Text 9411, 119 rader
Skriven 2006-01-28 10:52:54 av Mike '/m' (1:379/45)
Ärende: Wow! The Intel iMac Is Almost As Fast As The Quad Core Power Mac
========================================================================
From: Mike '/m' <mike@barkto.com>

http://www.macspeedzone.com/html/hardware/machine/performance_in_the_raw/06/1_2
3.shtml

===
Mac Performance In The Raw - Wow! The Intel iMac Is Almost As Fast As The Quad
Core Power Mac - How Macworld Pulled A "Not So Fast" One ...


Wednesday, January 25, 2006


We are pleased to report that our testing results show that the new Dual Core
Intel iMac, which clocks in at 2X 2.0GHz is almost as fast as the current
high-end Power Mac that has two Dual Core G5 processors running at 2.5GHz.

How can this be, you ask ... An ostensibly 2 processor machine nearly keeping
up with a faster, ostensibly 4 processor one? Easy, we used the same
methodology employed in the Macworld "First Look" review of the new iMacs, and
applied that to our comparison. (they found the Intel iMac, in general, only
10% to 25% faster than a similar speed single core processor G5 iMac)

But before we go any further, lets look at the astounding numbers, that prove
our case ... the incredible performance, of the Intel iMac ....

We found the following

When running a QuickTime encode the Power Mac Quad G5/2.5GHz took 84.85
seconds.
The Intel iMac Core Duo 2.0GHz took 97.02 seconds Advantage: Power Mac by 14%
.... Nothing to write home about ... Not even keeping up with the clock-speed
difference between the two machines


Not convinced .... I wasn't either ... Ok lets try something different. Lets
run two encodes at the same time .... just for fun. It is easy to do, just
duplicate the file and run the processes concurrently.

What scores did we get?

When running the QuickTime encodes the Power Mac Quad G5/2.5GHz took 86.25
seconds.
The Intel iMac Core Duo 2.0GHz took 176.60 seconds Advantage: Power Mac by 105%


Ok let's get this straight when doing twice the work it only takes the Power
Mac with its four processor cores, about 2 extra seconds, but takes the Intel
iMac an extra whopping 79+ seconds - almost twice as long as in the single
test?

What's wrong with this picture? What's wrong is processor capacity vs processor
usage.

If you visit your Utilities folder, in the Application folder on your Mac
(assuming you are running OSX), you will find a small application called
"Activity Monitor". From the Window menu of Activity Monitor you can launch a
window call CPU usage.

This will give you visual feedback about how much, of the processing capacity
of your machine, is being utilized at any given time. When we speed trial any
machine, we have the CPU usage window open while we go about our testing,
making note during each test of how much of the processor(s) are put to use.

Guess what we found for the two tests outlined above?

In the first test, where there was just one file being encoded, the Intel iMac,
on average, was using 87% of its processing capacity ... 13% was sitting around
with nothing to do

On the other hand the Quad G5 Power Mac was using less than half its capacity,
42%. A full 58% was waiting for its dance card to be filled.

When we ran the two QuickTime encodes at the same time, processor usage moved
to 87% for the Power Mac, and 100% for the Intel iMac. In other words the iMac
was maxed out, and the Power Mac had 13% capacity left before it would really
start to sweat.

This is where the Macworld "First Lab Tests" article falls a little flat ...
obscuring the processor capacity vs processor usage problem inherent with
mutiprocessor machines (or multi-core ... same difference). Using Macworld's
logic we could argue, given the data above, the Quad G5 Power Mac is only 14%
faster when running some of Apple's own applications. We think that this is
misleading, as we pointed out.

There are precious few applications that take complete advantage of multiple
processors, and of those only certain actions can use huge amounts of
processing capacity. If you are using this type of application, you probably
know it already.

We have long argued that, to really take full advantage of multiprocessor
machines, you need to be in a production type of environment, and have a
strategy for utilizing the significant resources these computers make
available. It is possible to do this, and we fault the Macworld article for not
pointing it out ... this was, after all, one of the reasons for OSX.

A multiprocessor machine will provide the user (even the casual user), with
more flexibility. However, truth be told, for most of us, most of the time, it
is over-kill. You need a kitten, and you're getting a tiger.

The Macworld article does say the machines have potential (we argue that
potential is already there), and also makes the point that old applications,
that do not run natively on the Intel processor, run about half as fast as they
would run on a G5 machine (which is what we found also).

From our perspective we have established that the iMacs are at least twice as
fast as their G5 counter parts, if you harness all the power. To us it appears
that the hardware is up to the task, but that there is some bottleneck in the
software (either at the application, or OS level), that prevents full
utilization of multiprocessors by single applications (in most cases). Because
you are able to max out the processors (even on the Quad machine), it seems
like the hardware can get data fast enough to the processors, so the capacity
is lacking elsewhere....
===

 /m

--- BBBS/NT v4.01 Flag-5
 * Origin: Barktopia BBS Site http://HarborWebs.com:8081 (1:379/45)