hardwarea-plusBeginner22 min read

What Is Peripheral Component Interconnect in Computer Hardware?

Also known as: Peripheral Component Interconnect, PCI, PCI slot, expansion slot, CompTIA A+ hardware

Reviewed byJohnson Ajibi· Senior Network & Security Engineer · MSc IT Security
On This Page

Quick Definition

Think of your computer motherboard as a city with main roads. PCI is a standard type of road that allows different devices, like a new Wi-Fi card or a graphics card, to connect and talk to the main computer brain (the CPU) and memory. It was the standard way to add extra capabilities to a desktop computer for many years, before being largely replaced by PCI Express.

Must Know for Exams

The Peripheral Component Interconnect (PCI) is a specific exam objective for the CompTIA A+ 220-1101 (Core 1) exam, under the domain '1.2 Motherboards'. The exam expects you to identify PCI slots based on their physical appearance compared to other expansion slots like PCIe x1, x4, x8, x16, and older standards like AGP. You will need to recognise that PCI slots are typically the same length (for 32-bit slots, about 85mm), older than PCIe, and are less common on modern motherboards.

The exam tests your knowledge of characteristics, not just visual identification. You must know that standard PCI operates at 33 MHz or 66 MHz, is a parallel bus, and has a maximum transfer rate of 133 MB/s for a 32-bit slot. This is often contrasted with PCIe, which uses serial lanes and has much higher speeds. The exam will ask which standard a specific network card or sound card likely uses based on the age of the system.

Furthermore, the A+ exam tests your ability to select the correct expansion slot for a given scenario. For example, a question might describe a technician installing a legacy parallel port card into an older server, and you must identify the PCI slot as the correct choice. The exam will not ask you to configure PCI settings, but it will test your knowledge of the slot's role as a 'legacy' standard. You must understand that while modern graphics cards need a PCIe x16 slot, legacy devices like TV tuner cards or some wireless networking cards from the mid-2000s would use PCI. The CompTIA A+ exam objectives explicitly list 'PCI' as a motherboard expansion slot type, making it a definite target for one or two questions on the exam.

Simple Meaning

Imagine your computer is a highly organised office building. The motherboard is the main floor plan. The CPU is the CEO in the corner office, and the RAM is the filing cabinet with all the current project files. Now, what if the CEO needs a new, specialised machine, like a high-speed printer for blueprints or a dedicated video conferencing unit? You wouldn't want to rebuild the whole office building. Instead, you have a standardised, empty space on the main floor—a 'slot'—that has a direct connection to the CEO's office and the filing cabinets. This slot works on a standard set of rules (a protocol), so any new machine that follows those rules can be plugged in, and the CEO instantly knows how to talk to it. That slot and its rulebook is the Peripheral Component Interconnect (PCI) standard.

In more technical but still plain terms, a computer's brain needs a way to talk to all the different parts that do specific jobs—like showing pictures on a screen (graphics card), connecting to the internet (network card), or playing sound (sound card). The PCI bus provides a common, high-speed pathway for this communication. It uses a shared set of wires (the bus) that connects the CPU's memory controller to the expansion slots. When you plug a PCI card into the slot, it automatically connects to this shared pathway. The computer's BIOS (Basic Input/Output System) or operating system then identifies the card, assigns it resources like memory addresses and interrupt requests (IRQs), and allows the software to use the hardware. This standardisation was a huge leap forward because it meant you could buy a network card from any manufacturer, plug it into any computer with a PCI slot, and it would work, without needing to manually set tiny switches or jumpers on the card. Before PCI, adding hardware was a complicated and often frustrating process of configuration.

Full Technical Definition

Peripheral Component Interconnect (PCI) is a local computer bus for attaching hardware devices in a computer. Developed by Intel in 1992, PCI became the industry standard for connecting expansion cards directly to the motherboard, replacing older buses like ISA and VESA Local Bus. The PCI specification defines both the physical slot (connector) and the electrical characteristics of the bus, including its protocols for data transfer, addressing, and device configuration.

A PCI bus consists of a set of parallel data lines (32-bit or 64-bit wide) and control lines. The bus is clocked at 33 MHz (standard) or 66 MHz (for PCI 2.1 and later). For a 32-bit, 33 MHz bus, the theoretical maximum data transfer rate is 133 MB/s. The bus is shared, meaning all devices connected to it must take turns communicating. The PCI specification supports bus mastering, which allows a PCI device to take control of the bus and transfer data directly to system memory without relying on the CPU, significantly improving performance. This direct memory access (DMA) is handled through bus mastering, not the original DMA controller on the motherboard.

Configuration space is a key innovation in PCI. Each PCI device has a 256-byte configuration space that the system's BIOS or operating system reads during boot. This space contains a predefined header with fields for Vendor ID, Device ID, Class Code (e.g., network controller, display controller), and Base Address Registers (BARs). The BARs are used to request memory or I/O address ranges. The system's plug-and-play software uses this configuration space to automatically assign resources such as interrupts and memory addresses without user intervention. This eliminated the manual jumper and DIP switch settings required by earlier ISA cards.

PCI uses a parallel bus architecture. While effective, this design has limitations. Multiple devices share the same electrical bus, and the bus frequency cannot easily be increased due to signal integrity issues (crosstalk, reflections, and timing skew). These limitations eventually led to the development of PCI Express (PCIe), a serial point-to-point interconnect that differs fundamentally from PCI. Despite being largely superseded by PCIe in modern systems, PCI remains important for legacy hardware and is still tested on the CompTIA A+ certification, particularly in the context of identifying expansion slots and understanding older system architectures.

Real-Life Example

Imagine a large, busy post office. The processing room in the back is the CPU and memory. The sorting tables are the PCI slots. The main conveyor belt running past all the tables is the PCI bus. Before the post office had a standard conveyor belt, each sorting table had its own unique, manual path to the processing room. A worker from the 'graphics' table had to hand-carry every envelope to the processing room, often waiting in line behind workers from the 'sound' table or the 'network' table. There were no standard rules, and if a new type of sorting table was added (a new expansion card), they had to build a completely new path to the processing room, which was slow and complex.

PCI changed this. The post office installed a standard, high-speed conveyor belt (the PCI bus). Every new sorting table is built with a standard interface (the PCI slot) that connects directly to this conveyor belt. Now, when a 'network card' table needs to send an envelope (data packet) to the processing room, they just place it on the conveyor belt. The belt uses a set of standard rules (the PCI protocol) to deliver the envelope directly to the right department in the processing room, and it can even bring a reply back without a worker having to manually fetch it. Furthermore, the post office manager (the BIOS) has a standard checklist for every new table. When a new 'sound card' table is added, the manager fills out a form (the configuration space) detailing what the table does, how many envelopes it needs to send at once, and where it should put its outgoing mail (memory addresses). This automatic setup happens instantly, without anyone on the floor needing to manually assign paths.

Why This Term Matters

Understanding PCI is crucial for any IT professional working with older, legacy systems, or even just troubleshooting compatibility. Many businesses run critical applications on hardware that is years or even decades old, and that hardware almost certainly uses PCI slots. You will encounter PCI-based devices in industrial control systems, point-of-sale (POS) terminals, embedded systems, and server management cards. Knowing how to identify a PCI slot versus an ISA or PCIe slot is a fundamental hardware skill.

From a troubleshooting perspective, understanding the shared bus nature of PCI helps you diagnose resource conflicts. While Windows has improved plug-and-play handling, issues with interrupt sharing (IRQ conflicts) can still occur in older systems. If you install a new PCI card and another unrelated device stops working, you suspect an address or interrupt conflict, a problem directly related to how the PCI bus assigns resources. Knowing that PCI devices use 256 bytes of configuration space makes it easier to understand tools like 'Device Manager' in Windows

Finally, PCI is the direct ancestor of PCI Express (PCIe), the standard found in every modern computer. Many of the concepts from PCI, including device enumeration, configuration space, class codes, and the idea of a hierarchical bus structure, were carried over to PCIe. Learning PCI provides a foundational understanding of how motherboards communicate with expansion cards. When you study PCIe, you will see the evolution from a shared parallel bus to a dedicated serial point-to-point connection, but the basic concepts of address allocation and device configuration remain very similar. This makes learning PCI a valuable stepping stone to understanding modern computer architecture for the A+ exam and beyond.

How It Appears in Exam Questions

In the CompTIA A+ exam, questions about PCI primarily fall into three categories: identification, compatibility, and historical context.

Identification and Comparison Questions: These are the most common. You might see an image of a motherboard with several different slots. The question will ask, 'Which of the following slots is a standard 32-bit PCI slot?' and you must visually identify it among PCIe x1, PCIe x16, and possibly an older ISA or AGP slot. Alternatively, the question will provide a description like, 'A technician is looking at a 33 MHz expansion slot that has a single notch in the connector. Which standard is this?' You must select PCI.

Compatibility and Selection Questions: These questions present a scenario. For example: 'A user has a legacy sound card with a 32-bit connector. The card needs to be installed in a computer that has only PCIe slots available. Which of the following is the best course of action?' The correct answer would be to use a PCI to PCIe adapter (or recommend a new PCIe sound card), highlighting your understanding that you cannot directly plug a PCI card into a PCIe slot. Another question might ask, 'A technician is building a system for a specific industrial control card. The manual specifies a PCI bus. Which of the following motherboard characteristics must be verified?' The correct answer would be that the motherboard has an available standard PCI slot.

Troubleshooting Questions: These questions are less common but test deeper understanding. A scenario might be: 'After installing a new PCI network card, a user reports that the existing PCI sound card is no longer functioning. What is the most likely cause?' The answer would be an IRQ or memory address conflict on the shared PCI bus. This tests your knowledge that PCI is a shared bus where resources must be managed. Finally, questions about data transfer speeds are common: 'What is the theoretical maximum throughput of a standard 32-bit, 33 MHz PCI slot?' with the answer being 133 MB/s.

Practise Peripheral Component Interconnect Questions

Test your understanding with exam-style practice questions.

Practise

Example Scenario

A local library uses an older computer system attached to a self-checkout kiosk. The kiosk uses a special parallel port card to communicate with the scanner. The library's current computer is old and running Windows 7. It is starting to fail, and they purchase a used but newer desktop computer to replace it.

The technician, Sarah, opens the newer computer to install the parallel port card. She sees the motherboard has several expansion slots: one long black slot with a single notch (PCIe x16), two short black slots with a single notch (PCIe x1), and three longer white or beige slots with two segments (the PCI slots). Sarah identifies the white PCI slots as the only option for the old parallel port card. She slots the card into an available PCI slot, secures it with a screw, and closes the case. When she boots the computer, Windows detects the new hardware and installs the drivers.

This scenario applies PCI because Sarah had to identify the correct expansion slot for a legacy device. The library's new computer might have been a late-model motherboard that still included a few PCI slots for exactly this purpose—backward compatibility. Sshe did not need to configure anything manually; the PCI bus's plug-and-play system handled the resource assignment. This demonstrates the real-world need for a technician to be able to identify PCI slots and understand their purpose for installing legacy hardware.

Common Mistakes

Confusing PCI with PCIe slots based on speed.

PCIe is a much faster, serial point-to-point standard. PCI is a shared parallel bus. Saying 'PCI is usually faster than PCIe' is the opposite of the truth. PCIe is the successor to PCI and completely different in architecture.

Remember: PCI = older, slower, parallel, shared bus. PCIe = newer, faster, serial, dedicated lanes.

Thinking all white expansion slots on a motherboard are PCI slots.

While many PCI slots were white or beige, PCIe x1 and x16 slots are also commonly found in black, white, or blue. Colour is not a reliable identifier. The key identifier is the physical length of the slot and the number of notches (segments) in the connector.

Learn the physical appearance. A standard 32-bit PCI slot has 120 pins arranged in two segments with a single notch. It is about 85mm long. Use the notch and length, not the colour, to identify it.

Believing PCI supports hot-swapping (like USB).

Standard PCI does not support hot-swapping. You must power down the computer completely before installing or removing a PCI card. Some specialised PCI standards (like CompactPCI) support hot-swap, but the standard desktop PCI does not.

Always turn off and unplug the computer before working on any standard PCI card. This is critical safety and hardware practice.

Assuming a clear audio or graphics card will only fit in a PCIe slot and ignoring PCI.

While modern audio and video cards are almost exclusively PCIe, many legacy cards (especially Sound Blaster or older Wi-Fi cards) were designed for the PCI bus. If a motherboard has PCI slots, one of these legacy cards might be the solution for a specific compatibility need.

Always check the card's connector type before installation. Count the pins and note the notch position. If it matches the PCI slot standard, it will work in a PCI slot.

Thinking PCI can achieve speeds close to modern standards.

The theoretical maximum for a 32-bit, 33 MHz PCI bus is 133 MB/s. A single PCIe 3.0 x1 lane has a throughput of about 1 GB/s. Modern PCIe 4.0 x16 slots can achieve 32 GB/s. PCI is a legacy standard with limited bandwidth.

Understand that PCI is an old standard. It is only relevant for compatibility tasks and is not used for high-performance devices like modern SSDs or GPU cards.

Exam Trap — Don't Get Fooled

The exam might show a picture of a motherboard with both a PCI slot and multiple PCIe x1 slots, and ask which slot you should use for a standard 32-bit sound card. An inexperienced learner might choose a PCIe x1 slot because it looks similar or is more modern. Memorise the physical keying of each slot.

A standard 32-bit PCI card will not physically fit into a PCIe x1 slot because the notch is in a different location and the slot length is shorter. If you are unsure, remember that a standard 32-bit PCI slot is always longer than a PCIe x1 slot (about 85mm vs. 25mm).

You cannot force a PCI card into a PCIe slot.

Commonly Confused With

Peripheral Component InterconnectvsPCI Express (PCIe)

PCIe is a completely different, much faster serial bus standard that replaced PCI. While PCI is a shared parallel bus, PCIe uses point-to-point serial lanes. They are not physically or electrically compatible. A PCI card cannot go into a PCIe slot and vice versa.

Think of PCI as a single-lane road that all traffic must share (parallel). PCIe is like a multi-lane highway where each device gets its own dedicated lane (serial). A car (a PCI card) designed for the single-lane road is too wide and has the wrong type of tyres for the highway, so it won't fit.

Peripheral Component InterconnectvsIndustry Standard Architecture (ISA)

ISA was the much older expansion standard that PCI replaced. ISA slots are typically black and much longer than PCI slots. ISA slots required manual configuration of resources (jumpers), while PCI introduced automatic plug-and-play configuration.

ISA is like an old manual typewriter that requires you to manually set the paper margins and ribbon. PCI is like a modern printer that auto-configures from your computer. They look different and are not compatible.

Peripheral Component InterconnectvsAccelerated Graphics Port (AGP)

AGP was a high-speed slot specifically dedicated to graphics cards, designed around the same time as PCI. It is physically different from PCI, often brown in colour, and located closer to the CPU. While PCI was a general-purpose bus, AGP was a dedicated point-to-point connection for video only.

AGP is like a dedicated express train line from the graphics department to the main office, bypassing the general city bus system (PCI). It served a specific purpose (graphics) that PCI was too slow for at the time.

Step-by-Step Breakdown

1

System Power-On and Reset

When the computer is powered on, the motherboard's chipset (specifically the reset logic) holds the PCI bus in a reset state. All PCI devices are inactive. This ensures a known, clean starting point before any configuration begins.

2

Device Enumeration by the BIOS

The BIOS or UEFI firmware takes control. It starts a process called 'PCI bus enumeration'. It provides the PCI bus with a clock signal and begins to probe each possible device slot number. It does this by trying to read the Vendor ID field in the Configuration Space of each device. If a device is present, the read is successful, and the BIOS discovers the device.

3

Configuration Space Reading

For each discovered device, the BIOS reads the 256-byte Configuration Space. This reveals critical information: Vendor ID (which company made it), Device ID (the specific model), Class Code (e.g., 'Network controller'), and the contents of the Base Address Registers (BARs). The BARs tell the system how much memory or I/O space the device needs.

4

Resource Assignment

The BIOS (or later, the operating system) assigns resources to each device. This includes memory addresses (mapped from the BAR requests), I/O addresses, and an interrupt request line (IRQ). The system keeps a table of assigned resources to avoid conflicts. It then writes these assignment values back into the device's Configuration Space registers.

5

Device Activation and Bus Mastering

Once the device has its resources, the system sets the 'Command Register' in the Configuration Space to enable bus mastering and memory/IO access. The device is now active. It can communicate with the CPU and system memory. It can also request control of the PCI bus (bus mastering) to perform direct memory access (DMA) transfers, moving data between the device and RAM without CPU intervention.

6

Normal Operation and Arbitration

When multiple devices want to use the PCI bus at the same time, an arbiter (part of the chipset) decides who goes first. A device signals its request, the arbiter grants control to one device at a time, the device completes its data transfer, and then the bus is released for the next device. This is the 'shared bus' nature of PCI in action.

Practical Mini-Lesson

As an IT professional, your practical interaction with PCI is likely limited to identification, installation, and legacy troubleshooting.

Identifying a PCI Slot: On a motherboard, look for an expansion slot that is approximately 85 millimetres long (about 3.4 inches). The connector is divided into two sections: a shorter front section (usually with 60 pins) and a longer back section (with 64 pins), separated by a single notch. The slot itself is typically white or beige, though colours vary. It is located parallel to other expansion slots. A key point: it is physically longer than a PCIe x1 slot (which is about 25mm) but shorter than most PCIe x16 slots (which are about 89mm but have a different notch pattern).

Installing a PCI Card: This is a simple but careful process. First, completely power down the computer and unplug the power cord. Press the power button to discharge any residual power in the capacitors. Open the case and locate an empty PCI slot. Remove the corresponding metal bracket from the back of the case. Align the PCI card's gold-edge connector with the slot, ensuring the notch on the card matches the notch in the slot. Press down firmly and evenly until the card is fully seated. Secure the card with the screw you removed. Close the case, plug the computer back in, and power it on. The system should automatically detect the new hardware.

Troubleshooting Resource Conflicts: This is the most challenging aspect. If after installing a PCI card another device stops working, or the new card doesn't work, a resource conflict is the prime suspect. In Windows, open Device Manager. If you see a yellow exclamation mark next to any device, right-click it, select Properties, and look at the 'Resources' tab. You may see an 'Input/Output Range' or 'Interrupt Request' that is listed as 'Conflicting device list'. The solution often involves manually changing the resources: uncheck 'Use automatic settings', select a different 'Basic configuration' from the drop-down, or manually change the IRQ or I/O range. If this fails, try moving the PCI card to a different PCI slot to force the BIOS to assign new resources on the next boot. If a legacy device requires a specific IRQ (like IRQ 7 for a parallel port), you may need to enter the BIOS and reserve that IRQ for legacy ISA devices, preventing PCI from using it. This manual configuration is rare on modern systems with good plug-and-play, but it is a classic troubleshooting skill for legacy hardware.

Memory Tip

Remember the 'P' in PCI stands for 'Parallel', which is a clue to its shared, single-channel nature. Later, the 'e' in PCIe stands for 'Express', linking to 'Express lane'—a fast, dedicated connection.

Covered in These Exams

Current Exam Context

Current exam versions that test this topic — use these objectives when studying.

Related Glossary Terms

Frequently Asked Questions

Can I install a PCI card into a PCIe slot?

No, you cannot. They are physically and electrically incompatible. The notches on the card and the slot are in different positions, so the card will not fit.

What is the maximum speed of a standard PCI slot?

A standard 32-bit, 33 MHz PCI slot has a theoretical maximum throughput of 133 MB/s. This is much slower than modern PCIe slots.

Is PCI still used in modern computers?

It is very rare on modern consumer motherboards. You might find one or two PCI slots on some professional or industrial motherboards for backward compatibility with legacy hardware.

What does 'plug-and-play' mean in the context of PCI?

It means that when you install a PCI card, the system automatically detects it, assigns resources (memory, I/O, IRQ), and loads the correct drivers without you needing to configure jumpers or DIP switches.

What is the purpose of the notch in the PCI slot?

The notch is a physical key. It prevents you from inserting a card designed for a different standard (like ISA or PCIe) into the PCI slot, and it ensures the card is inserted correctly.

Do I need to shut down my computer to install a PCI card?

Yes, you must completely shut down the computer and unplug it from the power source before installing or removing a PCI card. PCI does not support hot-swapping like USB or modern PCIe does in some implementations.

What is PCI bus mastering?

PCI bus mastering is a feature that allows a PCI device (like a network card) to take control of the bus and transfer data directly to system memory without involving the CPU, improving overall system performance.

What does '32-bit PCI' mean?

It refers to the width of the data bus. A 32-bit PCI bus can transfer 32 bits of data per clock cycle. There is also a 64-bit PCI version, which is longer and used mostly in servers.

Summary

Peripheral Component Interconnect (PCI) is a foundational hardware standard that defined how expansion cards were added to computers for over a decade. It introduced automatic configuration through its 256-byte configuration space and plug-and-play capabilities, a huge step forward from earlier manual standards. While it has been almost entirely replaced by the much faster, serial PCI Express (PCIe) standard, understanding PCI remains important for CompTIA A+ certification.

You must be able to identify a standard 32-bit, 33 MHz PCI slot by its physical appearance (85mm length, single notch) and distinguish it from other slots. You should know its key characteristics: a shared parallel bus, a maximum transfer rate of 133 MB/s, and the need for manual troubleshooting of resource conflicts in older systems. For the exam, focus on visual identification and the scenarios where you would use a PCI slot for legacy hardware compatibility.

Remember that PCI is a legacy technology, but its concepts provided the foundation for the modern computer architecture that runs nearly every device today.