Computer Architecture
Word, Byte and Capacity; Von Neumann vs Harvard; CISC vs RISC
C-CAT
Word, Byte and Capacity
13.1 Bit Groupings
| Name | Size | Notes |
|---|---|---|
| Bit | 1 binary digit | 0 or 1 |
| Nibble | 4 bits | One hex digit |
| Byte | 8 bits | Standard addressable unit in many systems |
| Word | CPU-dependent (16, 32, 64 bits) | "Natural" operand size |
A memory word is a group of bits stored together — may represent a number, opcode or characters.
13.2 Binary Prefixes (Computer Science)
| Symbol | Value | Power |
|---|---|---|
| K (Kilo) | 1024 | 2¹⁰ |
| M (Mega) | 1024 K | 2²⁰ |
| G (Giga) | 1024 M | 2³⁰ |
| T (Tera) | 1024 G | 2⁴⁰ |
Example conversions:
- 1 KB = 1024 bytes = 2¹⁰ bytes
- 1 MB = 1024 KB = 2²⁰ bytes
- 1 GB = 1024 MB = 2³⁰ bytes
13.3 Word vs Byte Examples
| Word size | Bytes per word |
|---|---|
| 16-bit | 2 |
| 32-bit | 4 |
| 64-bit | 8 |
Von Neumann vs Harvard
14.1 Von Neumann Architecture
- Single shared memory for both program and data.
- Single bus (typically) for memory access.
- Components: ALU, CU, one memory space, I/O.
- Cannot fetch instruction and data simultaneously on a single bus (von Neumann bottleneck).
- Examples: 8085, 8086, 80386, 80486, most early PCs.
┌──────────┐
│ CPU │
└────┬─────┘
│ one bus
┌────▼─────┐
│ Program │
│ + Data │ (same memory)
└──────────┘
14.2 Harvard Architecture
- Separate memories (and often separate buses) for program and data.
Allows simultaneous instruction fetch and data access.
- Examples: ARM (many cores), AVR, PIC, SHARC, Sun SPARC (Harvard-style paths).
┌──────────┐ Program bus ┌─────────────┐
│ CPU │───────────────────▶│ Program Mem │
└────┬─────┘ └─────────────┘
│ Data bus
└──────────────────────────▶ Data Mem
14.3 Modified Harvard
Many modern chips use Harvard internally (split L1 caches) but von Neumann externally — exams may call this modified Harvard.
CISC vs RISC
| Feature | CISC | RISC |
|---|---|---|
| Full name | Complex Instruction Set Computer | Reduced Instruction Set Computer |
| Instruction count | Large, variable format | Small, fixed format |
| Operand location | Often memory-to-memory | Register-to-register |
| Cycles per instruction | Often multiple | Often one |
| Control unit | Often microprogrammed | Hardwired, simpler |
| Examples | x86 (80386, 80486), VAX | ARM, AVR, PowerPC, MIPS |
| Memory access | Many instructions hit memory | MOV style loads/stores; ALU uses registers |
15.1 CISC Characteristics (Detail)
- One mnemonic may do much work (e.g.
LDAwith addressing modes). - Compiler can emit fewer instructions but each may be slower.
- Microcode ROM inside CPU interprets complex opcodes.
15.2 RISC Characteristics (Detail)
- Simple instructions execute quickly; pipelining is natural.
- Highly pipelined superscalar RISC cores dominate mobile and server (ARM, RISC-V).
- Memory access often
limited to explicit load/store (
LDA,STA,PUSH,POPon 8085 are CISC-style memory ops).
"RISC uses microprogrammed control" is usually false — that's CISC territory.
Continue learning
Related notes
Definition of AI; Need of AI
Artificial Intelligence
Introduction to Data Engineering; Big Data — The 5 V's; Types of Data
Big Data and Data Engineering
Introduction to C Programming; C Program Structure; Data Types and Variables
C Programming
What Is a Computer?; Machine Cycle: Fetch–Decode–Execute; CPU Organization
Computer Architecture
Put this topic into timed practice
Open mock tests when you want full-exam pacing, or keep drilling in practice mode.