Word to Gigabit

A word's size depends on the CPU architecture. In x86/x64 (Intel/AMD) documentation: word = 16 bits, DWORD = 32 bits, QWORD = 64 bits. In ARM 32-bit: word = 32 bits. In most modern 64-bit systems (excluding x86 documentation): word = 64 bits. When reading technical documentation, always check the architecture's definition, as "word" is not a universal fixed size.

In Windows API documentation and x86 architecture, a DWORD (Double Word) = 32 bits = 4 bytes, capable of holding values 0–4,294,967,295 (unsigned) or -2,147,483,648 to 2,147,483,647 (signed). DWORD is the most common fixed-width integer type in the Windows API, used for flags, handles, and return codes. The equivalent in modern C/C++ is uint32_t (unsigned) or int32_t (signed).

A CPU's word size determines: (1) the maximum addressable memory — a 32-bit CPU addresses up to 4 GiB (2³² bytes); a 64-bit CPU addresses up to 16 EiB (2⁶⁴ bytes); (2) the precision of integer arithmetic — a 64-bit word handles numbers up to ~18.4 × 10¹⁸ in a single instruction; (3) performance — operations on data smaller than the word size may require extra sign-extension instructions on some architectures.

Modern x86-64 CPUs (Intel Core, AMD Ryzen) have 64-bit general-purpose registers, so their native word size is 64 bits for most operations. However, x86 documentation maintains the legacy definition: "word" = 16 bits, DWORD = 32 bits, QWORD = 64 bits. This creates a confusing terminology mismatch between the architectural naming convention and the physical register size.

Memory alignment means storing data at addresses that are multiples of the data's size. A 32-bit word should be stored at an address divisible by 4 (bytes); a 64-bit word at an address divisible by 8. Misaligned access is either forbidden (causes a CPU fault) or penalised (requires two memory reads instead of one). Compilers automatically align variables; manual struct packing can create misalignment that causes subtle performance issues or crashes on strict architectures.

1 Gbps (gigabit) broadband delivers up to 125 MB/s, which is more than sufficient for most households. It supports dozens of simultaneous 4K streams, fast game downloads, and video conferencing with headroom to spare. The limiting factor is usually the Wi-Fi router (Wi-Fi 5 maxes out around 400–600 Mbps in practice) or the speed of the remote server you're downloading from.

10 Gbps networking is standard in data centers, server interconnects, and high-performance workstations doing large file transfers (video editing, database backups). It is increasingly available in prosumer home networking equipment. At 10 Gbps, a 1 TB file transfer takes about 13 minutes under ideal conditions.

One terabit equals 1,000 gigabits (SI). Terabit-per-second (Tbps) speeds are used in long-haul fiber optic cables and internet backbone infrastructure. A single transatlantic fiber cable typically carries hundreds of terabits per second across many multiplexed channels.

Wi-Fi 5 (802.11ac) delivers up to 3.5 Gbps theoretical, but typically 400–600 Mbps real-world on a single device. Wi-Fi 6 (802.11ax) reaches 9.6 Gbps theoretical and 600–900 Mbps practical per device, with better multi-device handling via OFDMA. Wi-Fi 6E extends the same technology into the uncongested 6 GHz band, improving real-world speeds to 1–2 Gbps. Wi-Fi 7 (802.11be) pushes the theoretical maximum to 46 Gbps using 320 MHz channels and 4096-QAM, with real-world single-device speeds expected around 2–5 Gbps — the first Wi-Fi standard to reliably exceed gigabit in practice.

Modern data centers handle enormous simultaneous traffic between thousands of servers — cloud computing, video streaming, and AI training all require massive internal bandwidth. 100 Gbps links between switches are now standard; 400 Gbps is increasingly deployed for spine connections. At these speeds, a single link can move 50 GB of data per second, keeping pace with NVMe storage arrays and GPU memory transfer rates.