Trying out Chromium (Google Chrome)

Due to tiredness with Firefox eating all my memory, getting out-of-memory killed for eating all of my memory, getting kind of slow (probably because it ate all of my memory) and an annoying layout bug in one website (which strangely didn’t appear on any other machine using Firefox except mine); I have decided to give Chromium a go (also, Melbourne is currently full of Google Chrome ads, this may have contributed to my decision).

Initial niggle was that it doesn’t seem to have a concept of minimum font size exposed in the UI. You can edit a config file to solve this, which is significantly less annoying than the extension that resizes your page 5 seconds after it loads.

Most of the extensions I use are available for Chrome, except I haven’t found an analogue for the very useful highlighter extension.

About Danielle

Danielle is an Australian software engineer, computer scientist and feminist. She doesn't really work on GNOME any more (sadly). Opinions and writing are solely her own and so not represent her employer, the GNOME Foundation, or anyone else but herself.
This entry was posted in Uncategorized. Bookmark the permalink.

20 Responses to Trying out Chromium (Google Chrome)

  1. Jon Smirl says:

    Chrome uses much more memory than Firefox. At least double. My Chrome right now is using 3GB of memory on a 24GB machine.

  2. Danielle says:

    @Jon: looking at some metrics, it seems like memory usage between Firefox 3.6 and Chrome are very similar, but that Chrome is better at releasing memory (because most of the memory is tied in up in the tab process). I’m also hoping that instead of the whole browser getting OOM killed, it will only be a single run-away tab.

  3. Jeff Walden says:

    What layout bug on what site?

  4. Danielle says:

    @Jeff: it was a layout bug on the new autostraddle.com. One image decided to wrap onto a row by itself. I found it strange that I couldn’t reproduce the problem on Firefox on another computer.

  5. Peter Dietas says:

    I don’t know but seems to me like I can never get used to Chrome, I’ve tried for a couple of weeks and is nothing like Firefox. Also I have to agree with Jon it takes a lot of memory our of your machine :(

  6. Janne says:

    I use both, at the same time, on fairly modest hardware. Frankly, I don’t notice much of a difference in speed resource use or responsiveness.

    Why do I use both? Accident, mostly. I slightly prefer the UI in Chrome; it’s a bit less cluttered and gives me more vertical space. On the other hand, I have at least one Firefox extension (Zotero) that I completely and utterly depend on, so I’m not about to dump Firefox either anytime soon. And now I have separate bookmarks in both browsers, so I use one browser for some sites, and the other for other ones.

    In practice I stay about 85% Firefox, 15% Chrome. If there was a way to seamlessly sync bookmarks between them I’d probably be closer to a 50-50 split I guess.

  7. Ben says:

    You’re referring to “Minimum Font”? Yeah, it’s certainly not the best solution. It doesn’t resize the page, though, it only changes the font sizes (minor nit-pick). You can thank me for it, or not. ;-)

    Anyway, Chrome 10 (I think) is supposed to have an option in the UI, which should make my extension obsolete, and editing the config file unnecessary. Until then….

  8. ethana2 says:

    I’ve been using Chrome and Chromium for a while now, but I have a really strong tendency to go to Epiphany and Midori until something drives me away, usually instability.

    I love WebKit, I love V8, and so on– but I’m also a sucker for AppMenu Bar and RGBA translucency compatibility, which means, basically– native GTK or bust. I also like using system codecs because media sandboxing is complete overkill on non-windows platforms for security in my opinion and I’d rather not waste the memory and disk space– assuming complete stability, Midori is my favorite. In practice, Chrome wins out.

  9. Jeff Walden says:

    I don’t hit any problems with a 4.0 nightly on that site, or at least I don’t hit anything that’s obviously a problem (I’m interpreting what you say to mean the long column of rows has a row that’s double-height). If you can pin down an exact build that reliably causes issues, send me an email and I’ll try to figure out what’s up and/or point Gecko hackers at it.

  10. Jeff Walden says:

    And while I’m talking, it would be great if there were some way to follow comments here without doing a one-off feed subscription or something just for this entry’s comments. (That’s the only reason I suggest sending email if you can pin it down — better for everyone to have the knowledge than just me, but I can’t promise I’ll follow comments here faithfully enough to see a comment if you make one, too long from now.) I use Subscribe to Comments on my own blog, after having seen others using it, and I’ve been quite happy with it.

  11. Danielle says:

    @Jeff: I’ve been meaning to try a Firefox 4.0 beta.

    Also, thanks for the pointer about the Subscribe to Comments plugin. Turns out that’s already available in this WordPress install, so I’ve just turned it on.

  12. Anonymous says:

    I have yet to figure out how people end up with Firefox using a giant pile of memory. I’ve left it running for days, opening and closing hundreds of tabs, with dozens open at any one time, and I’ve never seen it use unreasonable amounts of memory. Given the piles of interest at the moment in improving Firefox performance, you might consider reporting your observations as abnormal.

    You might also review your list of addons.

  13. me says:

    But chrome has policy “no prefereces/options. We know better than you what is the right configuration”

  14. muelli says:

    May I point you to that: http://developers.slashdot.org/story/06/02/14/2154224/Firefox-Memory-Leak-is-a-Feature

    If Firefoxes memory consumption is a problem for you, simply turn the features like the cache off :)

  15. Hm, I’ve been using Chromium for several months now, and there is a setting for this in the preferences, and has been for several stable builds. I’m using Fedora 14, so YMMV. When I go to preferences, I look “Under the Hood,” and then at the custom fonts settings. As far as I can tell, Chromium is respecting my settings there for a minimum font size. Note that the settings dialog itself has changed somewhat with newer builds (I’m on 679 now, but I think it was 634 before).

  16. Ben says:

    Here’s the Chrome bug, for those interested:
    http://crbug.com/7417

  17. jeff says:

    I’m glad to see someone else independently confirming the findings I posted on Planet GNOME back in September: http://jeff.ecchi.ca/blog/2010/09/19/free-my-memory/

    (Executive summary: Firefox is more “memory-efficient” per tab, but Chromium frees my memory *entirely* on tab close, which is what matters at the end of the day)

  18. Danielle says:

    @Paul: That’s for the default font size, not the minimum.

  19. Danielle says:

    Turns out you can reproduce the rendering bug I experienced in Chromium, just set your minimum font size high enough and it must cause a wrapping problem.

    I actually discovered my fonts were set to 96dpi instead of 120, which might have helped explain why some things were rending so small on this screen.

  20. skierpage says:

    My memory experience is exactly like yours with Firefox 4 nightly builds and now SeaMonkey 2.1betas on Kubuntu 10.10. I usually have 6-10 tabs open, and every few hours the browser eats up all available memory (1 GB RAM). My computer would be unusable for 15 minutes of non-stop paging while I waited for the oom-killer to kick in. I switched to SeaMonkey to conserve memory and finally added `ulimit -v 1280000` to .zshrc to make the browsers crash faster. I wish I knew what is triggering it!

Comments are closed.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>