Google Chrome Smashes Speedometer 3 Record With Massive Performance Gains

BrianFagioli writes: Google is flexing its engineering muscles today by announcing a record-breaking score on the Speedometer 3 benchmark with its Chrome browser. If you've felt like the web got snappier lately, this could be why.

According to the search giant, Chrome's latest performance improvements translate to real-world time savings. Believe it or not, that could potentially add up to 58 million hours saved annually for users. That's the equivalent of about 83 human lifetimes not wasted waiting for web pages to load!

A proper adblocker saves even more time!

[tliet]

Now that uBlock Origin is blocked, I'm wondering who still (voluntary) runs Chrome as their main browser.

Re: A proper adblocker saves even more time!

[ThurstonMoore]

Never have, never will.

Re: A proper adblocker saves even more time!

[Z00L00K]

In many cases the slowness experienced isn't the browser, it's the servers you try to access.
It's like measuring the horsepower of an engine while you have iron wheels on a dirt road. On a rainy day you don't have any traction at all.

Re: A proper adblocker saves even more time!

[JamesTRexx]

it's the servers you try to access

With servers, I presume you mean the websites built on bloated javascript frameworks and full of spyware links to social and anti-social media.
To paraphrase the old saying, what network hardware giveth, ever more bloated inane websites taketh away.

The only default permission in eMatrix (on Pale Moon) is CSS for the 1st party. Anything else must be neccessary and earned, and it has kept the browser snappy as wel. (includes uBlock to filter annoying elements)

Re:

[phantomfive]

Does PaleMoon allow you to prevent auto-video play?

Re:

[JamesTRexx]

https://forum.palemoon.org/viewtopic.php?t=25646/ [palemoon.org]

Re:

[AmiMoJo]

A lot of it is add-ons too. Taking uBlock Origin as an example, on Firefox the pattern matching is done in Javascript and performs worse than old Chrome, which did it with C, and much worse than new Chrome which just takes the patterns and matches them internally.

It's still well worth running uBO because the gains outweigh the cost, and the gap used to be much wider, especially on mobile. Still, it's an area where Firefox has been struggling to keep up for years. It's only really the fact that Chrome ditche

Re:

[phantomfive]

It varies a lot of course, but a common load profile is 2-4 seconds for network, then 6-8 seconds for rendering. Modern websites have to compile a lot of Javascript (like, the entire react framework or something. If the Javascript is improperly jammed together with WebPack, then ironically it's much, much slower, for Javascript that isn't even used on the page).

Re:

[gweihir]

There is always a large number of people without even basic insights...

Re:

[Tony Isaac]

uBlock Origin Lite still does a decent job, despite not being as capable as uBlock Origin.

Great news for Brave users

[Echoez]

.... because Brave will be even faster when combined with all of the ad blocking

Re:

[phantomfive]

Does Brave prevent videos from auto-playing?

A better measure of Google's efficiency

[hyades1]

If there was a benchmark for the speed with which a corporation loots and sells your personal information, I'm sure Google/Alphabet Inc. would be right up there at the top.

Re:

[KGIII]

Google is absolutely collecting your personal information.

They aren't really selling it, however. You can't go to Google and say, "Here's ten bucks. Let me see KGIII's information."

They're selling access to your metrics, however. You're profiled and assigned a market segment with pretty good accuracy. They know who you are, what your interests are, what your interests really are, all the pages you've gone to, etc.. (Royal you, not you specifically.)

The people paying Google are paying them to target people i

They probably just cheated again.

[devslash0]

There was an article here on slashdot some time ago about browser creators building in benchmark detection and special tweaks to perform better in benchmark environments.

It was probably the same this time.

If all they care about is benchmarks, they will build benchmarks that are good at benchmarks, not everyday use.

Re: and yet

[Anamon]

I keep hearing this online (though never ever from friends or coworkers). And I wonder how it's possible for experiences to be so different.

For me, Chrome is one of the most sluggish, laggy, leaky pieces of software I've ever used. It struggles to hande more than 10 open tabs, the UI keeps freezing, and it crashes all the time. This is true on all of my machines, across Windows and Linux, from a 20-year-old Thinkpad to a 2024 overpowered workstation.

Firefox, meanwhile? Handling my regularly four-digit open

Massive?

[avandesande]

Maybe if browser speed was measured in decibels....

Not interested in Googlenet Explorer

[xack]

All the real browser innovation is elsewhere, Chrome won't even implement vertical tabs.

it might sound small

[toxonix]

"Chrome’s engineering team claims to have improved the browser’s performance on this test by 10 percent since August 2024.
That might sound small, but it’s not."

Keep telling yourself that buddy.

Re:

[markdavis]

>" 'Chromeâ(TM)s engineering team claims to have improved the browserâ(TM)s performance on this test by 10 percent since August 2024. That might sound small, but itâ(TM)s not.'

Keep telling yourself that buddy."

LOL! This "Smashes" and "Massive Performance Gains" is just 10% on some benchmark! Hmm. I mean, perhaps that is impressive if it translates to real-world effect, but I wouldn't attribute the superlatives the article did.

Well, doesn't matter to me, I won't be using any Google-base

Re:

[Tony Isaac]

The headline wouldn't be nearly as dramatic without words like "smashes" and "massive". How else would they get people to click?

Re:

[Guignol]

Revolutionary AI assisted Slashdot headers crafting obliterates previous techniques, boosting activity by never seen before up to ten more clicks per month

So Google admits to murder?

[ukoda]

If "That's the equivalent of about 83 human lifetimes not wasted waiting for web pages to load!" then is Google effectively admitting that all those ads they serve are wasting human lifetimes?

We need leaner, better websites

[HnT]

There, I said it. We need way better and leaner websites without the unbelievable amounts of jank and bloat that has somehow become the sad normal nowadays.
I feel like I went from a slow aDSL in the 90s to gigabit and websites nowadays do not load any faster, because they bloated to the size of fully functional operating systems - and this will only get worse with more frameworks, wizards and vibecoding.

Re:

[viperidaenz]

Even with a corporate firewall blocking the ads here, DevTools reported this page as requiring 1.5MB of data downloaded. There's only 13 comments so far.

It's not just ads, it's the websites are now made poorly. Usually with Javascript rendering the content to ensure ads get displayed before content, so you stare at an ad while you wait.

I see what they did

[viperidaenz]

Areas powered by Oilpan garbage collection saw expanded usage to avoid reliance on slower memory allocators like malloc.

Garbage collection is quicker than malloc/free when you don't collect any garbage until after the benchmark has finished.

Re:

[gweihir]

Indeed. Lies, damned lies and then benchmarks.

Re:

[dsgrntlxmply]

If I recall correctly, Henry Baker ("List Processing in Realtime on a Serial Computer"), in a talk on his GC work, was asked if this was currently in use on Symbolics machines. The answer was something like "most people just save the world then reboot".

Wrong metric

[msauve]

>the equivalent of about 83 human lifetimes not wasted waiting for web pages to load!

That's miniscule compared to the human lifetimes wasted when Facebook does load.

Who cares?

[gweihir]

Browsers have been more than fast enough for at least half a decade, possibly more. And a whopping 83 human lifetimes per year? That is so tiny, it does not even register when you look at the biger picture.

Re:

[93 Escort Wagon]

Yeah, I'm trying to figure out the last time I thought the browser's innate speed was actually the problem. Typically, when a page is taking a long time to load, it seems to be an issue with communication between my laptop and the server.

And, with Chrome, there might be quite a few servers involved in loading a particular site's web page - most of which you probably don't know about.

Or they could inject fewer crap ads

[rumpledoll]

Or they could inject fewer shit coded ads. Page loading times have be constant from somewhere in the mid-2000's. Only time they were snappy was when the tag and was an issue.

Benchmarks for endusers are pointless now

[alfabravoteam]

Last benchmark that showed real benefit to users was ACID test. Telling off about the trash that IE always had been. Nowadays, gains are marginal on JS engines and rendering, while devs fall on same mistakes of old, relying on frameworks' magic to pull it all out, regardless of things like NPM dependency hell, lack of caching strategies, linting and so on. We had ugly images for a long time just to ensure no heavy resources were used; now everyone brute-forces networks because at least it's not a stream.

you are Little Red Riding Hood....

[Big Hairy Gorilla]

..and I'm the Big Bad Google!

So much the better to ingest all that delicious yummy data.

What?

[Virtucon]

Did they stop tracking every page you visit? Truly implement an incognito mode? Implement do not track as the default?

Big numbers

[algaeman]

If 2 billion people each save 2 minutes a year, that is 67 million hours saved annually. What would easily save even more money would be to prevent advertisers from writing huge profiling cookies or running cpu-burning javascript in order to make a text element blink.