Saved
Hardware · Tech history

PDP-8

1965

The PDP-8 didn’t invent the minicomputer—it weaponised price.

Worth your time?4.5/5Yes.Study the whole thing.What works, what does not, who it is for ↓
10:24in productionCh. 1 · The $18,500 Threshold/ 10:24 · ceiling 15 min
4 chapters · 10:24Ceiling 15:00

The brief is written and checked. The video joins this page when it is cut.4 chapters · 1 min read · 10:24 planned

Read the brief

The PDP-8 is a family of 12-bit minicomputers produced by DEC. It launched in 1965 as the first minicomputer to sell for under $20,000. Over 50,000 units were sold during its lifetime. The earliest model—the Straight-8—was introduced on 22 March 1965 and used diode–transistor logic on flip chip cards. It was the first computer sold for under $20,000 and became the best-selling computer in history at that time. Later models included the PDP-8/S (introduced August 1966), which used a serial, single-bit-wide data path, and subsequent models that returned to fully parallel implementations using TTL MSI logic.

Chapters & takeaways4
  1. 1:02
    The $18,500 Threshold

    It is the first computer under $20,000—and the first to make 'minicomputer' a meaningful category.

  2. 2:42
    Hardware in a Box

    The Straight-8’s diode–transistor logic on flip chip cards fits in a refrigerator-sized cabinet.

  3. 4:36
    Speed vs. Count

    The PDP-8/S trades speed for simplicity; later models abandon serial arithmetic for parallel TTL.

  4. 6:21
    Volume Over Velocity

    Over 50,000 units sold—not because it was faster, but because it was affordable and repairable.

1 more chapterFewer chapters
Worth your time?

Yes. Study the whole thing.

4.5/ 5
What works
  • Price point enabled mass adoption
  • Modular design supported field maintenance
  • 12-bit word suited instrumentation and control
  • Parallel-to-serial-to-parallel evolution exposed a real engineering constraint
What does not
  • It does not run modern operating systems.
  • It does not support networking as defined post-1970.
  • It does not include floating-point arithmetic in hardware.
  • It does not offer backward compatibility across all models.
Study it if
  • Historians of computing infrastructure
  • Engineers studying cost–performance trade-offs in logic design
  • Educators teaching pre-microprocessor digital systems
Skip it if
  • Developers seeking runtime environments
  • Architects evaluating scalability
  • Security researchers assessing attack surface
Who should study it, and who should skipHide who it is for
The written brief1 min read

What it is and the problem it solves

It is a family of 12-bit minicomputers built by DEC. It solves the problem of prohibitive cost in institutional computing. It targets labs, industrial controllers, and educational sites where a $20,000 ceiling was decisive.

How it works

It uses a 12-bit architecture with diode–transistor logic on flip chip cards. The earliest model processes data in parallel. The PDP-8/S uses a serial, single-bit-wide data path for arithmetic. Later models switch to TTL MSI logic and restore full parallelism.

What works

Its price point works: it sells over 50,000 units. Its modularity works: flip chip cards allow field repair and incremental upgrades. Its 12-bit word size works for real-time instrumentation and process control. Its parallel implementation in the Straight-8 delivers usable throughput for its target applications.

What does not

It does not establish a new instruction set architecture standard. It does not support multitasking, virtual memory, or high-level language compilation out of the box. It does not scale beyond single-user, real-time control tasks. Its serial arithmetic path sacrifices speed for component count reduction—and that compromise was abandoned in later models.

What it changes

It changes the price threshold for computer ownership. It redefines ‘computer’ as something that fits in a lab or factory room, not a climate-controlled vault. It proves that cutting logic gate count—even at the cost of speed—can expand market reach. It makes computing accessible to institutions that could not justify mainframe budgets.

Is it worth your time

Yes—if you work on the history of computing infrastructure or need a concrete reference point for cost-driven architectural trade-offs in early digital systems. No—if you need performance benchmarks, software ecosystems, or compatibility data: none are provided in the source material.

Same field · Hardware4 of 251
10:15
Multi-touchMulti-touch is a hardware-software coupling that enables exact, simultaneous spatial input. It worked in accelerator control rooms before personal computing existed. Its value lies not in novelty, but in proving that concurrency can be sensed, localised, and acted on—without abstraction layers or mode switches. It remains foundational—not because it scaled first, but because it solved concurrency before anyone asked for it.
10:48
IBM Personal Computer XTThe IBM Personal Computer XT is a hardware-locked evolution of the original PC. It adds local storage and expansion capacity without altering performance, compatibility, or user workflow at the instruction level. Its significance lies not in innovation but in consolidation: it turns optional accessories into baseline infrastructure.
8:52
Hard disk drive1956The hard disk drive is a rigid-platter magnetic storage device introduced by IBM in 1956 as the 350 for the 305 RAMAC. It enables random-access, non-volatile data storage. Its architecture—rotating platters, magnetisable film, movable read/write heads, servo positioning—remains unchanged. It dominated secondary storage from the early 1960s through the 2000s. Today, it faces displacement by SSDs in speed- and mobility-sensitive applications, though remains cost-effective for bulk storage.
9:51
Optical storageOptical storage fixes data in physical form using light — not electricity or magnetism. It succeeded where durability and distribution mattered more than speed or flexibility.
Up next in Technology

Philips

1891 · 9:13

Philips didn’t invent the future — it licensed the present, then pivoted before the market caught up.

9:13