Re: Looking for an application preloader...

<trryhend@xxxxxxxxx> wrote in message

What you are looking for probably does not exist.

That would be unfortunate.

One of the reasons Linux is faster than MS Windows is that we don't
not run lots of stuff in the background. If we start running lots of
applications in the background, (like Firefox), our OS will slow down
- it'll get slow and punchy and all bogged down like MS Windows.

I don't know what people do with Windows, but mine is certainly not "bogged
down". Which is not to say I haven't seen such setups, with ridiculous
amounts of background processes and useless icons in the systray, and every
bootup stalled for up to a minute while the system looks for a nonexistent
DHCP server... but a properly configured Windows box is very snappy.

Keeping it that way requires just a tiny bit of vigilance and common sense
which most users lack

You'll do well to realize that measuring the time an application loads
is not the way to benchmark a system. Some desktop users may think it
is, but it's not. Scientific benchmark tests indicate that Linux is 5
to 10 percent faster than Windows, but that does not mean Linux will
start up or load applications faster, it usually won't, but that is
largely because those applications are not already running in the

Sorry, I don't buy it. What matters is not some artificial "scientific"
benchmark of program speed; it is the speed in (the arguably immeasurable)
"real-world" patterns of use. And the speed of app launches matters in the
real world. Hiding behind artificial benchmarks isn't going to help;
ultimately the user experience is decided by what the user perceives.

Consider two cars; one gets 40 miles per gallon, the other 45. Clearly the
second car is more fuel-efficient. But if the second car, due to a smaller
gas tank, has to be refueled twice as often as the first, opinion about
which is better won't be quite unanimous. It isn't just about the numbers; I
might technically be spending less money on gas with the second car but I'll
curse it every time I'm standing out in the cold pumping gas when I'm
already late for work.

An artificial benchmark is useful, but limited.

I for one am pleased as punch to be a Linux user - I think everyone
should use Linux - but I'm not going to tell them that, or, at least
not unless they ask - I'm going to let others decide for themselves.
If they like what they see, they can try it - if they try it, they
will like it.

While I like some things about Linux, I see many shortcomings. I have been
chastised for pointing some of them out; in some cases I was told that what
I see as shortcomings are actually improvements that I just have to learn to
appreciate - sorry, but that's not the way the world works.


Relevant Pages

  • Re: project developers get the sack after converting to Open Source says ewic
    ... CELA, OEMLA, GLA (End user license agreement, Corporate User License ... In this case, between Windows and Linux. ... What is the penalty for publishing a benchmark without Microsoft's ...
  • Re: Looking for an application preloader...
    ... applications in the background,, our OS will slow down ... I don't know what people do with Windows, but mine is certainly not "bogged ... "Keeping it that way" in Linux does not take the vigilance of a MS ... Scientific benchmark tests indicate that Linux is 5 ...
  • Re: Optimal mid-range CPU selection?
    ... >> in a linux newsgroup some biased windows benchmark. ... Is your google broken perhaps? ... You are citing all the time tom's benchmark. ...
  • Re: Benchmark (Phoronix): FreeBSD 9.0-RC2 vs. Oracle Linux 6.1 Server
    ... benchmark; and while it is important to ... when it comes to a comparison to Linux. ... that "DragonflyBSD was more stable", when in fact I happen to wonder the ... application FreeBSd, were "FreeBSD shoulkd have the power to serve". ...
  • Re: Parallels vs. Fusion revisited
    ... Und denen ist die Performance herzlich egal. ... Mechanismen von sowohl Windows als auch dem Wirt-OS, ... Shared folders and mirrored folders respond slowly when accessed for ... In den Benchmark ist das womöglich überproportional ...