[MacDV] Re: G5 - is it really that fast?

Frank Flynn frank at declan.com
Wed Jun 25 13:09:21 PDT 2003


On 6/25/03 12:12 PM, "Macintosh Digital Video List"
<MacDV at lists.themacintoshguy.com> wrote:

> I read an article yesterday linked from MacNN about the G5 being not
> that fast. According to the article (written by a Mac user), the
> VeriTest used by Apple to compare speeds was manipulated a bit to
> Apple's favor. Despite this, Apple still didn't win out in everything.
> ...[snip]...  Anyone else see this and maybe have the link?

Yes - the page I think you're referring to is:

  http://www.haxial.com/spls-soapbox/apple-powermac-G5/

The article seem to raise some valid points about how the tests were
conducted to the G5's advantage and to the Pentium's disadvantage.

For example:
  "Apple/Veritest used a special fast malloc library on the G5
   benchmark, but did not use it on the Dell/Intel benchmark..."
Or
  "For both the Dell Dimension 8300 and the Dell Precision 650,
   Apple/Veritest performed the multi-processor "Rate" benchmarks
   with hyperthreading DISABLED...."

Not that it's unheard of for a vendor to try to skew tests of their products
in their direction just that if the G5's were as super fast as Apple / IBM
claim this should not be necessary.

It seems many Apple fans are flaming the heck out of this guy (well what did
he expect - praise?).  Personally I'd suggest not doing that, rather let's
wait until these units are available to _independent_ organizations who can
benchmark them in a more open and fair manner.  The truth will be obvious
then.

F

-- 
Frank Flynn
Poet, Artist & Mystic




More information about the MacDV mailing list