Epihany is pretty good since 3.0, very stable.Midori became pretty much an
Elementary browser and looks a little odd now, but Epiphany is well integrated,
it looks like a cute mini Firefox.I believe dependencies are like
Midori.Epiphany used to be not as good as Midori but the GNOME-devs
On 11/05/2011 05:51 PM, Julien Lavergne wrote:
So far, I don't see advantages to switch to Firefox. Of course, if
the benchmarks show a real difference, I may reconsider my opinion.
Tom's Hardware regularly benchmarks browsers, but sadly only under
Windows and OS X. The relevant page (Memory
Are those benchmarks from Tom's Hardware documented at all? Maybe we could
rebuild those benchmarks for linux? When they are run on a low-spec
machine then the results should be the same on other machines. User tests
should be done to verify the results aren't just laboratory results.
I might be
I have a P4 PC with 512MB RAM but I can drop that to 256.
I'm ready for such tests ;)
Also, I think it's important to ask the users to vote for that, not only
us. I can start a thread on Ubuntu Forum but I need to hear first from
Julien and others about that.
On Sun, Nov 6, 2011 at 11:45 AM,
I'm not very fan of a change unless we have a strong consensus for another
browser. But, this is my opinion :
- We should only consider Chromium or Firefox, because they have a strong
upstream and are well maintained. Browser is a very important part of a
system, and we really don't have
I also agree. Both browsers are widely accepted work equally well for general use. I think the question should be a) do we want to pick one and stick with it for the sake of consistency or b) always choose the lightest of the two?Tim-- Sent from my HP TouchPadOn Nov 5, 2011 9:07 PM, Julien
Le 11/06/2011 09:24 AM, Ali Linx a écrit :
Also, I think it's important to ask the users to vote for that, not
only us. I can start a thread on Ubuntu Forum but I need to hear first
from Julien and others about that.
Oh please don't, I imagine the mess and trolls for this tread :)
And also,
@ALI, is there a specific way that you're gonna make your benchmarks?
I'm saying it so we can compare our results :)
Whatever in your mind, let me know and I'll be more than glad to do it :)
Don't have specific ideas at the time being. I'm willing to open many tabs
and monitor that. Don't
Oh please don't, I imagine the mess and trolls for this tread :)
And also, it's not only a question of popularity, we need to consider
other aspects.
Yes, I realized it could be a bad idea for now. Let's first decide, we as a
team then we can go public.
So, I just sent to Jean-Pierre. Any
Le 11/06/2011 08:45 AM, Jonathan Marsden a écrit :
If (and this is a big IF) these results are similar to those in Lubuntu
on low-RAM machines, it might indeed be time to rethink use of Chromium.
Yes, it missed 2 points :
- memory management on Linux (we can't rely on Windows one to do proper
Thanks for the tests. You can also use about:memory of Chromium, to have
a second source of information.
Regards,
Julien Lavergne
Le 11/06/2011 04:16 PM, Jean-Pierre Vidal Piesset a écrit :
Just to get an idea, i've installed and run on my new laptop Epiphany
3.0.4, Firefox 7.0.1, Chromium
Just finished some testing using:
Chromium 15.0.874.102
Firefox 7.0.1
No extensions, no tab history (tabs are clean).
1 tab (logged into gmail)
chromium-browser: 50.1 + 24.7 + 9.1 = 83.9 MB
firefox: 97.2 MB
2 tabs (logged into gmail, ubuntuforums.org)
chromium-browser: 54.5 + 24.7 + 9.1 = 88.3
Thank you for your test.
I, however, have some doubt about the chromium test.
Since midori, epiphany, and chromium all uses webkit, what makes so many
differences?
The most memory-demanding parts should be webkit and the graphics.
I don't believe that the remaining parts can cause so much impact.
Le 11/06/2011 05:52 PM, PCMan a écrit :
Since midori, epiphany, and chromium all uses webkit, what makes so
many differences?
As far as I know, Chromium uses a patched webkit, that's probably why
the tests could be different between midori, epiphany and chromium.
Regards,
Julien Lavergne
Proposal for browser benchmarks:
- First of all, run the browsers without any add-on
- Take starts time
- Watch lxtask for memory usage cpu. Inside Chromium about:memory shows
the memory used by each browser running too. Htop?
- Watch the general behaviour
- Sites: youtube, gmail, google,
On Sun, 6 Nov 2011 14:29:10 -0300
Jean-Pierre Vidal Piesset jpx...@gmail.com wrote:
Proposal for browser benchmarks:
- First of all, run the browsers without any add-on
- Take starts time
- Watch lxtask for memory usage cpu. Inside Chromium about:memory shows
the memory used by each
*Maybe chromium has some hidden resource usage not covered by the test?*
To avoid this kind of doubt, I think it is better to make a new boot, open
the Chromium Browser (with tabs and common pages), and see how much of CPU
and RAM the *whole system* are using.
Then you reboot and do the same
Just a question.
Are we going to use some kind of tools/apps to do these tests? or we'll go
for the manual approach? I mean just open LXTask and take the figures from
there?
I'm going to do a fresh install for Lubuntu 11.10, run apt-get update and
then apt-get upgrade, install Firefox and make
I was suggesting some kind of script, which could write the results in a
log file. That way all the results will be the same on all machines and
are objective. I will need some help in creating that script probably, but
I'll be on IRC tomorrow and see what I can get bunged together.
With metta,
Make sure you are doing your tests with a clean browser profile. An
easy, safe way to do that is to create a new user and login to the new
account.
Jeremy
___
Mailing list: https://launchpad.net/~lubuntu-desktop
Post to :
On Sat, 5 Nov 2011 20:51:02 -0400
Julien Lavergne gi...@ubuntu.com wrote:
Hi,
I'm not very fan of a change unless we have a strong consensus for another
browser. But, this is my opinion :
- We should only consider Chromium or Firefox, because they have a strong
upstream and are well
Chromium-browser automatically checks for the latest version AFAIK. And the
script should be able to retrieve the latest version from the respective
vendor sites. If the installed version (also callable from the system)
doesn't match the latest version, then the test will fail.
With metta, Chris
If you can create such a script, it helps a lot in other areas, too.
We can use it to do benchmark for other components of lubuntu as well.
On Mon, Nov 7, 2011 at 2:55 AM, Chris cyber.dr...@gmail.com wrote:
I was suggesting some kind of script, which could write the results in a
log file.
On 6 November 2011 14:45, Chris cyber.dr...@gmail.com wrote:
Chromium-browser automatically checks for the latest version AFAIK. And the
script should be able to retrieve the latest version from the respective
vendor sites. If the installed version (also callable from the system)
doesn't match
I like the idea of a script.. at least for most of it. The one bugaboo
will be doing start up and shut down tests. Maybe we could use an
appropriate signal? I'm not sure which one would be used to cleanly
shutdown the program, but, for example, with SIGTERM we could do:
kill -n 15 $(pidof
25 matches
Mail list logo
