technologybriefs
Systems

Semiconductors

54
briefs
9:59
average
539 min
in total
4
inventors
Refine25
Fields
Hardware 252Software 115Web & cloud 18Security 30Semiconductors 54Energy 29Models 7Tools 18Agents 1Robotics 30Tech history 248Products 39Failures 5Internet culture 25
Verdict
Worth your time 581Worth it in part 1Worth skipping 38
Topics
Invention 241Computing 218Engineering 152Hardware 139Infrastructure 117Abstraction 51Programming 51Semiconductors 47All 140 topics →
All briefs54
9:37

Crucial

Crucial was the brand under which Micron Technology sold its consumer memory and storage products, until the brand was discontinued in 2026.
9:20

Chiplet

Chiplets don’t replace Moore’s Law—they bypass it with Lego logic and yield math.
9:27

ECC memory

ECC memory doesn’t prevent memory errors—it prevents you from ever knowing they happened unless they’re too big to fix.
9:33

EUV lithography

EUV lithography is not a refinement — it is a forced pivot into vacuum, reflection, and tin plasma, because light refused to keep shrinking.
9:45

Multigate device

One gate wasn’t enough — so engineers wrapped the channel in two, three, or more. That’s not evolution. It’s desperation.
12:10

Onsemi

onsemi doesn’t invent new physics — it industrialises Motorola’s legacy, one acquired wafer fab at a time.
10:49

RISC-V

UC Berkeley researchers · 2010

RISC-V is a free and open instruction set architecture developed in 2010 at UC Berkeley. Its specifications were published openly in 2011 and stewardship transferred to RISC-V International in 2015. It enables royalty-free processor design—but does not eliminate the engineering, tooling, or ecosystem work required to deploy it.

10:02

Wafer (electronics)

The wafer is not a component—it is the ground upon which computation is built, and its limits define what chips can do.
9:41

Graphics processing unit

The GPU didn’t accelerate graphics—it relocated computation, and in doing so, rewrote the rules of what a chip could be used for.
9:24

Photoresistor

It’s not a sensor — it’s a switch that blinks slowly and forgets the dark.
10:09

X86

x86 isn’t enduring because it’s good — it’s enduring because every fix creates new dependencies no one dares break.
10:49

Acorn Archimedes

ARM’s first desktop test was a closed, British, BBC-branded machine — not a global platform.
10:08

ARM architecture family

ARM isn’t about efficiency—it’s about offloading architectural control so others can build cheaper, cooler, more varied chips.
9:34

Broadcom

Broadcom isn’t building the future—it’s buying, bundling, and billing it.
10:10

Kingston Technology

Memory wasn’t commoditised until Kingston proved it could be validated, branded, and sold outside the OEM.
9:30

Mellanox Technologies

Mellanox didn’t invent speed—it weaponised standardisation against latency, then vanished into Nvidia before its full Ethernet promise could be tested.
9:55

Photolithography

Photolithography didn’t invent the microchip—it made mass production inevitable.
10:30

Qualcomm

Qualcomm doesn’t build networks—it builds the legal and technical tollbooths inside them.
9:15

Read-only memory

Read-only memory isn’t about limitation — it’s about guaranteeing what cannot change.
9:44

SK Hynix

Three vendors control the world’s memory—and SK Hynix is one of them, not by inevitability, but by surviving its own missteps.
9:18

System on a chip

SoCs shrink computers—but lock in obsolescence the moment they’re fabricated.
11:49

AI data center

AI data centers don’t just run models—they starve the rest of computing of memory.
11:21

FPGA prototyping

FPGA prototyping doesn’t prevent re-spins—it just moves the pain from silicon to the lab.
9:24

LED display

Nick Holonyak

The LED display is not Holonyak’s invention — it is HP’s. Holonyak’s 1962 GaAsP LED enabled it. HP’s 1969 Model 5082-7000 was the first intelligent, IC-integrated numeric display — fast, solid-state, low-power, and monochromatic. It displaced Nixie tubes but could not render text, graphics, or colour. Later all-LED flat panels remained monochrome and experimental.

8:39

Solid-state drive

SSDs killed the spinning disk—not by being perfect, but by making latency optional.
10:55

Western Digital

Western Digital wasn’t built to store data — it was built to test chips, and only stored bits after it stopped testing them.
10:06

Analog Devices

Precision signal conversion is not magic — it’s Analog Devices’ 1965 bet that the world would need chips that don’t lie about voltage.
10:16

Applied Materials

Applied Materials doesn’t make chips—it makes the machines that force atoms into place, one layer at a time.
10:26

ASML

ASML didn’t invent lithography—it weaponised integration, turning a crowded market into a single-point failure for global computing.
9:35

Commodore 64

Custom silicon—not marketing—made the C64 dominant. Everything else was cost engineering.
9:26

DDR4 SDRAM

DDR4 trades compatibility for capacity—and delivers on neither promise until 2014.
11:19

DDR5 SDRAM

DDR5 doubles capacity and cuts voltage—but leaves latency untouched and bandwidth gains conditional on cooling and controller design.
10:02

Flash memory

Fujio Masuoka

Flash memory is a block-erasable, non-volatile semiconductor memory invented by Fujio Masuoka at Toshiba in 1980. It builds on EEPROM but replaces byte-wise erasure with faster block-wise erasure — enabled by floating-gate transistor design. The name 'flash' was coined by colleague Shōji Ariizumi to reflect that speed. Masuoka and colleagues presented NOR flash in 1984 and NAND flash at the 1987 IEDM. Toshiba began marketing flash memory and commercially launched NAND flash — both in 1987. The first published implementation held only 8192 bytes but became the basis for larger-capacity versions. Flash does not support byte-level writes or in-place updates; erasure is all-or-nothing per block. It cannot replace RAM or true EEPROM where fine-grained modification is required. Flash changed how portable devices store firmware and user data, enabling compact, shock-resistant, low-power storage that scales without moving parts — a prerequisite for digital cameras, USB drives, SSDs and smartphones.

8:58

GlobalFoundries

A foundry built to escape AMD’s balance sheet became a specialist anchor—just as the industry stopped chasing nanometres.
9:54

NXP Semiconductors

NFC isn’t magic — it’s NXP’s tightly controlled radio handshake, built to replace plastic cards with chips you already carry.
10:13

PCI Express

PCI Express did not reinvent connectivity—it replaced a broken shared bus with a set of dedicated wires, each speaking in packets.
10:55

Samsung Electronics

Samsung wasn’t born a chipmaker—it bought its way in, merged its way up, and declared its way to relevance.
10:42

STMicroelectronics

Europe didn’t need another chipmaker — it needed a sovereign stack. ST built one, slowly, silently, and without hype.
9:45

Texas Instruments

Calling 'Texas Instruments (1930)' an invention confuses corporate founding with technological creation — and erases 24 years of actual engineering.
9:18

Microprocessor

The microprocessor did not invent the computer—it miniaturised its most fragile part into silicon.
10:39

Semiconductor

Semiconductors aren’t magic—they’re doped crystals whose usefulness begins only when quantum physics meets deliberate impurity placement.
8:33

Thermistor

Samuel Ruben · 1833

The thermistor is a semiconductor resistor whose resistance changes strongly with temperature. Michael Faraday discovered its NTC behaviour in silver sulfide in 1833. Commercial production did not begin until the 1930s due to manufacturing difficulty and limited applications. Samuel Ruben invented a commercially viable version in 1930.