RISC vs. CISC at 46

What was it about, and does it matter today?





Section E

References

Glossary




References

[MAIN]

This is a long article, so here's some mellow 80s synth pop to listen to while you're reading, from an appropriately named band. I like the restored version of the album. FYI The rest of the references are real...

https://mainframe-music.info/index.html

[RISC]

The Case for the Reduced Instruction Set Computer - D. Patterson / D. Ditzel.

https://people.cs.umass.edu/~emery/classes/cmpsci691st/readings/Arch/RISC-patterson.pdf

[RISC-I]

Paper describing the Berkeley RISC-I processor.

RISC I: A Reduced Instruction Set VLSI Computer - D. Patterson / C. Sequin.

https://people.eecs.berkeley.edu/~kubitron/courses/cs252-S07/handouts/papers/p216-patterson.pdf

[RETRO]

Retrospective: RISC 1: Reduced Instruction Set Computer - D. Patterson / C. Sequin.

https://www.researchgate.net/publication/220771227_Retrospective_RISC_I_A_Reduced_Instruction_Set_Computer

[MCode]

8086 microcode is complex stuff - and the 8086 is relatively simple!

How the 8086 processor's microcode engine works - K. Shirriff.

https://www.righto.com/2022/11/how-8086-processors-microcode-engine.html

[68K]

Design Philosophy Behind Motorola's MC68000 - T. Starnes.

http://www.easy68k.com/paulrsm/doc/dpbm68k1.htm

[HLLCA]

Retrospective on High-Level Language Computer Architecture - D. Ditzel / D. Patterson.

https://people.eecs.berkeley.edu/~kubitron/cs252/handouts/papers/p97-ditzel.pdf

[801]

The evolution of RISC technology at IBM - J. Cocke / V. Markstein.

https://acg.cis.upenn.edu/milom/cis501-Fall11/papers/cocke-RISC.pdf

[MIPS-X]

Architectural Tradeoffs in the Design of MIPS-X - P. Chow / M. Horrowitz.

https://archive.org/details/DTIC_ADA182299

[CISC]

Definition from the Computer Architecture Bible:

Computer Architecture: A Quantitative Approach - J. Hennessy / D. Patterson.

[Mashey]

A deep dive into CISC and RISC instruction set complexity - J. Mashey.

https://homepages.cwi.nl/%7Erobertl/mash/RISCvsCISC

[APX]

Proposal for upping the number of x86 registers to 32 - X86 Ecosystem advisory group.

https://x86ecosystem.org/blog/apx-201/

[Ditzel]

Conference slides - D. Ditzel.

https://microarch.org/micro55/media/keynote_ditzel.pdf

[SCJ]

Iterative processor development at Bell labs - S. Johnson.

https://hoult.org/1979_Johnson_A_32_bit_Processor_Design.pdf

[EM-1]

Tanenbaum Proposes Simpler Architecture in 1978.

Implications of Structured Programming for Machine Architecture - A. Tanenbaum.

https://research.vu.nl/ws/files/110789436/11056

[6600]

CDC 6600, the first supercomputer - HandWiki.

https://handwiki.org/wiki/CDC_6600

[88K]

Apple / NeXT 88110 development - Stories of Apple.

https://www.storiesofapple.net/the-88110-cpu-and-the-risc-workstations-that-never-were.html

[EV]

Electric Vlasic (pickle).

Characterisation of Organic Illumination Systems - Various.

https://web.archive.org/web/20080706151313/http://www.hpl.hp.com/techreports/Compaq-DEC/WRL-TN-13.pdf

[060UM]

M68060 User’s Manual - Motorola 1994.

https://www.nxp.com/docs/en/data-sheet/MC68060UM.pdf

[ARM-H]

A History of The ARM Microprocessor - D. Jaggar (Talks at Google).

https://www.youtube.com/watch?v=_6sh097Dk5k

[SPEC]

List of SPEC 92 / 95 results from early/mid 90s.

https://www.cs.toronto.edu/pub/jdd/spectable

[Lightroom]

Amiga Lightroom Benchmarks (FYI: My 2.8GHz Raspberry Pi 5 runs this in 47 seconds).

https://drive.google.com/file/d/1_VxGqSt2fMQbDxdZNO-SvA5Th0gfgp3n/view?pli=1

[HOMBRE]

Hombre, the future Amiga that never was :-(

https://grokipedia.com/page/Amiga_Hombre_chipset

Full Specs:

https://archive.org/details/Hombre_201808/Hombre_Presentation_part_1/

[Piranha]

Multi-core Alpha.

Piranha: A Scalable Architecture Based on Single-Chip Multiprocessing - Various.

https://www.microsoft.com/en-us/research/wp-content/uploads/2016/12/isca00.piranha.pdf

[Patterson]

Lex Fridman interviews David Patterson (includes RISC vs. CISC, RISC-V, and Wrestling).

https://www.youtube.com/watch?v=naed4C4hfAg

[X925]

Deep dive on the X925 core used in the RTX Spark.

Arm's Cortex X925: Reaching Desktop Performance - C. Lamb.

https://chipsandcheese.com/p/arms-cortex-x925-reaching-desktop

[Bench]

See below.


[M1RE]

Apple M1 Reverse Engineered - caohuajia.

https://github.com/caohuajia/apple-m1/blob/main/vol1%20M1%20Explainer.nb.pdf

[OLY]

Nvidia Olympus Architecture

https://nvdam.widen.net/s/nmw5vblpqd/nvidia_vera_cpu_architecture_whitepaper?nvid=nv-tblg-543584

[FAB]

Intel make chips for Apple, SpaceX and Nvidia.

href="https://www.macrumors.com/2026/07/13/apple-agreed-intel-chips-trump-tariff-talks/

[Amdahl]

Amdahl's law

href="https://lawsofsoftwareengineering.com/laws/amdahls-law/

[EAG]

x86 ecosystem advisory group.

https://x86ecosystem.org/

[Bench]

Contemporary benchmark links:

SPECrate2026 integer base rate-1 (Estimated Performance):

https://benchview-hjc-im.translate.goog/?_x_tr_sl=auto&_x_tr_tl=en&_x_tr_hl=en-US&_x_tr_pto=wapp&_x_tr_hist=true#dataset=data%2FSPECrate2026_int_base.csv

Above is linked from David Huang's Blog:

https://blog.hjc.im/spec-cpu-2017

PassMark Single Thread:

https://www.cpubenchmark.net/single-thread/

Cinebench 2024 Scores:

https://www.cpu-monkey.com/en/cpu_benchmark-cinebench_2024_single_core

Cinebench 2026 Single-threaded:

https://www.cpu-monkey.com/en/cpu_benchmark-cinebench_2026_single_thread

Geekbench 6 Single-threaded:

https://www.cpu-monkey.com/en/cpu_benchmark-geekbench_6_single_core

Early Nvidia Vera benchmarks, single + multi-threaded:

https://www.phoronix.com/review/nvidia-vera-benchmarks




Glossary

Note: Some of these terms are difficult to define as different vendors often use different terms and to add to the confusion they are often used interchangeably.


AArch64 / Arm64

The 64 bit instruction set defined by Arm. AArch64 is the official name, Arm64 is more informal.


Addressing mode

This determines how a value is interpreted in an instruction.


ARM / Arm

The company that defines the Arm architecture (i.e. instruction set architecture) and designs and licenses Arm processors (and more). Other companies such as Apple and Qualcomm design their own processors based on the Arm instruction set.

Note: "ARM" changed to "Arm" in a 2017 rebrand.


Cache

A small on-chip memory that is used to speed up access to data and instructions.


Cache line

A sub-division of memory in a cache. The smallest contiguous block of memory that can be transferred between memory and cache.


Code density

How much memory a piece of code takes up. Variable length instructions typically have higher code density.


Decode / decoder

The stage where an instruction is interpreted. This determines what operation it represents.


Exception / Fault

An exception is an internal event triggered by the processor while executing an instruction. It tells the processor to stop what it's doing and deal with what caused the exception.

A fault is a type of exception. See Interrupt.


Fetch

The stage where a processor gets (fetches) an instruction (or instructions) from memory into an internal buffer.


Instruction format

Describes the structure of an instruction.


Instruction set / Instruction Set Architecture (ISA) / architecture

The defined set of processor instructions and any effects they have.


Interrupt

An external signal that tells the processor to stop what it's doing and deal with what caused the interrupt. See exception.


Micro-architecture

The design or layout of the physical chip.


Micro-op (Micro-operation)

The basic operations that the processor performs. Many kinds of microprocessors, including RISC processors break instructions into micro-ops.


Memory page

Memory is typically divided into multiple blocks called pages. This is used to enable multiple-users, multitasking, and security.


Microcode

A type of code executed inside the processor. Usually involved in implementing an instruction, especially complex instructions.


Out-of-Order execution

Technique used by modern microprocessors to increase performance, involves executing instructions out of sequence where possible.


PAL code

DEC's name for a technique similar to micro-code but typically much simpler.

Used in many processors, including RISC processors.


Pipeline

A series of stages of execution in a processor operating like an assembly line. This enables multiple instructions to be in different stages of execution simultaneously.


Register

A small fast storage location inside a processor. Typically 32 or 64 bits long.


Superscalar

A processor that can issue more than one instruction per clock cycle.


Variable length instructions

Instructions whose size varies. These typically have high code density.





RISC vs. CISC at 46, Section A

RISC vs. CISC at 46, Section B

RISC vs. CISC at 46, Section C

RISC vs. CISC at 46, Section D

RISC vs. CISC at 46, Section E



© Nicholas Blachford 2026