Flopify - when would your device have been the fastest on Earth?

Ever wondered how your smartphone compares to a supercomputer from the past? Compare your phone to the computer that sent the Apollo rockets to the moon and many others 🚀.

Click your device in the dropdown or in the chart to see when it would have been the most powerful in the world.

We’ll constantly add new devices, so check back to see the latest updates.

How to explore this data

The chart tracks computing power (FLOPS) across history, from home devices to record-breaking supercomputers.

👆 Click a point on the chart or use the dropdown to explore a specific device.
📶 The trendline is based on the fastest computer of each year, click a device below the trendline to see when it would have been the most powerful in the world.
🙈 Use the filters at the top to show or hide different types of devices.
🤯 Switch to the linear scale to better understand the exponential growth of computing power over time.

What are FLOPS?

A FLOPS (Floating Point Operation per Second) is a measure of a computer’s ability to perform arithmetic calculations involving decimal numbers (floating-point numbers). Specifically, it refers to how many mathematical operations—like addition, subtraction, multiplication, or division—a processor can complete per second.

More FLOPS generally mean more raw computational power, but there are important caveats:

⚡ Power Efficiency – Not all FLOPS are equal—some devices achieve higher performance per watt, which isn’t reflected in raw FLOPS.
🧠 Specialised chips – Some chips may seem slower but are designed for highly to be effective for a very specific task which may not be reflected in its FLOPS (e.g. Neural Processing Units are highly optimised for specific AI tasks like machine learning).
💸 Cost per FLOPS – Supercomputers push the limits of performance, but consumer devices often offer better FLOPS per dollar.
⚛️ Quantum Computing: Quantum processors, like IBM’s Eagle, don’t use FLOPS but instead measure performance through qubits and quantum gate complexity. Their power scales differently, making direct comparisons with classical computers tricky.

Note on Accuracy: FLOPS can be measured in different ways, and some devices (particularly supercomputers) are upgraded over time. Additionally, certain devices are sold in multiple configurations. We’ve aimed to be consistent with the FLOPS scores we use and have selected the most accurate numbers available.

Interesting machines in history

Apollo Guidance Computer - powering missions to the moon

Developed by NASA for the Apollo space program, the Apollo Guidance Computer was a digital computer produced for the Apollo program that was installed on board each Apollo command module (CM) and Apollo Lunar Module (LM). The AGC provided computation and electronic interfaces for guidance, navigation, and control of the spacecraft and was a pioneering achievement in computing for its size and weight limitations during the 1960s moon missions.

British Codebreakers Colossus - first electronic digital computer

Colossus was the world’s first programmable electronic digital computer, designed to break the Lorenz-encrypted cipher used by the German military during World War II. It was composed of over 2,000 vacuum tubes, weighed over a ton and was also the first to use stored programs. Colossus kept secret for many years after World War II, with the UK government even denying their existence until the 1970s.

Zuse Z1 - one of the first mechanical computers ever built

The Zuse Z1 was a motor-driven mechanical computer designed by Konrad Zuse reading instructions from punched celluloid film, the Z1 was one of the first programmable computers, designed to perform basic arithmetic operations and store results, marking a significant step towards modern computing.

El Capitan - the world's fastest today

Currently the world's most powerful supercomputer. Developed for Lawrence Livermore National Laboratory, is expected to achieve peak performance of over 2 exaFLOPS. Its focus will be on advancing scientific research and national security applications, built on AMD’s cutting-edge CPUs and GPUs.

The Difference Engine - designed in 1822

Designed by Charles Babbage in 1822, the Difference Engine was an early mechanical computer intended to automate the calculation of mathematical tables - it was hand-cranked. While it was never fully constructed during Babbage’s lifetime due to technical and financial challenges, a working model of the first version was finally completed in 1991 at the Science Museum in London, demonstrating that Babbage’s design would have worked. The machine couldn't handle floating-point operations (it did integer arithmatic) so technically didn't have a FLOPS but it could make around 1 calculation every 180 seconds.

IBM Watson - AI supercomputer which won Jeopardy!

AI system known for winning first place prize of $1 million on Jeopardy! in 2011, using natural language processing and machine learning to answer questions. It was powered by IBM’s POWER7 processors and relied on a massively parallel architecture, enabling it to process vast amounts of data in real-time for applications in healthcare, customer service, and more.

Folding@home - distributed computing

Distributed computing project that uses the power of millions of personal computers around the world to simulate the folding of proteins, which is crucial for understanding diseases like Alzheimer’s, cancer, and COVID-19. It allows individuals to contribute their computer’s processing power to perform highly complex computations on protein structures that would otherwise require immense computational resources, with participants contributing by running software on their own devices.

The stories behind Flopify

Watch the story behind the technology and how your phone can beat supercomputers from history costing billions - and my journey building Flopify: