News
IndustryTSMC’s grip on advanced chipmaking will face competitionIndustryCadence joins European Chip Design Platform to support semiconductor startups and SMEsMaking chipsMarc Hijink on ASML’s Culture, Supply Chain, and GeopoliticsProcessorsIntel's post-Nova Lake CPUs take shape as Razor Lake desktop and laptop variants surfaceAI chipsHow Google’s New AI Chip Software Helps It Battle NvidiaIndiaMore IC Packaging Plants Emerge In IndiaIndustryUS Tightens China AI Chip Access as Nvidia, TSMC and Semiconductor Stocks SlideDesignAxelera AI moves into HPC with TitaniaMaking chipsChinese envoy says chip equipment export curbs threaten EU-China trade thawIndiaSEMICON India 2026 Day 2 Sees 25 MoUs, Strategic Collaborations Across Semiconductor Value ChainIndustryTSMC’s grip on advanced chipmaking will face competitionIndustryCadence joins European Chip Design Platform to support semiconductor startups and SMEsMaking chipsMarc Hijink on ASML’s Culture, Supply Chain, and GeopoliticsProcessorsIntel's post-Nova Lake CPUs take shape as Razor Lake desktop and laptop variants surfaceAI chipsHow Google’s New AI Chip Software Helps It Battle NvidiaIndiaMore IC Packaging Plants Emerge In IndiaIndustryUS Tightens China AI Chip Access as Nvidia, TSMC and Semiconductor Stocks SlideDesignAxelera AI moves into HPC with TitaniaMaking chipsChinese envoy says chip equipment export curbs threaten EU-China trade thawIndiaSEMICON India 2026 Day 2 Sees 25 MoUs, Strategic Collaborations Across Semiconductor Value Chain
EveryChip
IMP-16

IMP-16

A 1973 bit-slice microprocessor built from five PMOS chips

National Semiconductor's IMP-16 arrived in 1973 as the earliest microprocessor to handle 16-bit words across more than one chip. Four RALU slices plus a control ROM formed the set, and it went on to appear in a microcomputer, a Jacquard Systems machine and an Aston Martin car.

Share on WhatsApp

  • First multi-chip 16-bit microprocessor, released 1973
  • Four 4-bit RALU slices plus one CROM control chip
  • 715 kHz four-phase clock, PMOS logic
  • Addresses 64K words; no byte-level writes
  • Used in the Aston Martin Lagonda

What it is

IMP-16 was a 16-bit microprocessor put together from several separate integrated circuits, a layout known as bit-slice design. National Semiconductor brought it out in 1973. Four matching RALU parts each handled four bits of the register file and arithmetic, and a fifth chip, the CROM, held the control sequencing and microcode. Together they produced a full 16-bit word. It was the first microprocessor of its width to work this way, and it drew on the instruction style of the Data General Nova minicomputer.

How it came about

National Semiconductor designed the set and shipped it in 1973. The approach answered a real problem of the day: single chips could not yet carry a full 16-bit data path, so the work was split across several slices running side by side. Adoption stayed narrow. The IMP-16P microcomputer and Jacquard Systems' J100 used it, and it also turned up in the Aston Martin Lagonda, helped by the fact that National Semiconductor's chairman Peter Sprague held a large stake in the car maker.

Inside the chip

Each RALU slice carries four bits of the program counter, a group of registers, the arithmetic logic unit, a 16-word last-in first-out stack and status flags. Four 16-bit accumulators sit above that, and two of them double as index registers. Memory is addressed in 16-bit words, so single bytes cannot be written; up to 64K words, or 128K bytes, are reachable. A 16-bit multiplexer feeds condition bits such as zero, carry and overflow back into control. The instruction set follows the Data General Nova pattern, and basic instructions occupy one word.

How it was made

The parts were built with PMOS logic and sold in 24-pin dual in-line packages, five of them for a basic set and six with the extra chip. A four-phase non-overlapping clock running at 715 kHz drove the set, with a voltage swing from +5 to -12. No transistor total is given for the parts.

Where it went

Two computers relied on the chip set: National Semiconductor's own IMP-16P and the J100 from Jacquard Systems. Beyond those, take-up was slight. One unusual installation was the Aston Martin Lagonda, where the processor found a home in the car's electronics.

Versions and family

Adding the CROM II part, also labelled IMP-16A or 522D, brought 16-bit multiply and divide routines as well as byte handling, though byte access covered only the lower 32K words. Later single-chip 16-bit parts from the same company, PACE and INS8900, took over from the multi-chip design. They kept a related architecture but did not run the same binaries.

Why it matters

IMP-16 showed that a 16-bit processor could be assembled from smaller slices, a pattern that let builders scale word width by adding parts. Its Nova-style instruction set linked microprocessors to the minicomputer world. Sales were modest and the design was soon replaced, but the idea of parallel bit slices outlived the product.

Problems and reception

The set saw little use outside a handful of machines, and no faults or recalls are recorded. Its limits were practical ones: byte writes were impossible, and the extra chip was needed for multiply, divide and byte work.

Questions people ask about IMP-16

What kind of chip is the IMP-16?

It is a 16-bit microprocessor assembled from several integrated circuits rather than one. Four identical RALU slices each supply four bits of registers and arithmetic, while a CROM chip holds the control sequencing and microcode. National Semiconductor released the set in 1973.

Why use several chips instead of one?

In 1973 a single part could not carry a whole 16-bit data path. Splitting the job into 4-bit slices let each slice work in parallel, and the pieces combined into a 16-bit word. This bit-slice method also let designers widen a machine by adding more slices.

How much memory could it use?

The design reaches 64K words, equal to 128K bytes. Memory is written in whole 16-bit words, so there is no way to store a single byte. With the CROM II chip, byte operations become possible, but only across the lower 32K words.

What replaced it?

National Semiconductor's PACE and the INS8900, both single-chip 16-bit processors, took over from the multi-chip set. They followed a similar architecture, yet programs built for IMP-16 would not run unchanged on them.

Where did it actually turn up?

It powered the IMP-16P microcomputer and Jacquard Systems' J100. It also appeared in the Aston Martin Lagonda, a link that came about because National Semiconductor's chairman Peter Sprague was a major shareholder in Aston Martin.

Good reading and sources

Where this page comes from

The facts on this page come from the Wikipedia article "IMP-16", written by its contributors and used under the CC BY-SA 4.0 licence, and from its Wikidata record (CC0). EveryChip rewrote the text in its own words, with AI assistance, and compares every page with the article to make sure no passage is copied. It may be reused under the same CC BY-SA 4.0 licence with a link back.

It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-14.

Photo credits on this page

The small pictures above come from Wikimedia Commons under free licences. Some are cropped to fit.

The weekly digest

Every Monday: the chip of the day, new pages on EveryChip, and the conferences and anniversaries of the week ahead.