MB, MiB and the missing megabytes

One file, two labels

Our 1,500 kbps test export contains exactly 1,630,205 bytes. Divide by 1,000,000 and it is 1.630 MB. Divide by 1,048,576 and it is 1.555 MiB. The file has not lost data. The divisor has changed.

The capital B matters too

Video rates normally use bits per second. File sizes use bytes. Eight bits make one byte. A 10 Mbps connection carries at most 1.25 decimal MB per second before protocol overhead; it does not carry 10 MB each second. In this site, kbps means 1,000 bits per second, and MB means 1,000,000 bytes.

Where binary limits really bite

Storage makers state capacity in decimal units. A file manager that divides by 1,024 but still prints “GB” shows a 1 TB drive, 10¹² bytes, as about 931.3 GB. Nothing is missing; the divisor changed again.

File systems can impose a real binary limit. Microsoft’s comparison table lists the largest FAT32 file as 4 GiB, about 4.29 decimal GB, while exFAT and NTFS allow far larger files. A long recording copied to a FAT32 card or drive can fail at that size even when free space remains. Check the format before you blame the export.

Rates before protocol overhead. A connection rate in megabits divided by eight is the most decimal megabytes it can move each second.

Which one should you enter?

1. If a service documents its limit in bytes, use those bytes as the source of truth.
2. If your file manager shows a binary figure, multiply MiB by 1.048576 to enter decimal MB in the calculator.
3. If the label is ambiguous, inspect the exact bytes in file properties and leave room below the cap.

The calculator always labels decimal MB and GB. Our experiment tables also include the byte count so you can check the conversion independently.