Tuesday, October 1, 2002

For Great Prices On Upgrades Check The Vendors Below
Processor Upgrades
PowerMax	Small Dog
Graphics Cards
PowerMax	Small Dog
Hard Drives & Optical Drives
PowerMax	Small Dog
RAM & Other Memory
4AllMemory	MacConnection
SCSI, Firewire USB Cards
MacConnection
Closeout, Used & Refurbished Macs
Apple	ebay
PowerMax (refurb)	PowerMax (used)
Small Dog
Get your store listed in this space!

by David Engstrom

Below we have a performance comparison between the new Power Mac dual G4/1000 and the old Dual Processor G4/1000. The new machine costs $500 less than the old machine cost when it was first released. From a performance standpoint, have things changed much between the old and new?

Aside from drive performance there hasn't been a whole lot improvement. Our folder copy test showed a 55% speed improvement over the older GHz machine and our single file test a 13% improvement.

Otherwise, where there was improvement, it was pretty minimal. 4% in our on-screen graphics tests, 8% in burning CDs.

In some tests the new GHz machine came in behind the old one, like our AppleWorks Search & Replace test, where the new machine lagged the old by 5%.

In basic hardcore processing, both machines turned in the same score and the new machine did pull ahead by 17% in our Excel test.

Gaming performance was a mixed bag. Running Giants was 2% slower on the new machine, but that only translated to 1 frame per second. Quake III and 4x4 EVO2 both ran somewhat faster on the newer Tower.

One thing effecting the performance of the new Tower is that it only has half the L3 cache per processor that its predecessor has. We think that this held back the new Tower from performing better in some of the tests.

Another thing to note is that the old machine was running OS 10.1 and the new machine 10.2. Both the default operating systems that came with each machine. How much of a factor the new OS played in some of the scores below is something we will be testing soon (we have one of the old versions of the GHz Towers on its way to us). We don't believe that it would effect most of the test results below, but may have been a factor in some of the tests related to graphics performance, and perhaps the drive tests as well .... we will know shortly. In any case the new OS should have been an advantage to the new machine, and as it turned out, for the most part, they were pretty evenly matched.

We also have a 1.25 GHz machine in hand, and so should be able to provide you with detailed performance information on this machine in the very near future.

Difference and similarities of each machine

	Power Mac Dual G4/1000 (Winter 2002)	Power Mac Dual G4/1000 (current)	Additional Resources Power Mac Buyers Guide Power Mac InfoZone Power Mac Specifications and Reviews Page Power Mac Reviews Power Mac Performance Benchmarks Performance of all machines compared Performance article index Check prices at: .... Apple Store Clubmac MacConnection MacMall Macwarehouse MacZone Small Dog
Processors	G4/1000	G4/1000
L2 Cache	256k @ 1000 MHz	256k @ 1000 MHz
L3 Cache	2MB DDR SRAM per processor	1MB DDR SRAM per processor
Bus Speed	133 MHz	167 MHz
Memory	1 GB PC133 SDRAM	1 GB PC2100 266MHz DDR SDRAM
Graphics Card	NVIDIA GeForce4 MX 64MB of DDR SDRAM 4x AGP slot	ATI Radeon 9000 Pro with 64MB of DDR SDRAM
Drive	80GB Ultra ATA (7200 rpm)	80GB Ultra ATA/100; 7200 rpm
CD Drive	2x 6x8x 4x24x	2x 6x8x 4x24x
Operating System	10.1.2	10.2
Price	$2,999	$2,499

"Real World" Tests

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 the Power Mac Dual G4/1000 released in the Winter of 2002, which is set to 100%. For all scores, higher numbers are better.

Desktop Tests

Drive performance on the new Towers seem to be one area of improvement

It is SuperDrive vs SuperDrive

On screen redraws are only slightly better in the newer machine. . Let 1K Windows Bloom is a simple carbon application that opens and closes 1,000 windows.

Large document is scrolled from one end to the other using Classic OS 9.2.2 when booted in OS 10. Test shows the performance of on-screen graphics. It is possible that the older Tower is helped in the two test above by the fact that it has 2 MB of L3 cache per processor as opposed to only 1 MB per processor on the newer machine

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. and is a raw processing power test. The new Tower is hurt, we believe, by the smaller amount of L3 cache in the machine

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.

Encoding/Decoding Tests

A Sorenson encode compresses a QuickTime movie for streaming on the Web

Multitasking

A Sorenson encode and fractal render are performed at the same time

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. Frame rates here are 48 fps for the older machine and 45 fps for the current one. Again this may be a case where the new machine was hurt by the amount of L3 cache it has.