[vox-tech] Browser speed

[Peter Jay Salzman]

On Sun 13 Feb 05,  7:18 PM, Karsten M. Self <kmself at ix.netcom.com> said:
> on Fri, Feb 11, 2005 at 07:30:17PM -0500, Peter Jay Salzman (p at dirac.org) wrote:
> > In case you haven't read /. yet...
> > 
> > http://www.howtocreate.co.uk/browserSpeed.html#linspeed
> > 
> > I'm not even slightly surprised at the results.
> 
> I'm not even slightly impressed by the relevance.
> 
> "Speed" defined as "time until foo process is completed" is easy to
> measure.  It's also often highly irrelevant, though not fully.
> 
> If a browser allows the user to _start_ reading a page before it's fully
> loaded, TTR (time to read) rather than TTL (time to load).  But it's
> harder to measure.
 
True.  But I've always said that, in my perception, Opera is the fastest
to both begin rendering and finish rendering a typical webpage.

> I also see a significant differnece in browser responsiveness when task
> switching, window resizing, or pulling up dialogs.  For which Mozilla
> and Firebird fare far poorer than Galeon (1.3.x).

Galeon.  Ugh.  It was so right, and went so horribly, horribly wrong.

> Other factors of browser usability don't really relate to speed at all:
> 
>   - Stability.
> 
>   - Features.
> 
>   - UI access (a combination of features and how hard they are to get
>     to).
> 
>   - Crash / session recovery.
> 
>   - User preferences:  bookmarks, passwords, forms, etc.
> 
>   - Security.
> 
>   - Privacy.
 
Yeah.  Really, Opera wins most of that, except for stability, which is
pretty much a trump card and why i'm in week 3 of my Firefox experiment.

A gamer always puts a grain of salt in benchmarks but they're always fun to
look at.   ;)  In this case, the benchmark gibed with my personal
observation on my own system.

Pete

-- 
Save Star Trek Enterprise from extinction: http://www.saveenterprise.com

GPG Fingerprint: B9F1 6CF3 47C4 7CD8 D33E  70A9 A3B9 1945 67EA 951D