Is Jaguar any faster than the previous version of OS X?
People have been going back and forth about this around the
Net without anyone sitting down to do a dispassionate examination
of the new OS's performance potential.
To take some of the mystery out of the question, we decided
to put Jaguar through it's paces. We used our standard suite
of tests, the one that we run on each new Macintosh we benchmark.
Though our test suite is fairly processor focused, it includes
tests that will check drive & graphics performance. We
added a specific test that would check out Apple's Quartz
Extreme performance claims. Both OSes were installed on a
Dual G4/1GHz with 1 GB of RAM
The Quartz
Extreme feature of Jaguar relieves the processor from
most of the on-screen graphics processing. It shunts graphics
rendering over to the graphics card, which is much more efficient
at that type of work.
Quartz Extreme is the main performance improvement of OS
10.2. To take advantage of it you need to be running Jaguar on
a computer that has one of the following graphics cards: NVIDIA
GeForce2 MX, GeForce3, GeForce4 MX, or GeForce4 Ti or any
AGP-based ATI RADEON GPU. A minimum of 16MB VRAM is required
installed in a 2X or better AGP slot.
Scrolling and window redrawing & movement, should be
much faster in Jaguar than it is under 10.1. While we found
this to be the case for some of the Apple applications we
tested (such as the Preview App), other applications, such
as Internet Explorer showed little, if any improvement. Our
hunch is that applications need to be tweaked to take advantage
of Quartz Extreme and that Apple is ahead of the curve on
having, at lease some, of their applications up to speed.
In any case we found that Quartz Extreme worked as advertised
by Apple, significantly speeding up some on-screen graphics
and giving a moderate boost in other instances. So it looks
like this feature has potential.
In raw processing power, Jaguar has changed nothing. In our
processor intensive tasks, both operating systems turned in
basically the same score. However there were some little performance
increase surprises. iTunes 3 is significantly faster at converting
songs to the MP3 format. Classic boots much faster. And other
minor performance improvements popped up here and there.
On the downside, drive performance might be slightly poorer
under Jaguar, though it is pretty minimal, and one of our
game tests showed lower frame rates.
However another game showed better frame rates ... go figure.
Probably some difference in how each game balances processor
and graphic card resources.
Feedback:
"'It is our hunch that applications have to be written
to take advantage of Quartz Extreme"
No, applications automatically take advantage of it.
However, Quartz Extreme is simply a much, much faster
compositor. It can draw windows on top of each other a
lot faster than before. Creating the windows is pretty
much the same process it always was.
"We ran three different types of tests using Giants,
and Jaguar came in behind in all of them."
Games are typically not going to be affected by Quartz
Extreme, but if they are rewritten to take advantage of
the much better OpenGL extensions in 10.2, they get a
TON faster.
We haven't had a chance to rewrite Giants for this.
The first title we have coming out in which we're hoping
advantage of 10.2's OpenGL extensions is Freedom Force
(http://www.omnigroup.com/games/freedomforce/), so if
everything goes according to plan it should be quite a
bit zippier on newer graphics hardware.
The tests below are from our suite of real
world application tests. These tests feature a diverse selection
of applications commonly used by the Mac community. The test
suite was designed to render an accurate and well rounded
picture of a machine's performance. All of the tests below,
except for the game tests, were timed with a stopwatch. The
times were then converted to percentages, relative to OS 10.1.3,
which is set to 100%. For all scores, higher numbers are better.
Quartz Extreme?
For the test above we opened a document in Apple's 'Preview'
application and scrolled from one end to the other. The Preview
application can be found in the OS X Applications folder on
your hard drive. As you can see above Quartz Extreme is ....
well, pretty extreme [update: improved scrolling performance
is actually a result of Quartz 2D ... a companion feature
of Jaguar. You can read more about Quartz Extreme & Quartz
2D here].
We are not sure that is all that is going on here. This is
version 2 of the Preview application and there might be other
aspects of the application that were changed to make it perform
better. However, the Quartz Extreme function of Jaguar is
suppose to package screen graphics and off -load them for
the graphics card to deal with. This leaves the processor
free to take care of other tasks. We indeed saw this effect
during this test. Processor usage on our dual processor 1
GHz Tower dropped from 50% down to 5%, and the time to complete
the scroll dropped from about 86 seconds to a little over
2.
[Note: Apple claims a 2 to 3 times improvement when carrying
out functions that can take advantage of Quartz Extreme or
Quartz 2D]
A Photoshop scroll takes place while the application
is running in the Classic OS 9.2.2 layer
Desktop Tests
One of the speed improvement of Jaguar is the
speed at which Classic boots ... a real boon for when some
errant program running in Classic brings the whole Classic
layer crashing down. Classic boot time was about 18 seconds
in the updated OS
Slightly diminished performance, but probably
close to the margin of error
Same as above ... close to the margin of error
Don't have an explanation for why we are seeing
so much improvement here. Approximate 55% of the processing
capability of the machine is being used to carry out the above
function
Within the margin of error
Slightly better on screen graphics performance.
The load on the processor was the same under both tests leading
us to believe that the Quartz Extreme effect was not taking
place here. It is our hunch that applications have to be written
to take advantage of Quartz Extreme ... it doesn't fall like
ripe fruit when you throw an application at it. It has to
be picked by programmers. Someone please
correct us if this is not the case. However Jaguar is
also suppose to give a slight boost to graphics, even without
Quartz Extreme
Large Document & Database
Type Tests
A Macro
(series of complex actions) was run in Microsoft's Word program,
which is part of Office
X
A Macro (series of complex actions) was run
in Microsoft's Excel program, which is part of Office
X
Stresses the processing & memory systems
of the machine. This test takes place in a large AppleWorks
document. This is a raw processing power test.
Number Crunching & Rendering
Tests
The Fractal
program has been highly tuned to take advantage of the
G4 and is precisely the type of work that the G4 was made
for. It will also gobble up whatever processing capability
is present. This is a good test for assessing the fundamental
processing potential of each machine.
A Ripple Effect is applied to an iMovie
Encoding/Decoding Tests
A Sorenson
encode compresses a QuickTime movie for streaming on the Web
Preps QuickTime Movie for import into iMovie.
We believe that some change was made to the default settings
for DV Stream export function of QuickTime 6, and that the
diminished performance of the Jaguar machine was not due to
the OS. The machine when running 10.1.3 had QuickTime 5.0
installed
The better performance here is probably due
to improved encoding performance of iTunes 3 ... nice to see.
Under 10.1.3 iTunes 2.1 was used.
Multitasking
A Sorenson encode and fractal render are performed
at the same time. The QuickTime encode only soaks up 50% of
the processing power of each machine, leaving plenty of capability
left over for work to be done on the Fractal. The scores are
pretty much within the margin of error.
MP3 Encode, AppleWorks search & replace and
folder copy are all carried out at the same time. Better MP3
encoding puts Jaguar over the top
Gaming
Aside from the graphics card, Giants
is also tuned to take advantage of the processor as well.
Indeed, when run on a dual processor machine, you will see
Giants fly
ahead of the same game running on a single processor machine.
We ran three different types of tests using Giants, and Jaguar
came in behind in all of them. Frame rates here are 46 fps
for 10.1.3 and 41 fps for Jaguar.
However is was a completely different story
when running EVO2 4x4. Here fps were 24 for 10.1.3 and 30
for Jaguar
For Great Prices On Upgrades Check The Vendors Below
HandHelditems.com - Personalize your iPod with us. Shop hundreds of unique iPod accessories and save up to 80%.
Apple Store
- The size of a pack of gum, iPod shuffle weighs less than a car key. Which means there’s nowhere your skip-free iPod shuffle can’t go. Click Here
Copyright 1996-2007 by Cider Press Publishing LLC all rights reserved.
MacSpeedZone is not authorized, sponsored, or otherwise approved by Apple
Computer. Apple, the Apple logo, Macintosh, iPod, iBook, iMac, eMac, and
PowerBook are registered trademarks of Apple Computer, Inc. Additional
company
and product names may be trademarks or registered trademarks and are hereby
acknowledged.
For Great Prices On Upgrades Check The Vendors Below
