In the field of embedded engineering, selecting the appropriate controller is a critical decision that directly impacts system performance, cost, and design footprint. A microcontroller (commonly designated as an MCU, MC, UC, or μC) is essentially a small computer integrated onto a single Very Large Scale Integration (VLSI) integrated circuit chip. Unlike general-purpose computers, a microcontroller is typically dedicated to a single, specific task and runs a dedicated program stored in its read-only memory.
To help developers choose cost-effective solutions for specific applications, manufacturers provide diverse families of devices. These devices can be categorized based on their underlying memory architecture, vendor ecosystem, bus width, and memory configurations.
1. Classification by Memory Architecture
How a microcontroller's Central Processing Unit (CPU) accesses its program instructions and variable data defines its core architecture. On the basis of memory architecture, microcontrollers are divided into two primary types:
- Harvard Architecture: This layout utilizes completely separate memory spaces and hardware buses for program code (ROM) and data (RAM). This design enables the CPU to fetch instructions and read/write data simultaneously, significantly increasing processing speeds. The AVR microcontroller family is a prominent example of a Harvard architecture.
- Von Neumann Architecture: Also known as the Princeton architecture, this design uses a single, shared memory space and a common bus for both instructions and data. Because code and data cannot be accessed at the same time, Von Neumann systems require more machine cycles to execute instructions but feature a simpler physical chip layout.
2. Classification by Manufacturer / Vendor Ecosystem
Microcontrollers are heavily categorized by their service providers and silicon vendors. Each manufacturer offers a distinct instruction set, development environment, and hardware peripheral options:
- Intel: Pioneer of the highly enduring 8051 family, a classic CISC-based 8-bit architecture widely used in basic control systems.
- Atmel: Famous for the AVR family (including the Tiny, Mega, Classic, and Special Purpose lines), which are high-speed 8-bit RISC microcontrollers featuring on-chip Flash memory.
- Microchip: Manufacturer of the widely used PIC family of microcontrollers.
- Motorola / Freescale: Known for producing the HCS08 family and historical industry-standard chips like the Motorola 6811.
- Zilog: Manufacturer of the Z8 microcontroller series.
- Hitachi: Another major provider of dedicated microcontroller architectures.
3. Classification by Bus Width (Word Size)
The "word size" or bus width determines the maximum size of data that the CPU can process in a single instruction cycle.
- 8-Bit Microcontrollers: The absolute predominant family of microcontrollers. An 8-bit CPU works on only 8 bits of data at a time; larger data values must be broken down into 8-bit pieces to be processed. Despite this limitation, the single-byte word size is perfectly sufficient for the vast majority of embedded tasks, easily interfaces with standard logic circuitry, and aligns perfectly with byte-sized serial ASCII data communications. Low-end 8-bit chips are highly cost-effective and can integrate as little as 1000 bytes of ROM and 20 bytes of RAM on-chip.
- 16-Bit Microcontrollers: These devices provide a substantial jump in processing capabilities. An example is the Philips eXtended Architecture (XA) range, which features a 16-bit fully static CPU, a 24-bit program and data address range, and an enhanced instruction set optimized for multitasking and high-level programming language support.
- 32-Bit Microcontrollers: High-performance devices, such as the AVR32 family, designed to handle intensive data processing without the overhead of slicing data into smaller byte segments.
4. Classification by Onboard Memory Configuration
Because using a generic, "one-size-fits-all" device for all applications would introduce unnecessary complexity and cost, manufacturers classify family members by how their program memory is physically constructed:
- ROMless Devices: These microcontrollers contain no on-chip read-only memory. They rely entirely on external memory expansion via off-chip ROM to load and execute their instructions.
- ROM-Based Devices: These integrated circuits feature factory-programmed, non-volatile read-only memory directly on the silicon to hold permanent code.
- EPROM / One-Time Programmable (OTP) Devices: These utilize electrically programmable ROM, allowing developers to program the chip. OTP variants cannot be erased once programmed, while standard EPROMs require specialized ultraviolet exposure to clear.
- Flash Memory (FEEPROM) Devices: One of the most significant advancements in embedded design. These feature electrically erasable on-chip Flash memory that can be cleared and rewritten in seconds. This makes them highly reusable, cost-effective, and ideal for rapid product prototyping and iterative development.
Conclusion
Whether designing a simple microwave oven requiring low-cost 8-bit processing or a complex real-time industrial application, understanding the classifications of microcontrollers is essential. By evaluating devices based on their memory architectures, bit widths, and physical packaging options, system designers can successfully minimize operational power draw and overall hardware costs.
For all 2026 published articles list: click here:
…till the next post, bye-bye & take care