Clock frequency (CPU) over time
Nominal operating clock. Says nothing about work done per cycle and must never be used as a proxy for performance across architectures.
- Device scope
- cpu
- Method
- Nominal clock
- Plotted in
- hertz
Chart
73 figures from 65 machines, spanning 1966 to 2025. The lowest clock frequency here is 0.98525 MHz (Commodore 64, 1982) and the highest is 4.41 GHz (Apple Mac mini (M4, 2024), 2024). That is a range of about 4476× within one comparability group; the figures are plotted in Hz. Points use the release date of the machine the figure belongs to. The table below lists absent figures with their recorded reasons.
Data
| Machine | Released | Figure | Confidence | Sources |
|---|---|---|---|---|
| Apollo Guidance Computer (Block II) Apollo Guidance Computer Block II processor | 1966 | 2.048 MHz Master oscillator rate, stated by the report as 2.048 megapulses per second. | Vendor-rated | |
| Saturn Launch Vehicle Digital Computer Saturn Launch Vehicle Digital Computer | 1966 | 2.048 MHz Four clocks per bit, giving 512 kilobits per second in serial operation. | Vendor-rated | |
| Apple II Configuration with unspecified RAM | June 1977 | 1.023 MHz | Vendor-rated | |
| Atari 2600 NTSC launch configuration | September 1977 | 1.19 MHz | Vendor-rated provisional | |
| Intel 8086 PC Intel 8086 | 1978 | 5 MHz | Vendor-rated provisional | |
| Atari 800 NTSC, 48 KiB RAM | November 1979 | 1.79 MHz | Vendor-rated provisional | |
| Atari 800 PAL, 48 KiB RAM | November 1979 | 1.77 MHz | Vendor-rated provisional | |
| IBM Personal Computer 5150 Original 16 KiB base configuration | August 1981 | 4.77 MHz | Vendor-rated | |
| BBC Micro Model B Retail Model B | December 1981 | 2 MHz | Vendor-rated | |
| Intel 80286 PC Intel 80286 | 1982 | 12.5 MHz | Vendor-rated | |
| ZX Spectrum 48K PAL 48K | April 1982 | 3.5 MHz | Vendor-rated provisional | |
| Commodore 64 NTSC | August 1982 | 1.02273 MHz NTSC machines. The manual states this as the system frequency underlying its RS-232 baud-rate formulas. | Vendor-rated | |
| Commodore 64 PAL | August 1982 | 0.98525 MHz PAL machines. The manual states this as the system frequency underlying its RS-232 baud-rate formulas. | Vendor-rated | |
| MSX MSX1, 64 KiB RAM | 1983 | 3.58 MHz | Vendor-rated provisional | |
| Apple Macintosh 128K Original retail Macintosh | January 1984 | 7.8336 MHz | Vendor-rated | |
| Amstrad CPC 464 PAL retail machine | June 1984 | 4 MHz | Vendor-rated provisional | |
| Intel 80386DX PC Intel 80386DX | 1985 | 25 MHz | Vendor-rated provisional | |
| Atari 520ST 520ST retail configuration | June 1985 | 8 MHz | Vendor-rated provisional | |
| Nintendo Entertainment System NTSC retail console | October 1985 | 1.66 MHz As stated by Nintendo for the console, with no region named. | Vendor-rated | |
| Sega Master System NTSC retail console | June 1986 | 3.58 MHz | Vendor-rated provisional | |
| Commodore Amiga 500 NTSC | 1987 | 7.15909 MHz NTSC machines with the stock system clock, derived from the 28.63636 MHz base crystal. | Vendor-rated | |
| Commodore Amiga 500 PAL | 1987 | 7.09379 MHz PAL machines with the stock system clock, derived from the 28.37516 MHz base crystal. | Vendor-rated | |
| Acorn Archimedes A310 A310 retail configuration | June 1987 | 8 MHz | Vendor-rated provisional | |
| NEC PC Engine Japanese retail console | October 1987 | 7.16 MHz | Vendor-rated provisional | |
| Sega Mega Drive NTSC retail console | October 1988 | 7.67 MHz | Vendor-rated provisional | |
| Intel 80486DX PC Intel 80486DX | 1989 | 33 MHz | Vendor-rated provisional | |
| Nintendo Game Boy Original retail handheld | April 1989 | 1.05 MHz | Vendor-rated provisional | |
| SNK Neo Geo AES AES retail console | April 1990 | 12 MHz | Vendor-rated provisional | |
| Super Nintendo Entertainment System NTSC retail console | August 1991 | 2.68 MHz As stated by Nintendo, without naming the memory region the rate applies to. | Vendor-rated | |
| Commodore Amiga 1200 NTSC | October 1992 | 14.32 MHz NTSC machines. Half the 28.63636 MHz NTSC base crystal, which gives 14.31818 MHz exactly. | Vendor-rated | |
| Commodore Amiga 1200 PAL | October 1992 | 14.19 MHz PAL machines. Half the 28.37516 MHz PAL base crystal, which gives 14.18758 MHz exactly. | Vendor-rated | |
| Commodore Amiga 4000 CYBERSTORM 68060 accelerator | October 1992 | 50 MHz | Vendor-rated | |
| Commodore Amiga 4000 Standard 68040 | October 1992 | 25 MHz | Vendor-rated | |
| Intel Pentium PC Intel Pentium | 1993 | 100 MHz | Vendor-rated provisional | |
| Sega Saturn PAL retail console (MK-80200-50) | November 1994 | 28.4 MHz Master and slave SH2 of the PAL console, model MK-80200-50. | Vendor-rated | |
| Sony PlayStation Retail console | December 1994 | 33.8688 MHz System clock of the customized R3000A, per table 2-1. Identical in every region. | Vendor-rated | |
| Nintendo 64 Retail console | June 1996 | 94 MHz R4300 CPU, from the manual's clock speeds and bus bandwidth list. | Vendor-rated | |
| Intel Pentium II PC Intel Pentium II | 1997 | 300 MHz | Vendor-rated provisional | |
| Silicon Graphics Octane Octane 1X 250MHz R10k | 1997 | 250 MHz MIPS R10000, processor chip revision 3.1, as reported in the SPEC result configuration. | Vendor-rated | |
| Silicon Graphics Octane Octane 2X 250MHz R10k | 1997 | 250 MHz Both processors run at the same rate. The clock is per processor, not a total. | Vendor-rated | |
| Sega Dreamcast Retail console | November 1998 | 200 MHz SH-4 internal clock. The external bus runs at 100 MHz and the peripheral clock at 50 MHz, all derived from a 33.3 MHz PLL output. | Vendor-rated | |
| Sony PlayStation 2 Retail console | March 2000 | 294.912 MHz Emotion Engine core clock. Identical in every region, the console having no video-standard-derived clock. | Vendor-rated | |
| Game Boy Advance Original retail handheld | March 2001 | 16 MHz | Vendor-rated provisional | |
| Nintendo GameCube Retail hardware | September 2001 | 485 MHz The IBM "Gekko" processor of the shipped console. | Vendor-rated | |
| Microsoft Xbox Retail hardware | November 2001 | 733 MHz | Vendor-rated | |
| Intel Pentium 4 PC Intel Pentium 4 | 2002 | 3.0 GHz | Vendor-rated provisional | |
| Nintendo DS Original retail handheld | November 2004 | 67 MHz | Vendor-rated provisional | |
| Microsoft Xbox 360 Retail hardware | November 2005 | 3.2 GHz | Vendor-rated | |
| Intel Core 2 Duo E6600 PC Intel Core 2 Duo E6600 | July 2006 | 2.4 GHz | Vendor-rated provisional | |
| Nintendo Wii Retail hardware | November 2006 | 729 MHz | Vendor-rated provisional | |
| Sony PlayStation 3 Launch console | November 2006 | 3.2 GHz PowerPC core and all SPEs, per the announcement of May 2005, eighteen months before the console shipped. | Vendor-rated provisional | |
| Apple iPhone (1st generation) | June 2007 | 412 MHz The single ARM1176JZF-S core of the shipped phone, deliberately set below the core's capability to save power. | Vendor-rated provisional | |
| Apple iPhone 3GS | June 2009 | 600 MHz | Measured provisional | |
| Intel Core i7-2600K PC Intel Core i7-2600K | January 2011 | 3.40 GHz | Vendor-rated provisional | |
| Nintendo Wii U Retail hardware | November 2012 | 1.24 GHz The three PowerPC 750-derived cores of the shipped console. | Vendor-rated provisional | |
| Apple iPhone 5s | September 2013 | 1.3 GHz | Vendor-rated provisional | |
| Microsoft Xbox One Retail hardware | November 2013 | 1.75 GHz The eight Jaguar cores of the shipped console. | Vendor-rated | |
| Sony PlayStation 4 Launch console (CUH-1000 series) | November 2013 | 1.6 GHz The eight Jaguar cores of the shipped console, as reported by two independent technical publications. | Vendor-rated | |
| Apple iPhone 6s | September 2015 | 1.85 GHz | Vendor-rated provisional | |
| Nintendo Switch Docked mode | March 2017 | 1.02 GHz The four Cortex-A57 cores of the shipped console, which run at the same rate in both modes. | Vendor-rated provisional | |
| Nintendo Switch Handheld mode | March 2017 | 1.02 GHz The four Cortex-A57 cores of the shipped console, undocked. | Vendor-rated provisional | |
| Apple iPhone X | November 2017 | 2.39 GHz | Vendor-rated provisional | |
| Apple iPhone 12 | October 2020 | 2.99 GHz | Measured provisional | |
| Apple Mac mini (M1, 2020) | November 2020 | 2064 MHz The four Icestorm efficiency cores, which AnandTech calls Thunder cores. | Measured | |
| Apple Mac mini (M1, 2020) | November 2020 | 3.2 GHz The four Firestorm performance cores, in single-threaded work and, given thermal headroom, in all-core loads too. | Measured | |
| Microsoft Xbox Series X Retail hardware | November 2020 | 3.8 GHz Eight Zen 2 cores with simultaneous multithreading disabled. | Vendor-rated | |
| Apple iPhone 15 Pro | September 2023 | 3.78 GHz | Measured provisional | |
| Intel Core i9-14900K PC Intel Core i9-14900K | October 2023 | 3.2 GHz P-core base frequency; the E-core base frequency is 2.4 GHz. | Vendor-rated provisional | |
| Raspberry Pi 5 | October 2023 | 2.4 GHz | Vendor-rated | |
| Apple Mac mini (M4, 2024) | November 2024 | 4.41 GHz | Measured provisional | |
| Nintendo Switch 2 Docked mode | June 2025 | 1.101 GHz | Measured provisional | |
| Apple iPhone 17 | September 2025 | 4.26 GHz | Measured provisional | |
| Apple iPhone 17 Pro | September 2025 | 4.26 GHz | Measured provisional | |
| IBM System/360 Model 50 | April 1964 | unknown The checked documents state internal cycle durations, but do not identify a nominal CPU oscillator frequency. | not plotted | — |
| DEC PDP-8 DEC PDP-8 processor | 1965 | unknown The checked brochure specifies a computer-cycle duration rather than a nominal processor oscillator frequency. | not plotted | — |
| Apple Mac mini (M1, 2020) | November 2020 | not applicable | not plotted | — |
| Sony PlayStation 5 | November 2020 | not applicable Sony specifies the processor as running at a variable frequency with a ceiling. There is no fixed frequency to state, which is a fact about the machine rather than a gap in the sources. | not plotted | — |
Why this chart holds one group only
Every point on a chart is implicitly compared with every other point on it. This chart uses one comparability group: one metric, one device scope, one method . There is no chart of all clock frequencies over time because they are not one quantity.
Points use the release date of the machine a figure belongs to, including figures recorded against a part. A release date applies only to a machine that shipped. What this metric measures.