Vector3int16¶
A Vector3 with signed 16-bit integers for components.
Description¶
The Vector3int16 data type represents a vector in 3D space with a signed 16-bit integer for its components. It is similar to Vector3 in that it allows for the same arithmetic operations, but it lacks commonly used vector functions.
Vector3int16 should not be confused with:
- Vector3, a more precise and complete implementation for 3D vectors.
- Vector2int16, a similar implementation for 2D vectors.
For each component:
- The lower bound is -215, or -32,768.
- The upper bound is 215 − 1, or 32,767.
Converting to Vector3¶
To convert a Vector3int16 to a Vector3, construct a Vector3 by passing each component of the Vector3int16 to Vector3.new():
local vector3int16 = Vector3int16.new(1, 2, 3)
local vector3 = Vector3.new(vector3int16.X, vector3int16.Y, vector3int16.Z)
print(vector3) --> 1, 2, 3
Do not pass an entire Vector3int16 to Vector3.new(),
as the constructor interprets a Vector3int16 as a 0 within its
parameters without producing an error. This can lead to silent logic
errors if you do something like:
local vector3int16 = Vector3int16.new(1, 2, 3)
local vector3 = Vector3.new(vector3int16)
print(vector3) --> 0, 0, 0
Math Operations¶
The following math operations are valid for the Vector3int16 data type. For all operations, be mindful of the bounds associated with signed 16-bit integers, described earlier.
| Operation | Description |
Datatype.Vector3int16 + Datatype.Vector3int16 |
Produces a Datatype.Vector3int16 whose components are the sum of the operands' respective components. |
Datatype.Vector3int16 - Datatype.Vector3int16 |
Produces a Datatype.Vector3int16 whose components are the difference of the operands' respective components. |
Datatype.Vector3int16 * Datatype.Vector3int16 |
Produces a Datatype.Vector3int16 whose components are the product of the operands' respective components. |
Datatype.Vector3int16 / Datatype.Vector3int16 |
Produces a Datatype.Vector3int16 whose components are the quotient of the operands' respective components. The results of the
division are rounded down. |
Datatype.Vector3int16 * number |
Produces a Datatype.Vector3int16 whose components are the product of the respective Datatype.Vector3int16 components and the number (factor). This operation is commutative. |
Datatype.Vector3int16 / number |
Produces a Datatype.Vector3int16 whose components are the quotient of the respective Datatype.Vector3int16 components and the number (divisor). The results of the
division are rounded toward zero. |
Constructors¶
new Supported Returns a new Vector3int16 from the given x, y, and z components.
Returns a new Vector3int16 from the given x, y, and z components.
Returns a new Vector3int16 from the given x, y and z components. Non-integer components are rounded down.
The components must fall within the range [-215, 215). If outside this range, integer overflow may occur. For example, providing 32,768 (equal to 215) as a component overflows the 16-bit integer, and so the component will be -32,768 (equal to -215) instead.
Properties¶
X Supported The x-coordinate of the Vector3int16.
The x-coordinate of the Vector3int16.
The x-coordinate of the Vector3int16, also accessible in its lower-case variant.
Y Supported The y-coordinate of the Vector3int16.
The y-coordinate of the Vector3int16.
The y-coordinate of the Vector3int16, also accessible in its lower-case variant.
Z Supported The z-coordinate of the Vector3int16.
The z-coordinate of the Vector3int16.
The z-coordinate of the Vector3int16, also accessible in its lower-case variant.
Math operations¶
+ Check Produces a Vector3int16 whose components are the sum of the operands' respective components.
Produces a Vector3int16 whose components are the sum of the operands' respective components. Be mindful of the bounds associated with signed 16-bit integers, described earlier.
- Check Produces a Vector3int16 whose components are the difference of the operands' respective components.
Produces a Vector3int16 whose components are the difference of the operands' respective components. Be mindful of the bounds associated with signed 16-bit integers, described earlier.
* Check Produces a Vector3int16 whose components are the product of the operands' respective components.
Produces a Vector3int16 whose components are the product of the operands' respective components. Be mindful of the bounds associated with signed 16-bit integers, described earlier.
/ Check Produces a Vector3int16 whose components are the quotient of the operands' respective components. The results of the di…
Produces a Vector3int16 whose components are the quotient of the operands' respective components. The results of the division are rounded down. Be mindful of the bounds associated with signed 16-bit integers, described earlier.
* Check Produces a Vector3int16 whose components are the product of the respective Vector3int16 components and the number (fact…
Produces a Vector3int16 whose components are the product of the respective Vector3int16 components and the number (factor). This operation is commutative. Be mindful of the bounds associated with signed 16-bit integers, described earlier.
/ Check Produces a Vector3int16 whose components are the quotient of the respective Vector3int16 components and the number (div…
Produces a Vector3int16 whose components are the quotient of the respective Vector3int16 components and the number (divisor). The results of the division are rounded toward zero. Be mindful of the bounds associated with signed 16-bit integers, described earlier.
Nova binding inventory¶
| Library | Members |
|---|---|
new |
X, Y, Z, x, y, z |
Roblox reference: Vector3int16