Previous | Tutorial index | Next

Tutorial 5: Evolution of Modern Computer Construction

Learning Objective

Explain how the construction of modern computers has evolved.

5.1 Introduction: The Incredible Shrinking Computer

If you held a modern smartphone in one hand and stood next to the ENIAC—a machine that weighed 30 tons and filled a room—you would be holding more computing power in your palm than existed in the entire world in 1945. This transformation did not happen by accident. It is the result of a relentless series of technological breakthroughs, each one making computers smaller, faster, cheaper, and more accessible.

This tutorial traces the physical evolution of computers from the vacuum tube to the microprocessor and beyond. Understanding this history is not merely academic—it explains why computers are designed the way they are, why software must account for hardware limitations, and what physical constraints will shape the future of computing.

5.2 The Vacuum Tube Era (1940s–1950s): The Birth of Electronic Computing

5.2.1 What is a Vacuum Tube?

Before electronic computers, mechanical and electro-mechanical machines (like Babbage's Analytical Engine and IBM's punch-card tabulators) relied on gears, relays, and switches. The vacuum tube (also called a thermionic valve) changed everything.

A vacuum tube is a glass cylinder from which most air has been removed. Inside are metal electrodes: a cathode (which emits electrons when heated) and an anode (which collects them). By controlling the flow of electrons between these electrodes, a vacuum tube can act as an amplifier (boosting a signal) or a switch (turning current on or off). It was this switching ability—the ability to represent the binary states of 0 and 1—that made vacuum tubes the foundation of the first electronic computers.

Key limitation: Vacuum tubes were fragile, generated enormous heat, consumed vast amounts of electricity, and burned out frequently (average lifespan: a few thousand hours). ENIAC had so many tube failures that it was often down for maintenance.

5.2.2 ENIAC: The First General-Purpose Electronic Computer

ENIAC (Electronic Numerical Integrator and Computer) was unveiled to the public on February 14, 1946 at the University of Pennsylvania. It was designed by John Mauchly and J. Presper Eckert, Jr. for the U.S. Army to calculate artillery range tables.

Specifications that boggle the modern mind:

Feature ENIAC
Vacuum tubes 18,000
Weight 30 tons (approximately the weight of six elephants)
Dimensions 50 feet long × 30 feet wide (1,500 square feet—about the size of a large classroom)
Power consumption 150 kilowatts (enough to power a small neighborhood)
Speed 5,000 additions per second; 360 multiplications per second
Cost $400,000 (about $6 million in today's money)
Programming By physically rewiring plugboards and setting switches—reprogramming could take days

Why it mattered: ENIAC proved that electronic computing was feasible. It demonstrated that machines could perform calculations thousands of times faster than mechanical calculators.

5.2.3 UNIVAC I: The First Commercial Computer

UNIVAC I (Universal Automatic Computer) was developed by Eckert and Mauchly's company (later acquired by Remington Rand) and delivered to the U.S. Census Bureau in 1951. It was the first commercially successful computer.

Feature UNIVAC I
Vacuum tubes 5,200
Weight 29,000 pounds (13 metric tons)
Dimensions 14.5 feet long × 7.5 feet high × 9 feet wide
Power consumption 125 kW
Speed ~1,905 operations per second
Clock speed 2.25 MHz

Key innovation: UNIVAC used magnetic tape for external storage, replacing the slower punched cards. It also used acoustic delay-line memory, where data was stored as sound waves traveling through tubes of mercury.

5.2.4 The Limits of the Vacuum Tube Era

By the mid-1950s, it was clear that vacuum tubes were a dead end. Computers could not grow much larger without becoming impossibly unreliable and power-hungry. A new technology was desperately needed.

5.3 The Transistor Era (1950s–1960s): The Solid-State Revolution

5.3.1 The Invention of the Transistor

In 1947, three scientists at Bell Labs—John Bardeen, Walter Brattain, and William Shockley—invented the transistor. Unlike a vacuum tube, a transistor is a solid-state device made of semiconductor material (typically silicon or germanium). It has no moving parts, no glass to break, no filament to burn out.

How a transistor works: A transistor has three layers: the emitter, base, and collector. A small current or voltage applied to the base controls a much larger current flowing between the emitter and collector. This allows the transistor to act as a switch or amplifier—just like a vacuum tube, but vastly smaller, more reliable, and more efficient.

Why the transistor was revolutionary:

5.3.2 The First Transistorized Computers

IBM was quick to adopt the new technology. In 1959, IBM introduced the IBM 7090, a fully transistorized version of its vacuum-tube-based IBM 709.

Feature IBM 709 (Vacuum Tube) IBM 7090 (Transistor)
Technology Vacuum tubes Transistors
Speed Baseline 6 times faster than the 709
Cost (1960) $2,900,000 or $63,500/month rent

The IBM 7090 became a popular mainframe for scientific and technological applications. By 1960, vacuum-tube computers had become completely uncompetitive.

5.3.3 The Impact of the Transistor

The transistor did more than just make computers faster. It initiated the era of miniaturization. Computers that once filled rooms could now fit in a closet. This made computers commercially viable for businesses, universities, and government agencies.

However, transistors were still discrete components—each one was a separate physical part that had to be individually soldered onto a circuit board. As computer complexity grew, so did the number of transistors, and the problem of wiring them together became a bottleneck.

5.4 The Integrated Circuit Era (1960s–1970s): The Chip is Born

5.4.1 The Invention of the Integrated Circuit

In 1958, Jack Kilby at Texas Instruments demonstrated the first integrated circuit (IC)—a single piece of semiconductor material containing multiple transistors and other components, all connected internally. Independently, Robert Noyce at Fairchild Semiconductor developed a similar design using a planar process that made mass production practical.

Key insight: Instead of wiring together individual transistors, an integrated circuit builds the transistors and the wires connecting them onto a single silicon chip. This eliminated the need for manual assembly of discrete components.

5.4.2 The First IC-Based Computers

IBM was again at the forefront. In 1964, IBM announced the System/360, a family of compatible mainframe computers. The System/360 was developed during the transition from discrete transistors to integrated circuits.

The System/360 was revolutionary not just for its technology but for its compatibility—all models in the family could run the same software, a concept that is now standard but was groundbreaking at the time.

5.4.3 The Impact of the Integrated Circuit

The IC marked the beginning of exponential growth in computing power. Instead of adding transistors one at a time, manufacturers could now add them by the thousands, then tens of thousands, then millions—all on a single chip.

This set the stage for the most transformative development in computer history.

5.5 The Microprocessor Era (1970s–Present): A Computer on a Chip

5.5.1 The First Microprocessor: Intel 4004

In November 1971, Intel unveiled the Intel 4004, the world's first commercially available microprocessor. The 4004 was originally designed for a Japanese desk calculator, but its implications were far broader.

Feature Intel 4004
Transistors 2,300
Data width 4 bits
Clock speed 740 kHz
Cost at launch $60

The astonishing claim: The Intel 4004 packed "as much computing power as the ENIAC"—a machine that weighed 30 tons, into a chip that weighed less than an ounce.

5.5.2 The Personal Computer Revolution

The microprocessor made personal computers possible. While mainframes and minicomputers still cost tens or hundreds of thousands of dollars, a microprocessor-based computer could be built for a few thousand—or even a few hundred.

Key milestones in the personal computer revolution:

5.5.3 Moore's Law: The Engine of Progress

In 1965, Gordon Moore (co-founder of Intel) observed that the number of transistors on a chip was doubling approximately every year. He later revised this to every two years. This observation became known as Moore's Law.

The numbers tell the story:

Moore's Law has driven the exponential growth in computing power for over five decades. However, as transistors approach atomic scales, physical limits are becoming apparent—the end of Moore's Law is now a serious topic of discussion.

5.5.4 LSI and VLSI: Putting it All Together

The microprocessor era was enabled by two key manufacturing advances:

These technologies allowed not just the CPU but also memory, graphics, and other functions to be integrated onto a single chip—the system-on-a-chip (SoC) used in smartphones and tablets today.

5.6 The Modern Era: From Room-Sized to Pocket-Sized

5.6.1 The Smartphone: A Computer in Your Pocket

The modern smartphone contains more computing power than the supercomputers of the 1990s. It is a testament to the miniaturization driven by Moore's Law and VLSI technology. A typical smartphone SoC contains:

5.6.2 The Rise of Cloud Computing

Cloud computing represents a shift in how we think about computer construction. Instead of owning physical hardware, users access computing resources over the Internet. This was made possible by:

Cloud computing allows users to access virtually unlimited computing power on demand, without owning the underlying hardware.

5.6.3 The Emergence of AI Hardware

Artificial intelligence, particularly deep learning, has driven the development of specialized hardware:

These specialized chips are optimized for massive parallelism, trading off general-purpose flexibility for AI-specific performance.

5.6.4 Quantum Computing: The Next Frontier

Quantum computing represents a fundamentally different approach to computation. Instead of bits (0 or 1), quantum computers use qubits, which can exist in multiple states simultaneously (superposition). This allows quantum computers to solve certain problems—like factoring large numbers or simulating quantum physics—exponentially faster than classical computers.

Current state: Quantum computers are still in their infancy. They are large, expensive, and require extreme cooling (near absolute zero). However, major tech companies and governments are investing billions in quantum research.

5.7 Summary Table: The Evolution of Computer Construction

Era Years Key Technology Size Example Speed Example Cost Example Key Machine
Vacuum Tube 1940s–1950s Vacuum tubes Room-sized (1,500 sq ft) 5,000 additions/sec $400,000 (1946) ENIAC
Transistor 1950s–1960s Discrete transistors Closet-sized 6× faster than tubes $2.9M (1960) IBM 7090
Integrated Circuit 1960s–1970s ICs on silicon chips Desk-sized Thousands of ops/sec Hundreds of thousands IBM System/360
Microprocessor 1970s–present LSI/VLSI Desktop to pocket-sized Billions of ops/sec $60 (4004 in 1971) Intel 4004
Modern Present–future AI, cloud, quantum Pocket-sized (phone) to global (cloud) Trillions of ops/sec Pay-per-use (cloud) Smartphones, cloud data centers

5.8 The Unifying Principle: Miniaturization and Democratization

The entire history of computer construction can be seen as a single, relentless trend: making computers smaller, cheaper, and more powerful. This has democratized computing:

The physical construction of computers has evolved from room-sized collections of fragile glass tubes to microscopic patterns etched onto silicon chips. This evolution has been driven by physics, materials science, and human ingenuity—and it shows no signs of stopping.

5.9 Quizzes

Quiz 1: The Vacuum Tube Era

1. What was the primary function of a vacuum tube in early electronic computers?

Answer(B) To act as a switch or amplifier

2. Approximately how many vacuum tubes did the ENIAC contain?

Answer(B) 18,000

3. How much power did ENIAC consume?

Answer(C) 150 kW

4. What was the primary limitation of vacuum tube computers?

Answer(B) They were unreliable, generated massive heat, and consumed huge amounts of power

5. UNIVAC I was significant because it was:

Answer(B) The first commercially successful computer

Quiz 2: The Transistor and Integrated Circuit Eras

6. Who invented the transistor at Bell Labs in 1947?

Answer(C) Bardeen, Brattain, and Shockley

7. Compared to vacuum tubes, transistors are:

Answer(B) Smaller, more reliable, and more energy-efficient

8. The IBM 7090 was significant because it was:

Answer(B) A fully transistorized version of an earlier vacuum-tube computer

9. The integrated circuit was invented in 1958 by:

Answer(B) Jack Kilby (and independently by Robert Noyce)

10. The IBM System/360 was revolutionary because:

Answer(B) It offered a family of compatible computers that could run the same software

Quiz 3: The Microprocessor Era

11. When was the Intel 4004, the first commercially available microprocessor, introduced?

Answer(B) 1971

12. How many transistors did the Intel 4004 contain?

Answer(B) 2,300

13. What was the approximate cost of the Intel 4004 at launch?

Answer(B) $60

14. Gordon Moore's prediction that the number of transistors on a chip doubles approximately every two years is known as:

Answer(B) Moore's Law

Quiz 4: The Modern Era

15. Which of the following is a characteristic of cloud computing?

Answer(B) Computing resources are delivered over the Internet on demand

16. GPUs (Graphics Processing Units) have become important for AI because:

Answer(B) They excel at the massive parallel computations required for neural networks

17. Quantum computers use which fundamental unit of information?

Answer(C) Qubit

18. The Intel 80486 microprocessor, released in 1989, contained more than:

Answer(C) 1 million transistors

5.10 Exercises

Exercise 1: Timeline Construction

Instructions: Create a visual timeline of computer hardware evolution from 1940 to the present. Include the following milestones with dates, key specifications, and a one-sentence significance statement:

  1. ENIAC (1946)
  2. UNIVAC I (1951)
  3. Transistor invented (1947)
  4. IBM 7090 (1959)
  5. Integrated circuit invented (1958)
  6. IBM System/360 (1964)
  7. Intel 4004 microprocessor (1971)
  8. First personal computers (mid-1970s)
  9. IBM PC (1981)
  10. Modern smartphone (2007–present)

Format: You may draw this by hand, create a digital diagram, or present it as a written list with clear visual spacing.

Exercise 2: Comparison Chart

Instructions: Complete the following table comparing the key characteristics of each era:

Feature Vacuum Tube Era Transistor Era IC Era Microprocessor Era Modern Era
Years
Key Technology
Typical Size
Approx. Speed
Primary Use
Key Limitation
Representative Machine

Exercise 3: Comparative Analysis

Instructions: Write a short paragraph (150–200 words) comparing the ENIAC and the Intel 4004. Include at least three specific points of comparison (size, speed, cost, etc.) and explain why the comparison is significant.

Sample Answer The ENIAC and the Intel 4004 represent two extremes of computing history, separated by just 25 years. ENIAC weighed 30 tons, consumed 150 kW of power, and cost $400,000, while the 4004 weighed less than an ounce, consumed milliwatts, and cost just $60. ENIAC contained 18,000 vacuum tubes and filled a room; the 4004 packed 2,300 transistors onto a chip the size of a fingernail. Yet astonishingly, the 4004 had roughly the same computing power as the ENIAC. This comparison illustrates the exponential pace of technological progress driven by Moore's Law, showing how computing power that once required a room full of equipment can now fit in your pocket.

Exercise 4: Moore's Law Calculations

Instructions: Answer the following questions based on Moore's Law (transistor count doubles approximately every two years).

  1. The Intel 4004 (1971) had 2,300 transistors. According to Moore's Law, approximately how many transistors would you expect on a chip in:

  2. If a chip had 1 million transistors in 1989 (like the Intel 80486), approximately how many transistors would Moore's Law predict for:

  3. Modern high-end CPUs have over 50 billion transistors. If Moore's Law continues, approximately when might we reach 1 trillion transistors on a chip? Show your calculation.

Answers 1. (a) 1973: ~4,600; (b) 1975: ~9,200; (c) 1981: ~73,600; (d) 1991: ~2.35 million. 2. (a) 1991: ~2 million; (b) 1995: ~8 million; (c) 2001: ~128 million. 3. Starting from 50 billion (~5 × 10¹⁰), doubling every 2 years. To reach 1 trillion (10¹²), we need about 4.3 doublings (since 2^4.3 ≈ 20; 50B × 20 = 1T). At 2 years per doubling, that's about 8.6 years. So approximately 2034–2035.

Exercise 5: Technology Adoption Timeline

Instructions: Research and write a short paragraph about ONE of the following topics:

  1. The first personal computer: What was it, who made it, and why was it significant?

  2. The first smartphone: What was it, when was it released, and how did it change computing?

  3. The history of cloud computing: What key technologies made it possible, and when did it become mainstream?

  4. The history of GPUs: How did graphics processors evolve from gaming to AI?

5.11 Homework Questions

Answer the following questions in complete sentences. Each response should be 3–5 sentences unless otherwise specified.

Short Answer Questions

1. Explain why vacuum tubes were eventually replaced by transistors. What specific advantages did transistors offer?

Sample AnswerVacuum tubes were replaced by transistors because transistors were dramatically smaller, more reliable, consumed far less power, and generated much less heat. While vacuum tubes frequently burned out and required constant maintenance, transistors were solid-state devices with no moving parts and much longer lifespans. This made computers more practical, affordable, and commercially viable.

2. What was the significance of the integrated circuit in the evolution of computer construction?

Sample AnswerThe integrated circuit allowed multiple transistors and their interconnections to be manufactured on a single silicon chip, eliminating the need to wire together individual discrete components. This drastically reduced the size and cost of computers while increasing their speed and reliability. The IC was the essential enabling technology for the microprocessor and all modern computing.

3. How did the invention of the microprocessor change who could afford and use computers?

Sample AnswerThe microprocessor put a complete computer's processing power onto a single inexpensive chip, making it possible to build computers for thousands or even hundreds of dollars instead of millions. This transformed computing from an expensive tool for governments and large corporations into a personal technology accessible to individuals, small businesses, and schools, ultimately leading to the personal computer revolution.

4. Explain Moore's Law in your own words. Why has it been so important for the evolution of computers?

Sample AnswerMoore's Law is the observation that the number of transistors on a chip doubles approximately every two years, leading to exponential growth in computing power at decreasing cost. This predictable doubling has driven the entire computer industry for over five decades, enabling the miniaturization and democratization of computing from room-sized mainframes to pocket-sized smartphones.

5. What is the difference between a general-purpose CPU and a specialized AI accelerator like a GPU or TPU?

Sample AnswerA general-purpose CPU has a few powerful cores optimized for sequential, complex logic and is designed to handle a wide variety of tasks. In contrast, a GPU or TPU has thousands of simpler cores designed for massive parallel computation, making them far more efficient for the matrix multiplications and parallel operations required by neural networks. Specialized AI hardware trades general-purpose flexibility for dramatically higher performance on specific AI workloads.

Essay Questions

Answer the following questions in 300–500 words each.

6. Trace the evolution of computer construction from ENIAC to the modern smartphone. What were the three most important technological breakthroughs, and how did each one change what computers could do and who could use them?

Suggested outline:

7. Discuss the concept of Moore's Law. What physical and economic factors have driven it, and what are the challenges to its continuation? How might the end of Moore's Law change the future of computer construction?

Suggested outline:

Research Questions

These questions require additional research beyond the tutorial content.

8. Research the "Stretch" computer project (IBM 7030). How did this project contribute to the development of transistorized computers, and what lessons did IBM learn from its failure to meet performance goals?

9. Research the history of the Apollo Guidance Computer, which was used in the Apollo moon missions. How did integrated circuits make this computer possible, and what were its specifications compared to modern computers?

10. Research the current state of quantum computing. What are the leading approaches (superconducting qubits, trapped ions, etc.), and what are the main technical challenges that must be overcome before quantum computers become practical for everyday use?

Previous | Tutorial index | Next