Through the years, motherboards have had many types of connectors for attaching all types of computer hardware.

To make sense of all these connectors, let’s look at how these hardware components connect to the motherboard.

We’re first going to look at how the CPU connects to the motherboard.

The CPU connects to the motherboard through the CPU socket.

The most common socket for CPU’s is known as “zero insertion force” or ZIF.

This socket type keeps the contacts of the CPU and motherboard permanently together. It does so without adding any extra pressure on the great number of small pins that make up this connection.

Avoiding damaging any of these pins is important, so this socket has a plastic build and metal contacts designed for compatible CPUs to fit perfectly.

The socket also has a latch for keeping the CPU and motherboard connectors in place.

Depending on the manufacturer, the socket can have the male pins on the CPU or on the motherboard socket: The Pin Grid Array (PGA) configuration has the pins on the CPU and Land Grid Array (LGA) has the pins on motherboard’s socket.

Now, let’s look at RAM connectors…

Most motherboards have at least two of these next to the processor.

RAM is not only important for the PCs performance but also for its upgrade potential.

These connectors are normally known as DIMM slots.

Depending on the type of memory the motherboard supports, for example, DDR2, DDR3, or DDR4, these slots will have specific notches to match a specific type of RAM.

RAM is kept in place by a set of tabs at one or both ends of the connector on the motherboard.

With regards to hard drives, CD/DVD drives, or Blu-ray drives, most motherboards have at least a pair of SATA connectors.

These work for both hard disks and solid-state drives, otherwise known as SSDs.

The SATA cable, however, only transfers data and doesn’t give power to the drives. The drives need an extra four-pin cable that comes directly from the power supply.

SSDs, come in the 2.5-inch format as well as a form factor known as M.2, which can have a SATA or PCIe bus interface and connectors.

M.2 connectors can be connected to the PCIe slots with an adaptor, but today’s motherboards include a port for latching the drive parallel to the motherboard, attached to it by its connectors on one edge and by a screw on the other.

It is worth mentioning that for storage, before SATA was the IDE drive and its connector.

The IDE connector has a ribbon cable and requires its drives to be set as “master”, the booting drive, or “slave” through a set of 4 pairs of pins in the back of the drive known as jumpers.

Old floppy disks connectors are similar to the IDE channel, but a little shorter with 46 pin holes instead of 40.

Let’s move to the peripheral connectors.

The most common are the PCI or “Peripheral Component Interconnect” and PCI Express bus ports.

These are dedicated to adding extra hardware for better performance or for extra features.

Introduced in 1992, PCI was originally designed to replace ISA or “Industry Standard Architecture”.

While ISA hardware was limited by its 8 MHz clock speed and 8 and 16 bit bus, PCI initially featured a 32-bit bus and 33 MHz clock speed.

Later PCI models even came with a 64-bit bus and/or 66 MHz of clock speed. This was the case with the PCI-X bus.

This was around the time when videogames and other computer programs were becoming more and more dependent on graphic acceleration. The PCI bus architecture had limitations for dealing with that much data transfer.

As a result, the Accelerated Graphics Port or AGP was created in 1997.

While PCI had to make an extra step through the southbridge for reaching to the CPU, AGP communicated to the CPU through the northbridge and was able to texture map directly into RAM through Direct Memory Execute or DIME.

The AGP was also able to request multiple instructions at a time, while PCI could only request one, a technology known as pipelining.

PCI-X’s 64-bit bus and AGP’s special design for graphics would only last a few years.

The PCI Express bus, showed up in 2004. PCI, as well as PCI-X and AGP, rely on parallel communication. PCIe, like SATA, USB, and other devices, uses serial communication.

Now, parallel communication can transfer more bits at the same time than serial, which does so only one bit at the time. However, at high clock speeds parallel communication breaks down due to the electromagnetic interference it creates when many communication wires work at the same time.

Because PCI uses multiple wires for transferring data, the data doesn’t arrive at their destination at the same time.  This means the computer has to wait for all the bits of an instruction before being able to use that data.

Parallel communication also uses the same wire for both sending and receiving data. This is known as half-duplex.

Serial, on the other hand, has a full-duplex system, in which fewer wires are used more efficiently: two for sending data and another two for receiving it. This set of pairs is known as a lane. PCIe slots are classified by their number of lanes: 1x, 4x, 8x, or 16x.  16x is the most common for video cards.

PCIe also consumes less power and like SATA supports hot-swapping, meaning that it can be attached and detached from the motherboard while the computer is running.

Like AGP, most motherboard’s PCIe ports are directly connected to the CPU or northbridge.

In the past PCI ports were used for all types of hardware.

For example, we can find PCI NIC’s (Network Interface Controllers). NIC’s give computers the ability to connect to a network.

Notebook motherboards also have PCI ports.  They can be Mini-PCI, or PCIe cards which are mostly used for network cards.

Connection ports can also be expanded, as with SATA or USB expansion cards.

Riser cards are used by PCI connectors for setting expansion cards horizontally when they are too big to fit vertically into a computer.

Let’s now look at connectors that provide power to the motherboard.

For receiving electricity from the power supply, ATX motherboards have two sets of connectors: one with 24 pins and another of 4 or 8 pins. These lock into their ports by latches.

Some motherboards designed for multiple video cards can also have additional power connectors.

3 or 4 pin fan connectors are often located at the edges of the motherboard depending on the model, although one can also be found near to the CPU socket for its own fan. Water cooling pumps and LED strips with the right voltage can also use these connectors.

USB 2.0 and 3.0 headers have 9 and 19 pins respectively.

Next to the USB headers are the front panel or system panel connectors for power and reset buttons, as well as for frontal led lights.

Each of these has two pins depending on the motherboard model and can have a distinct color or name for identifying its position.

Next to them is the optical drive audio connector, a four-pin connector for CD-ROM drivers to play audio, and the front panel audio connector. There are also similar pins for connecting an internal motherboard speaker.

At the back of the PC case motherboards have many external connectors for input/output peripherals known as rear I/O: Here you can find USB and ethernet ports, jack connectors for speakers and microphones, as well as video ports such as VGA or HDMI.

Motherboards also offer legacy support by including PS/2 serial ports for keyboard and mouse connectors; some even have parallel ports for old printers.