Calculation Methodology
This page explains exactly how our converters work — the formulas, the constants, the precision rules, and the assumptions. We believe in total transparency, so you can verify our math against yours.
The Core Relationship
Every conversion on this site is built on two facts:
- 1 byte = 8 bits. This is exact, by definition.
- 1 Mbps = 1,000,000 bits per second. Data rates always use decimal (SI) prefixes, per IEC 80000-13 and universal industry convention.
From these, the core formulas follow:
Bytes/s = Mbps × 1,000,000 ÷ 8 = Mbps × 125,000Mbps = Bytes/s × 8 ÷ 1,000,000MB/s = Mbps ÷ 8(decimal megabytes)KB/s = Mbps × 125(decimal kilobytes)
SI vs IEC: How We Handle Both
"Kilobyte" and "megabyte" have two competing definitions, and our calculators support both explicitly instead of silently picking one:
| Unit | Standard | Value in Bytes |
|---|---|---|
| 1 KB (kilobyte) | SI / decimal | 1,000 B |
| 1 KiB (kibibyte) | IEC / binary | 1,024 B |
| 1 MB (megabyte) | SI / decimal | 1,000,000 B |
| 1 MiB (mebibyte) | IEC / binary | 1,048,576 B |
| 1 GB (gigabyte) | SI / decimal | 1,000,000,000 B |
| 1 GiB (gibibyte) | IEC / binary | 1,073,741,824 B |
Our calculators default to SI (decimal) because that is what ISPs, speed tests, and network equipment use. The IEC (binary) toggle is provided because operating systems and many applications report file sizes and transfer speeds in binary units. Megabits per second (Mbps) always use decimal bits regardless of the toggle — there is no "binary Mbps" in ISP marketing or networking practice.
Conversion Factors We Use
| From | To | Factor |
|---|---|---|
| 1 byte | bits | 8 (exact) |
| 1 Mbps | bits/s | 1,000,000 (exact) |
| 1 Mbps | Bytes/s | 125,000 (exact) |
| 1 Mbps | KB/s (SI) | 125 (exact) |
| 1 Mbps | MB/s (SI) | 0.125 (exact) |
| 1 Gbps | Mbps | 1,000 (exact) |
| 1 MiB/s | Mbps | 8.388608 (exact) |
Because bits, bytes, and SI prefixes are all defined exactly, every factor above is exact — there are no measurement approximations in data-rate conversion, only rounding choices in display.
Precision Standards
- Internal calculations use IEEE 754 double-precision floating-point numbers (15–17 significant digits).
- Displayed results round to the decimals setting you choose (default 2, adjustable 0–10).
- The hidden input field always carries at least 6 decimal places so round-trip conversions don't drift.
- Worked examples in articles round to 2–3 decimal places for readability, with rounding stated explicitly.
What Our Calculators Don't Model
Our converters compute the exact mathematical equivalence between units. They intentionally do not subtract real-world overhead — TCP/IP headers, retransmissions, Wi-Fi interference, server throttling — because overhead varies by network (typically 5–15%). Where overhead matters, our articles say so explicitly and show how to estimate it. Any future tool that applies an overhead assumption (like a download-time calculator) will label the assumption clearly and let you turn it off.
Sources
All definitions and factors are verified against:
- NIST — Prefixes for Binary Multiples
- IEC — Prefixes for binary multiples (IEC 80000-13)
- BIPM — The International System of Units (SI)
- Official platform documentation for real-world figures (e.g., streaming bandwidth requirements published by Netflix and YouTube)
Found an Error?
If our methodology differs from your calculation and you think we're wrong, please tell us via the contact page. We'll investigate within 48 hours — see our Editorial Standards for the full accuracy promise.