Speedometer 3.0: A Shared Browser Benchmark for Web Application Responsiveness

Contributors from Apple, Google, Microsoft, and Mozilla, writing for BrowserBench: Since the initial version of the Speedometer benchmark was released in 2014 by the WebKit team, it has become a key tool for browser engines to drive performance optimizations as users and developers continue to demand richer and smoother experiences online.

We're proud to release Speedometer 3.0 today as a collaborative effort between the three major browser engines: Blink, Gecko, and WebKit. Like previous releases (Speedometer 2 in 2018 and Speedometer 1 in 2014), it's designed to measure web application responsiveness by simulating user interactions on real web pages. Today's release of Speedometer 3.0 marks a major step forward in web browser performance testing: it introduces a better way of measuring performance and a more representative set of tests that reflect the modern Web.

This is the first time the Speedometer benchmark, or any major browser benchmark, has been developed through a cross-industry collaboration supported by each major browser engine: Blink/V8, Gecko/SpiderMonkey, and WebKit/JavaScriptCore. It's been developed under a new governance model, driven by consensus, and is hosted in a shared repository that's open to contribution. This new structure involves a lot of collective effort: discussions, research, debates, decisions, and hundreds of PRs since we announced the project in December 2022.

Speedometer 3 adds many new tests. We started designing this new benchmark by identifying some key scenarios and user interactions that we felt were important for browsers to optimize. In particular, we added new tests that simulate rendering canvas and SVG charts (React Stockcharts, Chart.js, Perf Dashboard, and Observable Plot), code editing (CodeMirror), WYSIWYG editing (TipTap), and reading news sites (Next.js and Nuxt.js).

AI?

[dragande]

I'm utterly astonished, not a single whisper about AI in this entire post...

Re:Good

[narcc]

Performance wouldn't be a problem at all if modern development practices weren't so astonishingly poor. Between the complete lack of thought given to memory management and the countless third-party libraries that themselves include countless third-party libraries, it's a wonder that things work as well as they do.

User time is more important than developer time!

I don't think anyone under 40 has thought about how much of their own time they waste trying to avoid writing code.

Better instrumentation tools are good to see.

Ironically, "better" tools are what enable the monstrosities created by today's agile "software development by accretion".

Re:

[AmiMoJo]

The developers aren't entirely to blame here. Management demands things be done quickly and cheaply, and how much of your RAM they waste is of little concern to the bottom line.

Re:

[narcc]

I'm talking about the performance implications of poor memory management, not the amount of memory needed.

As for "demands things be done quickly and cheaply", I explain in my post that developers waste considerable time and effort in foolishly trying save time and effort with countless third-party libraries. That could be a manager problem in some cases, but odds are good that its a developer problem.

Re:

[ls671]

Well, "Web Application Responsiveness"? We didn't need much benchmarking before web application became as bloated as they are today when web developers knew what they were doing and understood how the network and other things work.

Re:

[ls671]

Infinity before your web app responds? Interesting! /s

I don't need a benchmark suite to know they suck

[Seven Spirals]

Turn off Javascript, override all CSS with your own, reject cookies outright across the board. If you can manage that, you might have a decent web experience if the site doesn't break completely or try to render itself in a narrow line down the left hand side of your screen. However, since those technologies have become synonymous with web applications, they are slow, janky, and extremely untrustworthy. I don't trust corporations to fluff each other then try to assure me everyone is above board (people lik

Re:

[IWantMoreSpamPlease]

I have yet to run across a website that works properly without JS and/or CSS enabled. Most can work (somewhat) with JS disabled/blocked, but without CSS give it up, the site is busted beyond recognition.

Re:

[Seven Spirals]

I just clobber their CSS with some known-decent stuff, but yeah, breaks about 40% of all sites I visit beyond being able to read them. If so, I leave.

Is it the app that is slow?

[QuietLagoon]

... or having to load up all those slow-loading ads?

SeaMonkey

[king*jojo]

Doesn't seem to even run the test

I guess that's the best possible score

score interpretation

[DarthVadersSaber]

can anyone share some insight how to interpret the score that was shown? it seems like a random arbitrary score is shown and details just show how long something takes to load. in short what score is good and what is bad?