diff --git a/src/lib.rs b/src/lib.rs
index c2c641e..4dfb80f 100644
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -84,6 +84,7 @@
 //! ```
 
 pub mod physics;
+pub mod vec;
 
 use std::{
     cell::Cell,
@@ -187,8 +188,10 @@ pub struct Sprite {
     pub rect: Rect,
     surf: Surface<'static>,
     angle: f64,
+    is_platform: bool,
 }
 
+
 impl Sprite {
     /// Create a new Sprite. The `path` is relative to the current directory while running.
     ///
@@ -197,7 +200,7 @@ impl Sprite {
     /// # use cat_box::*;
     /// let s = Sprite::new("duck.png", 500, 400).unwrap();
     /// ```
-    pub fn new<P: AsRef<Path>>(path: P, x: i32, y: i32) -> Result<Self> {
+    pub fn new<P: AsRef<Path>>(path: P, x: i32, y: i32, plat: bool) -> Result<Self> {
         let ops = RWops::from_file(path, "r")?;
         let surf = ops.load()?;
 
@@ -208,6 +211,7 @@ impl Sprite {
             rect: dest_rect,
             surf,
             angle: 0.0,
+            is_platform: plat,
         })
     }
 
@@ -219,7 +223,7 @@ impl Sprite {
     /// let bytes = include_bytes!("../duck.png");
     /// let s = Sprite::from_bytes(bytes, 500, 400).unwrap();
     /// ```
-    pub fn from_bytes<B: AsRef<[u8]>>(bytes: B, x: i32, y: i32) -> Result<Self> {
+    pub fn from_bytes<B: AsRef<[u8]>>(bytes: B, x: i32, y: i32, plat: bool) -> Result<Self> {
         let ops = RWops::from_bytes(bytes.as_ref())?;
         let surf = ops.load()?;
 
@@ -230,6 +234,7 @@ impl Sprite {
             rect: dest_rect,
             surf,
             angle: 0.0,
+            is_platform: plat,
         })
     }
 
diff --git a/src/vec.rs b/src/vec.rs
new file mode 100644
index 0000000..927dbc6
--- /dev/null
+++ b/src/vec.rs
@@ -0,0 +1,472 @@
+//! Types representing directions and locations in 2d and 3d space.
+//! 
+//! NOTE: `Vec3`, `Vec3Int`, and `Direction3` don't exist yet. Coming soon!
+//! 
+//! This crate contains 6 major types:
+//!  - [`Vec2`], a 2d float vector
+//!  - [`Vec2Int`], a 2d integer vector
+//!  - [`Direction2`], a 2d cardinal direction
+//!  - [`Vec3`], a 3d float vector (TODO)
+//!  - [`Vec3Int`], a 3d integer vector (TODO)
+//!  - [`Direction3`], a 3d cardinal direction (TODO)
+//! 
+//! All the types implement the expected [`From`]s and all the relevant operator traits.
+
+#![warn(clippy::pedantic)]
+#![warn(missing_docs)]
+#![allow(clippy::must_use_candidate)]
+
+
+
+use std::{ops::{Add, AddAssign, Div, DivAssign, Mul, MulAssign, Neg, Sub, SubAssign}, fmt::Debug};
+
+// Direction
+/// A cardinal direction in a 2d plane.
+///
+/// Conversions to a [`Vec2`] or [`Vec2Int`] assume that East is positive-x and South is positive-y.
+#[derive(Debug, Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash)]
+pub enum Direction2 {
+    /// North, or Vec2::from((0, -1))
+    North,
+    /// North, or Vec2::from((0, 1))
+    South,
+    /// North, or Vec2::from((1, 0))
+    East,
+    /// North, or Vec2::from((-1, 0))
+    West,
+}
+
+#[allow(clippy::enum_glob_use)]
+impl Direction2 {
+    /// Flips this `Direction` around both the x- and y-axes.
+    pub fn flipped(self) -> Self {
+        self.flip_x().flip_y()
+    }
+
+    /// Flips this `Direction` around the x-axis.
+    pub fn flip_x(self) -> Self {
+        use Direction2::*;
+        match self {
+            East => West,
+            West => East,
+            v => v,
+        }
+    }
+
+    /// Flips this `Direction` around the y-axis.
+    pub fn flip_y(self) -> Self {
+        use Direction2::*;
+        match self {
+            North => South,
+            South => North,
+            v => v,
+        }
+    }
+}
+
+// ...and related op impls
+impl Neg for Direction2 {
+    type Output = Self;
+
+    fn neg(self) -> Self::Output {
+        self.flipped()
+    }
+}
+
+#[allow(clippy::enum_glob_use)]
+impl From<Direction2> for Vec2 {
+    fn from(v: Direction2) -> Self {
+        use Direction2::*;
+        match v {
+            North => (0.0, -1.0).into(),
+            South => (0.0, 1.0).into(),
+            East => (1.0, 0.0).into(),
+            West => (-1.0, 0.0).into(),
+        }
+    }
+}
+
+#[allow(clippy::enum_glob_use)]
+impl From<Direction2> for Vec2Int {
+    fn from(v: Direction2) -> Self {
+        use Direction2::*;
+        match v {
+            North => (0, -1).into(),
+            South => (0, 1).into(),
+            East => (1, 0).into(),
+            West => (-1, 0).into(),
+        }
+    }
+}
+
+impl Mul<f32> for Direction2 {
+    type Output = Vec2;
+
+    fn mul(self, rhs: f32) -> Self::Output {
+        Vec2::from(self) * rhs
+    }
+}
+
+impl Mul<i32> for Direction2 {
+    type Output = Vec2Int;
+
+    fn mul(self, rhs: i32) -> Self::Output {
+        Vec2Int::from(self) * rhs
+    }
+}
+
+// Vec2
+/// A set of 2 [`f32`]s representing a location or direction in the 2d plane.
+#[derive(Clone, Copy, Default, PartialEq)]
+pub struct Vec2 {
+    /// The x component of the vector.
+    pub x: f32,
+    /// The y component of the vector.
+    pub y: f32,
+}
+
+impl Debug for Vec2 {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        f.debug_tuple("Vec2").field(&self.x).field(&self.y).finish()
+    }
+}
+
+impl Vec2 {
+    /// Creates a new `Vec2` with the given x- and y-values.
+    ///
+    /// It is often simpler, and preferred, to just write `(x, y).into()`.
+    pub const fn new(x: f32, y: f32) -> Vec2 {
+        Self { x, y }
+    }
+
+    /// Gets the squared magnitude of the vector.
+    ///
+    /// Useful for comparisons as it is faster to calculate than `magnitude`.
+    pub fn sq_magnitude(self) -> f32 {
+        self.x * self.x + self.y * self.y
+    }
+
+    /// Gets the magnitude of the vector.
+    pub fn magnitude(self) -> f32 {
+        self.sq_magnitude().sqrt()
+    }
+
+    /// Gets the squared distance from this vector to `rhs`.
+    ///
+    /// Useful for comparisons as it is faster to calculate than `dist`.
+    pub fn sq_dist(self, rhs: Self) -> f32 {
+        (self - rhs).sq_magnitude()
+    }
+
+    /// Gets the distance from this vector to `rhs`.
+    pub fn dist(self, rhs: Self) -> f32 {
+        (self - rhs).magnitude()
+    }
+
+    /// Normalizes the vector, making its magnitude `1`.
+    pub fn normalized(self) -> Self {
+        self / self.magnitude()
+    }
+
+    /// Rounds the vector to a [`Vec2Int`].
+    ///
+    /// This uses `as i32` under the hood, and as such comes with all the same unfortunate edge cases. Beware.
+    pub fn rounded(self) -> Vec2Int {
+        #[allow(clippy::cast_possible_truncation)]
+        Vec2Int {
+            x: self.x as i32,
+            y: self.y as i32,
+        }
+    }
+}
+
+impl From<(i32, i32)> for Vec2 {
+    fn from(v: (i32, i32)) -> Self {
+        Vec2Int::from(v).to_f32()
+    }
+}
+
+impl From<(f32, f32)> for Vec2 {
+    fn from(v: (f32, f32)) -> Self {
+        Self { x: v.0, y: v.1 }
+    }
+}
+
+impl From<Vec2> for (f32, f32) {
+    fn from(v: Vec2) -> Self {
+        (v.x, v.y)
+    }
+}
+
+impl PartialEq<(i32, i32)> for Vec2 {
+    fn eq(&self, other: &(i32, i32)) -> bool {
+        self == &Self::from(*other)
+    }
+}
+
+impl PartialEq<(f32, f32)> for Vec2 {
+    fn eq(&self, other: &(f32, f32)) -> bool {
+        self == &Self::from(*other)
+    }
+}
+
+// ...and related op impls
+impl Neg for Vec2 {
+    type Output = Self;
+
+    fn neg(self) -> Self::Output {
+        self * -1.0
+    }
+}
+
+impl Add for Vec2 {
+    type Output = Self;
+
+    fn add(self, rhs: Self) -> Self::Output {
+        Self {
+            x: self.x + rhs.x,
+            y: self.y + rhs.y,
+        }
+    }
+}
+
+impl Add<Direction2> for Vec2 {
+    type Output = Self;
+
+    fn add(self, rhs: Direction2) -> Self::Output {
+        self + Self::from(rhs)
+    }
+}
+
+impl<T> AddAssign<T> for Vec2
+where
+    Vec2: Add<T, Output = Self>,
+{
+    fn add_assign(&mut self, rhs: T) {
+        *self = *self + rhs;
+    }
+}
+
+impl<T> Sub<T> for Vec2
+where
+    Vec2: Add<T, Output = Self>,
+{
+    type Output = Self;
+
+    fn sub(self, rhs: T) -> Self::Output {
+        -(-self + rhs)
+    }
+}
+
+impl<T> SubAssign<T> for Vec2
+where
+    Vec2: Sub<T, Output = Self>,
+{
+    fn sub_assign(&mut self, rhs: T) {
+        *self = *self - rhs;
+    }
+}
+
+impl Mul<f32> for Vec2 {
+    type Output = Self;
+
+    fn mul(self, rhs: f32) -> Self::Output {
+        Self {
+            x: self.x * rhs,
+            y: self.y * rhs,
+        }
+    }
+}
+
+impl Div<f32> for Vec2 {
+    type Output = Self;
+
+    fn div(self, rhs: f32) -> Self::Output {
+        Self {
+            x: self.x / rhs,
+            y: self.y / rhs,
+        }
+    }
+}
+
+impl MulAssign<f32> for Vec2 {
+    fn mul_assign(&mut self, rhs: f32) {
+        *self = *self * rhs;
+    }
+}
+
+impl DivAssign<f32> for Vec2 {
+    fn div_assign(&mut self, rhs: f32) {
+        *self = *self / rhs;
+    }
+}
+
+// Vec2Int
+/// A set of 2 [`i32`]s representing a location or direction in the 2d plane.
+#[derive(Clone, Copy, Default, PartialEq, Eq, Hash)]
+pub struct Vec2Int {
+    /// The x component of the vector.
+    pub x: i32,
+    /// The y component of the vector.
+    pub y: i32,
+}
+
+impl Debug for Vec2Int {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        f.debug_tuple("Vec2Int").field(&self.x).field(&self.y).finish()
+    }
+}
+
+impl Vec2Int {
+    /// Creates a new `Vec2` with the given x- and y-values.
+    ///
+    /// It is often simpler, and preferred, to just write `(x, y).into()`.
+    pub const fn new(x: i32, y: i32) -> Vec2Int {
+        Self { x, y }
+    }
+
+    /// Gets the squared magnitude of the vector.
+    ///
+    /// Useful for comparisons as it is faster to calculate than `magnitude`.
+    pub fn sq_magnitude(self) -> i32 {
+        self.x * self.x + self.y * self.y
+    }
+
+    /// Gets the magnitude of the vector.
+    pub fn magnitude(self) -> f32 {
+        #[allow(clippy::cast_precision_loss)]
+        (self.sq_magnitude() as f32).sqrt()
+    }
+
+    /// Gets the squared distance from this vector to `rhs`.
+    ///
+    /// Useful for comparisons as it is faster to calculate than `dist`.
+    pub fn sq_dist(self, rhs: Self) -> i32 {
+        (self - rhs).sq_magnitude()
+    }
+
+    /// Gets the distance from this vector to `rhs`.
+    pub fn dist(self, rhs: Self) -> f32 {
+        (self - rhs).magnitude()
+    }
+
+    /// Casts this vector to a [`Vec2`].
+    ///
+    /// This uses `as f32` under the hood, and as such comes with all the same unfortunate edge cases. Beware.
+    pub fn to_f32(self) -> Vec2 {
+        #[allow(clippy::cast_precision_loss)]
+        Vec2 {
+            x: self.x as f32,
+            y: self.y as f32,
+        }
+    }
+}
+
+impl From<(i32, i32)> for Vec2Int {
+    fn from(v: (i32, i32)) -> Self {
+        Self { x: v.0, y: v.1 }
+    }
+}
+
+impl From<Vec2Int> for (i32, i32) {
+    fn from(v: Vec2Int) -> Self {
+        (v.x, v.y)
+    }
+}
+
+impl PartialEq<(i32, i32)> for Vec2Int {
+    fn eq(&self, other: &(i32, i32)) -> bool {
+        self == &Self::from(*other)
+    }
+}
+
+// ...and related op impls
+impl Neg for Vec2Int {
+    type Output = Self;
+
+    fn neg(self) -> Self::Output {
+        self * -1
+    }
+}
+
+impl Add for Vec2Int {
+    type Output = Self;
+
+    fn add(self, rhs: Self) -> Self::Output {
+        Self {
+            x: self.x + rhs.x,
+            y: self.y + rhs.y,
+        }
+    }
+}
+
+impl Add<Direction2> for Vec2Int {
+    type Output = Self;
+
+    fn add(self, rhs: Direction2) -> Self::Output {
+        self + Self::from(rhs)
+    }
+}
+
+impl<T> AddAssign<T> for Vec2Int
+where
+    Vec2Int: Add<T, Output = Self>,
+{
+    fn add_assign(&mut self, rhs: T) {
+        *self = *self + rhs;
+    }
+}
+
+impl<T> Sub<T> for Vec2Int
+where
+    Vec2Int: Add<T, Output = Self>,
+{
+    type Output = Self;
+
+    fn sub(self, rhs: T) -> Self::Output {
+        -(-self + rhs)
+    }
+}
+
+impl<T> SubAssign<T> for Vec2Int
+where
+    Vec2Int: Sub<T, Output = Self>,
+{
+    fn sub_assign(&mut self, rhs: T) {
+        *self = *self - rhs;
+    }
+}
+
+impl Mul<i32> for Vec2Int {
+    type Output = Self;
+
+    fn mul(self, rhs: i32) -> Self::Output {
+        Self {
+            x: self.x * rhs,
+            y: self.y * rhs,
+        }
+    }
+}
+
+impl Div<i32> for Vec2Int {
+    type Output = Self;
+
+    fn div(self, rhs: i32) -> Self::Output {
+        Self {
+            x: self.x / rhs,
+            y: self.y / rhs,
+        }
+    }
+}
+
+impl MulAssign<i32> for Vec2Int {
+    fn mul_assign(&mut self, rhs: i32) {
+        *self = *self * rhs;
+    }
+}
+
+impl DivAssign<i32> for Vec2Int {
+    fn div_assign(&mut self, rhs: i32) {
+        *self = *self / rhs;
+    }
+}