Mebibyte to Gigabyte

MB (megabyte) = 1,000,000 bytes (SI decimal). MiB (mebibyte) = 1,048,576 bytes (IEC binary). MiB is 4.86% larger. This gap is why a file manager on Linux showing "512 MiB" of free RAM and a marketing sheet showing "512 MB" of RAM are technically different: the marketing sheet refers to fewer bytes.

The original CD-ROM standard defined capacity as 74 minutes of audio or 650,000,000 bytes. Technically this is 650 MB in SI terms, or approximately 620 MiB (since 650,000,000 ÷ 1,048,576 ≈ 620). However, the CD industry loosely used "MB" to mean 650 × 10⁶ bytes. Some media used 700 MB (≈ 668 MiB). This inconsistency is a classic example of the pre-IEC ambiguity.

Docker reports image sizes in decimal MB (e.g., "docker images" shows 150 MB), but the underlying layer storage on disk uses binary-aligned block sizes. A "150 MB" Docker image actually occupies roughly 143 MiB on disk before compression. Compressed layers further complicate things: a 150 MB uncompressed image might only transfer 50 MB over the network. Container registries like Docker Hub display compressed sizes, while "docker images" shows uncompressed — leading to frequent confusion in CI/CD pipeline size budgets.

One mebibyte (1,048,576 bytes) holds about: one minute of MP3 audio at 128 kbps (≈ 960 kB, so slightly under 1 MiB), a medium-resolution JPEG photo (0.5–2 MiB), about 200 pages of plain text, or the complete text of a short novel. A typical Linux kernel image at boot is 8–12 MiB compressed.

RAM chips are physically organized as binary address grids — each address line doubles the capacity, so sizes must be exact powers of 2 (4 GiB = 2³² bytes, 8 GiB = 2³³ bytes). USB flash drives use NAND flash that is also binary internally, but manufacturers reserve variable amounts for wear levelling, bad block management, and controller firmware. A "64 GB" USB drive might have 64 GiB of raw NAND but only expose 59.6 GiB (64 × 10⁹ ÷ 2³⁰) to the user — the label uses decimal marketing, unlike RAM which honestly reflects binary sizing.

Hard drive manufacturers measure 1 TB as 1,000,000,000,000 bytes (decimal). Windows displays storage in gibibytes (binary) but historically labelled them as "GB" — so 1,000,000,000,000 bytes ÷ 1,073,741,824 ≈ 931 GiB, which Windows displayed as "931 GB". macOS (since 10.6) correctly reports the same drive as "1 TB" using decimal GB. The drive is not lying; the OS was using a binary unit with a decimal label.

8 GB RAM is the current minimum for gaming; 16 GB is the recommended standard for most modern games at 1080p and 1440p; 32 GB benefits heavily multitasking systems or games with large open worlds. Memory-intensive tasks like video editing, 3D rendering, and running large language models locally typically require 32–64 GB or more.

A 4K movie in H.264 or H.265 encoding is typically 50–100 GB on Blu-ray; streaming services compress aggressively to 15–25 GB for 4K HDR content. Netflix's 4K streams average about 7 GB per hour; the downloaded version via the Netflix app for offline viewing is roughly 3–6 GB per hour at high quality settings.

1 GB of mobile data supports roughly: 2–3 hours of music streaming, 1 hour of HD video streaming, 2–3 hours of web browsing, or 30–60 minutes of video calling. Social media apps with autoplay video are heavy consumers — TikTok and Instagram Reels can use 300–600 MB per hour of active use.

AI model sizes vary enormously. GPT-2 (2019) is about 1.5 GB; Llama 2 7B is roughly 13 GB in float16 precision; Llama 2 70B is about 130 GB. GPT-4-class models are estimated at 500+ GB. Quantised (compressed) versions are smaller: a 4-bit quantised 7B model fits in about 4 GB, runnable on a modern laptop. Training requires far more — the training dataset, gradients, and optimizer states for a 70B model can occupy 1–2 TB of GPU memory across a cluster. The trend toward larger models is driving consumer GPU memory from 8 GB to 16–24 GB as a baseline for local AI inference.