The formatted disk capacity is always less than the "raw", unformatted capacity specified by the disk's manufacturer, because some portion of each track is used for sector identification and for gaps (empty spaces) between sectors and at the end of the track.
Disks are either hard sectored or soft sectored. Hard sectoring was at first preferred because a less-sophisticated disk controller can use it.1 In hard sectored 8-inch and 5+1⁄4-inch formats, each track is divided into a particular number of sectors determined when the disk is manufactured. Holes are punched in the magnetic media to indicate where each sector should start (in an area closer to the center of the disk than is used for magnetic recording). An additional hole is punched near one of the sector holes to identify the start of the track. A sensor in the drive detects the holes in the magnetic media as they align with a hole in the disk jacket.
Data is generally written at a fixed number of bits per second, with only a very small percentage of variation due to component tolerances, so given the nominal speed that the disk rotates it is possible to calculate the number of degrees a given number of bytes will occupy when written. In practice the motor speed varies, especially from one drive to another, resulting in those bytes occupying more degrees of the track at high motor speeds or fewer degrees at low motor speeds.
Improved technology removed hard sectoring's cost advantage, and most computers use soft sectoring.2 When a soft sectored disk is low-level "formatted", each track is written with a number of bytes calculated to fit within 360 degrees at the highest expected motor speed. Special bit patterns are written right before the location where a sector should start, and serve as identifiers, similar to the punched holes used by hard sectored disks. Thus, the full constellation of punched holes is not needed, and only a single hole is retained, to indicate the start of the track (3+1⁄2-inch disks use an alignment pin rather than a hole). If the motor is spinning any slower than the highest acceptable speed, which is usually the case, the data will fit in fewer than 360 degrees, resulting in a gap at the end of the track. Additionally, if a sector were to be rewritten on a drive running faster than the drive was running when the track was formatted, the new data would be larger (occupy more degrees of rotation) than the original sector. Therefore, during formatting a gap must be left between sectors to allow a rewritten sector to be larger without overwriting the following sector.
Commodore's Amiga used an unusual format which got closer to the disk's raw (unformatted) capacity by eliminating the gaps between sectors and simplifying the identification data. This meant that individual sectors could not be rewritten; the Amiga instead simply rewrote the entire track.
Formatted capacity numbers are based on the resulting number of logical sectors and the byte payload they can carry; that is, they depend on the physical parameters and modulation, but are independent of a particular file system. Sometimes floppies are superformatted to use more tracks or sectors per track to accommodate slightly more data. Some floppy-based Linux distributions utilize such techniques. For comparison purposes, formatted capacities given in this section assume standard disk geometries as they are supported by common operating systems in their default configuration.
The maximum usable capacity is file system and configuration specific and always lower than the formatted capacity, since the file system occupies a number of the available sectors for control structures as well.
Most floppy disks used by PCs use the FAT12 file system format, which imposes certain practical defaults on the logical geometry in order to be recognizable by all operating systems. Sometimes disks may use a more exotic file system.
SSDD originally referred to single sided, double density, a format of (usually 5+1⁄4-inch) floppy disks which could typically hold 35 to 40 tracks of nine 512-byte, or 18 256-byte, sectors each. Only one side of the disc was used, although some users did discover that punching additional notches into the disc jacket would allow the creation of a flippy disk, which could be manually turned over to store additional data on the reverse side.
Single-sided disks began to become obsolete after the introduction of IBM PC DOS 1.1 in 1982, which added support for double-side diskette drives with a capacity of 320 KiB to the IBM 5150 PC. In 1983, PC DOS 2.0 pushed the formatted capacity to 180 KiB single-sided or 360 KiB double-sided by using nine, instead of eight, sectors per track.
Main article: List of floppy disk formats
Throughout the 1970s and 1980s, many different disk formats were used, depending on the hardware platform. Variables included the size of media (nominal 8-inch, 5+1⁄4-inch, 3+1⁄2-inch and others), the encoding of data on the media (FM, MFM, M²FM, GCR), the number of disk tracks, single or double sided, the number of sectors in each track, and hard or soft sectoring. Even media that was physically interchangeable between two systems might not have been usable owing to particulars of the arrangement of data on the disk.
Thompson, Lester E. (September 1984). "Floppy-Disk Formats / Shedding some light on tracks, sectors, fields, and encoding schemes". BYTE. p. 145. Retrieved 2025-04-10. https://archive.org/details/BYTE_Vol_09-10_1984-09_Computer_Graphics/page/n155/mode/1up?view=theater ↩