Macworld’s Speedmark tests of Apple’s latest high-end desktop system—the eight-core 2.8GHz Mac Pro released earlier this week—show that it’s notably faster than the four-core 2.6GHz Mac Pro it replaces in Apple’s product line. Even more impressively, the new Mac Pro comes close to matching the speed of the previous-generation eight-core 3GHz Mac Pro model, suggesting that internal changes to the Mac Pro line are bolstering performance. This new set of Mac Pros feature several internal improvements, including a faster system bus, faster memory, and two Quad-Core Intel Xeon processors as standard equipment.
However, not all the results were promising for Apple’s latest pro desktop model. The eight-core 2.8GHz system lagged in some of our tests, results we attribute to its somewhat sluggish Seagate hard drive.
As with the previous generation of Mac Pros, Apple offers up a “recommended configuration” as well as a long list of customizable options. Macworld bought the recommended configuration—a $2,799 Mac Pro with two Quad-Core Intel Xeon processors (for eight cores total) running at 2.8GHz. The machine ships with 2GBs of 800MHz DDR2 fully buffered ECC RAM, a 320GB internal hard drive, ATI Radeon HD 2600 XT graphics card with 256MB of GDDR3 memory running in a dedicated, 16-lane PCI Express 2.0 card slot.
Optional upgrades to the recommended configuration include faster processors running at 3.0GHz and 3.2GHz (for $800 and $1,600, respectively), faster graphics cards including a NVIDIA Quadro FX 5600 with 1.5GB of graphics memory for an additional $2,850. There is also an option to use 15,000-rpm Serial Attached SCSI drives. It’ll cost you $650 to upgrade your standard 7,200-rpm 320GB drive to a 300GB SAS drive and each additional 300GB SAS drive costs $800. An $800 Mac Pro RAID PCI Express card is required to use these internal SAS drives. One final configuration consists of a single Quad-Core 2.8GHz processor for $500 less than the $2,799 price of the model we tested.
We’ve ordered each one of the four Mac Pro processor configurations and plan to test each one when they arrive. For now, let’s focus on the Mac Pro we did get our hands on—the eight-core 2.8GHz model. We compared it to Apple’s previously recommended configuration, a Quad-Core (or two Dual-core) 2.66GHz model. We also included a custom configuration from the last Mac Pro generation—a desktop with two Quad-Core 3GHz Processors. Finally, our reference systems included a pair of build-to-order Macs: the 2.8GHz iMac Core 2 Duo, which has the same clock speed at the new Mac Pro, and the fastest available MacBook Pro, a 2.66GHz Core 2 Duo laptop.
2.8GHz 8-Core Mac Pro Benchmarks
| Speedmark 5 | Adobe Photoshop CS3 | Cinema 4D XL 10.5 | Compressor 3 | iMovie HD | iTunes 7.5 | Unreal Tournament 2004 | Finder | HandBrake | Pro App Multitask | |
|---|---|---|---|---|---|---|---|---|---|---|
| OVERALL SCORE | SUITE | RENDER | MPEG2 ENCODE | AGED EFFECT | MP3 ENCODE | FRAME RATE | ZIP ARCHIVE | H.264 ENCODE | SUITE | |
| Mac Pro Xeon/Dual 2.8GHz Quad Core | 314 | 0:49 | 0:16 | 0:46 | 0:30 | 0:47 | 100.8 | 3:51 | 0:54 | 1:22 |
| Mac Pro Xeon/2.66GHz Quad Core | 258 | 0:51 | 0:28 | 1:20 | 0:39 | 0:52 | 87.8 | 5:15 | 1:55 | 1:25 |
| Mac Pro Xeon/Dual 3GHz Quad Core | 317 | 0:44 | 0:15 | 1:02 | 0:34 | 0:51 | 95.4 | 3:41 | 0:57 | 0:58 |
| 24-inch iMac Core 2 Duo/2.8GHz* | 268 | 0:48 | 0:47 | 1:45 | 0:41 | 0:50 | 97.8 | 3:46 | 2:39 | 1:47 |
| MacBook Pro Core 2 Duo/2.6GHz* | 209 | 1:14 | 0:51 | 1:56 | 0:47 | 1:02 | 79.1 | 4:45 | 2:53 | 2:21 |
| >Better | <Better | <Better | <Better | <Better | <Better | >Better | <Better | <Better | <Better |
Best results in bold. Reference system in italics. * Denotes build-to-order system
Speedmark 5 scores are relative to those of a 1.5GHz Core Solo Mac mini, which is assigned a score of 100. Adobe Photoshop, Cinema 4D XL, iMovie, iTunes, and Finder scores are in minutes:seconds. All systems were running Mac OS X 10.5.1 with 2GB of RAM. The Photoshop Suite test is a set of 14 scripted tasks using a 50MB file. Photoshop’s memory was set to 70 percent and History was set to Minimum. We recorded how long it took to render a scene in Cinema 4D XL. We used Compressor to encode a 6minute:26second DV file using the DVD: Fastest Encode 120 minutes – 4:3 setting. In iMovie, we applied the Aged Film effect from the Video FX. menu to a one minute movie. We converted 45 minutes of AAC audio files to MP3 using iTunes’ High Quality setting. We used Unreal Tournament 2004’s Antalus Botmatch average-frames-per-second score; we tested at a resolution of 1,024 by 768 pixels at the Maximum setting with both audio and graphics enabled.. We created a Zip archive in the Finder from a 2GB folder. For the Professional Application Multitasking suite, we recorded how long it took Photoshop to run our standard test suite while a longer Cinema4D task and our Compressor encode test ran in the background.—Macworld Lab testing by James Galbraith, Jerry Jung, and Brian Chen
Overall, the new 2.8GHz Mac Pro was a solid performer, handily beating the previous recommended Mac Pro configuration in just about all tests. In our overall system benchmark, Speedmark 5, the new 2.8GHz Mac Pro got a 22-percent higher score than the older 2.66GHz Mac Pro. Some individual tests, like Cinema 4D and Compressor showed the new 2.8GHz with a more impressive performance advantage, with the 2.66GHz model taking about 74-percent longer to complete those tests than the new model. In other tests, like the ones involving iTunes MP3 encoding and our Photoshop CS3 test suite, performance gains between the 2.8GHz and 2.66GHz Mac Pro systems were more modest.
When comparing the new eight-core 2.8GHz Mac Pro results to the older eight-core 3GHz Mac Pro, you can see that the systems are neck and neck, with the half of our tests showing the new system to be faster and the other half giving the performance advantage to the older system. Speedmark shows the older 3GHz Mac Pro with a higher score—317, to the 2.8GHz model’s 314 tally. But consider that each core on the new Mac Pro runs 200MHz slower than those on the 3GHz model. That indicates some of the internal upgrades are having a positive effect on the 2.8GHz Mac Pro’s performance. For instance, despite the slower clock speed, the 2.8GHz model beat the 3GHz Mac Pro in our HandBrake, iTunes and iMovie and Compressor tests.
Comparing the new 2.8GHz Mac Pro to other reference systems, the new Mac Pro comes out ahead. The new Mac Pro was 50 percent faster than the MacBook Pro in overall performance and finished some tests, including Handbrake and Cinema 4D, in about a third of the time it took the MacBook Pro. The 2.8GHz iMac, with its super-fast 500MB hard drive, proved itself again to be quite a speedy all-in-one system, giving the new Mac Pro a run for its money in both the iTunes and Photoshop tests.
Hard drives proved to be a deciding factor in another test as well.
Our Professional Application Multitasking suite tests how well a system can perform while running Photoshop, Compressor and Cinema 4D simultaneously. In that test, the 3GHz system posted the fastest time by quite a noticeable margin. The eight-core 2.8GHz system took 24 seconds longer than the 3GHz model, while the quad-core 2.66GHz Mac Pro trailed the new model by just three seconds. Keep in mind that these systems are running with just 2GB of RAM (standard shipping amount for Mac Pros) and so hard drive speeds will be a factor in tests like this multitasking suite.
When we swapped the 250GB Western Digital drive in the 2.66GHz Mac Pro for the 320GB Seagate hard drive in the 2.8GHz Mac Pro, multitasking performance on the 2.66GHz model slowed by 14 percent to 1 minute, 37 seconds and sped up considerably by 31 percent, to just 57 seconds on the 2.8GHz machine (which would have bested the 3GHz Mac Pro’s multitasking suite score).
We have more of the new 800MHz ECC RAM on order and will run these tests again once we receive it. Be sure to check back soon for more Mac Pro benchmarks as our systems trickle in, and for Macworld’s full review of the new 2.8GHz 8-Core Mac Pro.
[James Galbraith is Macworld Lab director.]