RISC Single Chip
IBM single-chip microprocessor for entry-level workstations
The RISC Single Chip (RSC) is a microprocessor designed and made by IBM. It condensed the multi-chip POWER1 design into one die using 0.8 micron CMOS technology. Running at up to 45 MHz, it powered entry-level IBM RS/6000 workstations.
- Single-chip implementation of the POWER1 architecture
- Operates at 33 MHz or 45 MHz
- Contains approximately one million transistors
- Uses 0.8 micron CMOS fabrication process
- Features an 8 KB unified instruction and data cache
What it is
The RISC Single Chip, known as RSC, is a microprocessor created by International Business Machines. It serves as a compact version of the earlier POWER1 central processing unit. While POWER1 used multiple chips, the RSC fits all functions onto a single piece of silicon. This design targeted entry-level models in the IBM RS/6000 workstation family. The chip integrates memory control and input output logic directly on the die. It operates at frequencies of 33 or 45 megahertz. Its compact form allowed IBM to reduce costs for lower-end systems while maintaining the POWER instruction set architecture.
How it came about
IBM developed the RSC to simplify the complex multi-chip POWER1 design. The goal was to create a feature-reduced implementation that could fit on one die. This approach addressed the need for cheaper workstations without abandoning the existing software ecosystem. The chip was fabricated using complementary metal-oxide semiconductor technology. By integrating previously separate components, IBM reduced the physical footprint and power requirements. The resulting processor supported industry standard memory modules instead of custom cards, further lowering system costs for customers buying entry-level hardware.
Inside the chip
The processor contains nine functional units on its die. These include fixed point and floating point execution units, a branch processor, and memory management logic. The fixed point unit handles integer math and address generation using a three-stage pipeline. The floating point unit lacks register renaming to save space. It uses a 32-bit wide multiply-add array. For double-precision tasks, operands are split and processed in two passes. An 8 KB unified cache stores both instructions and data. This cache is two-way set associative and uses a least recently used replacement policy. The memory bus is 72 bits wide, with eight bits reserved for error correction.
How it was made
IBM manufactured the RSC using a 0.8 micron CMOS process with three levels of wiring. The die measures 14.9 mm by 15.2 mm, covering an area of roughly 226 square millimetres. It holds approximately one million transistors. The chip is housed in a ceramic pin grid array module measuring 36 mm square. This package features 201 signal pins. The design requires a 3.6 volt power supply. At 33 MHz operation, the processor consumes only 4 watts of power. This low power draw and small form factor were key advantages over the larger POWER1 modules.
Where it went
The RSC powered entry-level workstations in the IBM RS/6000 family. Specific models included the Model 220 and Model 230. These machines used the chip to provide affordable access to the POWER instruction set. The integration of memory and I/O controllers simplified motherboard design. Systems used standard SIMMs for memory, which were cheaper and more available than custom cards. The chip also formed the basis for other specialized processors. The RAD6000, a radiation-hardened version, was derived from this design for space applications.
Versions and family
The RSC is a direct derivative of the multi-chip POWER1. It removed some features to fit on a single die, such as register renaming in the floating point unit. The PowerPC 601 is closely related, described as a cheaper and less powerful version of the RSC with added PowerPC instructions. The RAD6000 is a radiation-hardened variant based on the RSC design. These variants show how the core architecture was adapted for different markets, including consumer desktops and space exploration hardware.
Why it matters
The RSC demonstrated that complex RISC architectures could be integrated onto a single chip. It bridged the gap between high-end multi-chip servers and affordable workstations. By using standard memory components, it helped lower the total cost of ownership for IBM RS/6000 systems. The design influenced subsequent single-chip implementations of the POWER architecture. Its success in entry-level markets proved the viability of consolidating memory and I/O functions on the processor die.
Questions people ask about RISC Single Chip
What is the RISC Single Chip?
The RISC Single Chip, or RSC, is a microprocessor developed by IBM. It is a single-die version of the multi-chip POWER1 CPU, designed for entry-level workstations.
How much power does the RSC consume?
The processor consumes 4 watts of power when operating at 33 MHz. It requires a 3.6 volt power supply to function correctly within IBM RS/6000 systems.
What kind of cache does the RSC have?
It features an 8 KB unified cache that stores both instructions and data. The cache is two-way set associative and uses a least recently used replacement policy.
Which computers used the RSC processor?
The RSC was used in entry-level models of the IBM RS/6000 family, specifically the Model 220 and Model 230 workstations.
Good reading and sources
- RISC Single Chip on WikipediaEncyclopedia
- More photos on Wikimedia CommonsPhotos
- Wikidata recordOpen data
The facts on this page come from the Wikipedia article "RISC Single Chip", 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.