Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Dear Chrome, Slow Your Roll (massivegreatness.com)
15 points by ssclafani on Sept 28, 2012 | hide | past | favorite | 8 comments


Something that people who are not terribly familiar with browsers often don't get is that it's not all about the browser. Content plays a major role.

I hear "Firefox 1.0 was fast and lean and had no bloat" all the time but if you download Firefox 1.0 and run it, you'll find it awfully slow and clumsy. (And the same is true for Chrome 1.0.)

So, what's changed. It's the content, stupid.

Back in 2004, when Firefox 1.0 shipped, a web page was usually a simple document that was a couple hundred kilobytes. Today, a "web page" is an application that might be multiple megabytes even if it's just a seemingly simple list of status updates.

Back in 2004, you might have had one web app, maybe Gmail, open all the time in a tab. Today, you probably have several of these web apps open in tabs all of the time, Gmail, Facebook, Twitter, Spotify, etc. and they're considerably larger and richer than they were 8 years ago.

Yes, browsers have added features. But they've also added massive amounts of performance and efficiencies that were absent in earlier browser releases. It's not as simple as "browsers get bloated over time."


Take every piece of software and OS, and you will find conflicting views of whether it goes faster or slower with time. You buy a new device, everything feels snappy, but one year down the line, things feel slow. But the problem is, this is based on expectations, based on hardware, based on usage pattern, etc.

What we need is a objective software speed measure based in UI terms (because that is what people in the end notice). Such as "time to start" and "time to get to usable state". This should be measured automatically after each build, and normalized to hardware. Only then can both developers and users see hard evidence that software, e.g. Chrome, is now running slower than before. Because the cost of adding a feature is otherwise invisible to the developer, while the user's keep getting into anecdotal discussions of what is faster or slower, comparing wildly different hardware capabilities.


The data disagree with you. Pick a benchmark, any benchmark, and you'll see significant improvements in Chrome performance over time. It starts faster, it renders pages faster, it runs JS faster. I can't comment on stability, because I haven't seen the data and because Chrome is highly reliable for me and everyone I know.


For me, turning plugins such as flash into "click to play" seemed to speed up things a lot in osx / chrome. The setting is a bit hidden, but you can find it from settings -> show advanced -> content settings -> plug-ins.


If the slow down is only noticed on OSX, it's a OSX problem especially when OSX 10.7.5 just came out. Most application that runs perfectly on window tend to run poorly on OSX.

... Here comes it hate mail :(


I'm curious, what does the author of the post, or others, think Chrome added that was unnecessary?


Agree. The author complains a lot without naming any specific features that are considered "bloated." As a web developer, I love all the features that Google rolls into Chrome with each update; especially in the developer tools section.


+1 this post seemed devoid of any real examples to support the complaint. I've been quite happy with Chrome for awhile now. If they have been adding unnecessary bloat, I haven't noticed.




Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: