) const { return x * other.y - y * other.x; }
 };
 
+class vec3
+{
+public:
+  float x;
+  float y;
+  float z;
+
+  vec3 operator+(const vec3& other) const { return { x + other.x, y + other.y, z + other.z }; }
+  vec3 operator-(const vec3& other) const { return { x - other.x, y - other.y, z - other.z }; }
+  vec3 operator*(const vec3& other) const { return { x * other.x, y * other.y, z * other.z }; }
+  vec3 operator/(const vec3& other) const { return { x / other.x, y / other.y, z / other.z }; }
+
+  float operator%(const vec3& other) const { return x * other.x + y * other.y + z * other.z; }
+  vec3 operator^(const vec3& other) const
+  {
+    return
+    {
+      y * other.z - z * other.y,
+      z * other.x - x * other.z,
+      x * other.y - y * other.x
+    };
+  }
+};
+
+class vertex2
+{
+public:
+  vec2 position;
+  vec2 texture_coordinates;
+};
+
 class triangle
 {
 public:
-  vec2 a;
-  vec2 b;
-  vec2 c;
+  vertex2 a;
+  vertex2 b;
+  vertex2 c;
 
   triangle fix_winding() const;
-  bool inside(const vec2& p) const;
+  vec3 barycentric(const vec2& p) const;
+  bool contains(const vec2& p) const;
 
   vec2 min() const;
   vec2 max() const;
+
+private:
+  float area_double() const;
 };
 
 class gfx2d
@@ -48,7 +83,7 @@ public:
   gfx2d& operator=(const gfx2d&) = delete;
   gfx2d& operator=(gfx2d&&) = delete;
 
-  void draw_triangle(triangle tri, pixel p = 0xFF, bool ignore_winding = true);
+  void draw_triangle(triangle tri, const buffer<pixel>& texture, bool ignore_winding = true);
 
 private:
   oled* screen;
diff --git a/src/2D.cpp b/src/2D.cpp
index 345ba0e..64f57a9 100644
--- a/src/2D.cpp
+++ b/src/2D.cpp
@@ -1,13 +1,16 @@
 #include "picoled/gfx/2D.h"
+#include "picoled/buffer.h"
 
 #include <cmath>
 
+#include <pico/stdio.h>
+
 namespace picoled::gfx
 {
 
 triangle triangle::fix_winding() const
 {
-  if (((b - a) ^ (c - a)) < 0)
+  if (((b.position - a.position) ^ (c.position - a.position)) < 0)
   {
     return { a, c, b };
   }
@@ -17,11 +20,19 @@ triangle triangle::fix_winding() const
   }
 }
 
-bool triangle::inside(const vec2& p) const
+vec3 triangle::barycentric(const vec2& p) const
 {
-  const float abp = (b - a) ^ (p - a);
-  const float bcp = (c - b) ^ (p - b);
-  const float cap = (a - c) ^ (p - c);
+  const float total = area_double();
+  const float u = triangle{ b, a, p }.area_double() / total;
+  const float v = triangle{ a, c, p }.area_double() / total;
+  return vec3{ u, v, 1 - u - v };
+}
+
+bool triangle::contains(const vec2& p) const
+{
+  const float abp = (b.position - a.position) ^ (p - a.position);
+  const float bcp = (c.position - b.position) ^ (p - b.position);
+  const float cap = (a.position - c.position) ^ (p - c.position);
 
   return abp >= 0 && bcp >= 0 && cap >= 0;
 }
@@ -30,8 +41,8 @@ vec2 triangle::min() const
 {
   return
   {
-    fminf(fminf(a.x, b.x), c.x),
-    fminf(fminf(a.y, b.y), c.y)
+    fminf(fminf(a.position.x, b.position.x), c.position.x),
+    fminf(fminf(a.position.y, b.position.y), c.position.y)
   };
 }
 
@@ -39,12 +50,19 @@ vec2 triangle::max() const
 {
   return
   {
-    fmaxf(fmaxf(a.x, b.x), c.x),
-    fmaxf(fmaxf(a.y, b.y), c.y)
+    fmaxf(fmaxf(a.position.x, b.position.x), c.position.x),
+    fmaxf(fmaxf(a.position.y, b.position.y), c.position.y)
   };
 }
 
-void gfx2d::draw_triangle(triangle tri, pixel p, bool ignore_winding)
+float triangle::area_double() const
+{
+  // don't divide by two since it would cancel out
+  // in the barycentric coordinate calculation anyway
+  return std::abs((b.position - a.position) ^ (c.position - a.position));
+}
+
+void gfx2d::draw_triangle(triangle tri, const buffer<pixel>& texture, bool ignore_winding)
 {
   if (ignore_winding) { tri = tri.fix_winding(); }
 
@@ -55,7 +73,15 @@ void gfx2d::draw_triangle(triangle tri, pixel p, bool ignore_winding)
   {
     for (float j = min.y; j <= max.y; j += 0.75f)
     {
-      if (tri.inside({ i, j })) { screen->write(i, j, p); }
+      const vec2 position = { i, j };
+      if (tri.contains(position))
+      {
+        const vec3 barycentric = tri.barycentric(position);
+        const float u = barycentric % vec3{ tri.a.texture_coordinates.x, tri.b.texture_coordinates.x, tri.c.texture_coordinates.x };
+        const float v = barycentric % vec3{ tri.a.texture_coordinates.y, tri.b.texture_coordinates.y, tri.c.texture_coordinates.y };
+        const pixel read = texture.read_uv(u, v, buffer_read_mode::clamp);
+        screen->write(i, j, read);
+      }
     }
   }
 }
diff --git a/src/SSD1306.cpp b/src/SSD1306.cpp
index 566dada..04d10f4 100644
--- a/src/SSD1306.cpp
+++ b/src/SSD1306.cpp
@@ -81,7 +81,7 @@ ssd1306::~ssd1306()
 void ssd1306::update_impl()
 {
   // clear display to zero
-  memset(&buffer[1], 0, get_width() * get_height() * sizeof(*buffer));
+  memset(&buffer[1], 0, get_width() * get_height() * sizeof(*buffer) / 8);
 
   for (int x = 0; x < get_width(); x++)
   {
diff --git a/src/buffer.cpp b/src/buffer.cpp
deleted file mode 100644
index e69de29..0000000