Basic / Arithmetic Types

• Bool
• Int, UInt, Byte
• Float

Width-Specified Types

Signed	Unsigned	Float
Int8	UInt8
Int16	UInt16
Int32	UInt32	Float32
Int64	UInt64	Float64

Integer

Int is

• Int64 (like in Swift and Go) on virtually all modern systems,
• Int32 on systems with a 32-bit address bus only (i.e. old/small platforms, like Win32/x86),
• Int16 on systems with a 16-bit address bus (i.e. very old/small microcontrollers like AVR/ATmega328/Arduino Uno, 6502, Z80).

Simply use Int,

• not Size, SSize, nor PtrDiff,
• not Long nor LongLong.

Boolean

It is Bool, not bool nor Boolean.

Byte

Byte is std::byte, i.e. not the same type as UInt8 for parameter overloading. It is used to represent raw memory, for binary buffers, serialization, or low-level I/O.

Floating Point

Float is

• Float64 / double precision (like in Python, unlike C/C++).
• Float32 on old/small platforms only (i.e. those with hardware support for Float32 but not for Float64).

With modern processors double precision is very fast. When concerned with memory bandwidth, cache size, and SIMD performance, choose one of the smaller floating-point types like Float32 or (if available) BFloat16, Float16, Float8, maybe even Float4.