🚀 go-pugleaf

In <docjoe-2008950813260001@ip9.wilmington.de.interramp.com> docjoe@interramp.com (Joe Ragosta) writes:

>In article <danpop.808872265@rscernix>, Dan.Pop@mail.cern.ch (Dan Pop) wrote:
>
>> I definitely get more work done on Linux than on MacOS, no matter how
>> much faster the Mac processor is.  Your studies are worth zilch to me.
>
>OK. I'll just take your word for it. I'm sure you've run a controlled
>side-by-side study comparing MacOS to Linux. Have you ever even used a
>Mac?

Of course not.  Everybody knows you're the only Mac user around.  No other
mortal had the privilege to use that wonderful thing.

>BTW, have you considered the fact that the applications that are used in
>the real world aren't available on Linux? How does that affect
>productivity?

All the applications used in the segment of real world I'm working are
available on Linux.  Many aren't on the Mac, or don't provide equivalent
functionality.  I couldn't care less about the applications I don't use
and I don't need.  As I made clear in my previous article, I was speaking
only for myself.  Everybody else's mileage may vary.
>> 
>> >Besides, if the MacOS is so slow, can you explain why the PowerMac 9500
>> >was approximately twice as fast as a fast Pentium in side-by-side
>> >application tests?
>> 
>> That's easy.  What OS was the Pentium running?  What applications were
>> involved in the tests?  You can prove anything you want to prove by using
>> "real world" applications.  All you have to do is to choose the applications
>> according to what you want to prove.  This is why the industry recognizes
>> the SPEC suites and not "side-by-side application tests" as a measure of
>> performance.  Get a clue.
>
>Bullshit. No one buys a computer to run SPEC suites. They buy computers to
>run applications. How fast the computer runs the applications you use is
>the impotrant issue. You could argue about which applications to test and
>that's OK. The best test for someone contemplating buying a computer is to
>test the applications they're going to be running (which probably doesn't
>included SPEC).

As usual, you're wrong.  If you want to buy a machine to run a single
app, you can benchmark that app and decide which machine is better.
If you want to buy a general purpose machine, SPEC is your best way to
compare the relative performances.  At least, this is the industry
consensus.  Of course, you know better.

And many people's applications are similar to one or more of those from
the SPEC suite.  They can request the individual results and base their
choices on the results of the relevant apps.

>This has been done by a number of computer magazines and independent labs.
>In all cases, the PowerMac 9500 was approximately twice the speed of a
>fast Pentium. If you're running Excel, the difference will be about 15%.
>If you're running Photoshop, it will be >>100%.

I'm running neither Excel nor Photoshop nor any other kind of "office
productivity" package.
>
>If you're so concerned about the effect of OS on the Pentium, where are
>your studies showing that OS/2, NT, or Linux are so much faster? Do you
>really believe that your favorite OS will make up for a 2X difference
>between a Pentium running WIndows and a PowerMac?

It depends on the nature of the application being run, of course.  If the
app is cpu intensive, the OS won't make much of a difference.  If it's
I/O bound, the OS will make a significant difference.  And even more so
if I have to run more than one application at a time, all of them competing
for cpu and I/O resources.

Dan
--
Dan Pop
CERN, CN Division
Email: Dan.Pop@mail.cern.ch 
Mail:  CERN - PPE, Bat. 31 R-004, CH-1211 Geneve 23, Switzerland