|Page (1) of 1 - 05/23/06||email article||print page|
13-Inch MacBook Benchmarks 1Apple Final Cut Studio rendering and encoding tests
For this first round of benchmarks tests, we'll compare the 13-inch MacBook with a dual 2.0 GHz G5 and a 2.16 GHz MacBook Pro running various rendering and encoding tests in Final Cut Studio 5.1. The applications in Studio are, of course, Universal Binaries, so they're designed to run natively on PowerPC and Intel hardware. The results are surprising. Like the 15-inch MacBook Pro, the 13-inch MacBook surpasses the performance of a dual 2.0 GHz G5 in many cases. But it also holds its own against the MacBook Pro, coming in just slightly slower in most tests. Some components of Mac OS X have been updated since we ran our MacBook Pro benchmarks, and, as a result, the MacBook comes in faster in some tests--significantly in some Compressor encoding tests.
For these benchmarks, we ran a 2.0 GHz 13-inch MacBook against a 2.16 GHz 15-inch MacBook Pro and a dual 2.0 GHz G5 desktop. All of the systems were equipped with 2 GB RAM. The MacBook has a 64 MB Intel graphics processing system. The MacBook Pro has a 256 MB ATI Radeon Mobility X1600. And the G5 has a 256 MB ATI Radeon 9600. All systems were running Final Cut Studio 5.1, which is the first Universal Binary version of Apple's video production, motion graphics ad DVD authoring suite.
Final Cut Pro 5.1
Contrary to the bizarre rumor, the MacBook can indeed run Final Cut Pro, and it can run it quite well, if these tests are any indication. For Final Cut Pro, we ran four individual tests to compare performance on a variety of functions and in both HD and SD. In most cases, the results were fairly close. The 13-inch MacBook beat the dual G5 in two of four tests, tied it in one and lost out by one second in another. All of the tests were rendering tests, which, of course, doesn't tell the whole story. But it does give a good indication of the power of a computer's CPU.
Here are the results.
The first test involved rendering two DVCPro HD 720p60 clips in a sequence. Each clip had a single effect applied to it. The first one used Color Corrector 3-Way. The second used the Curl effect. Between the two, a Swing transition was applied. This was the only test in which the MacBook actually beat the MacBook Pro (by a second), tying the G5. I have no explanation for why the MacBook would beat the MacBook Pro here.
The second test also involved two DVCPro 720p60 clips. The first had Pond Ripple and Lens Flare effects applied to it. The second had Gaussian Blur, Fisheye and Color Balance filters applied to it. Between the clips was a Ripple Dissolve transition. The MacBook came in right in between the MacBook Pro and the G5.
The third test involved three layers of 720p HD footage with seven total clips and four transitions. The transitions included Page Peel, Center Split Side, Channel Map and Ripple Dissolve. Filters involved in this test included Color Key, Luma Key, Pond Ripple, Lens Flare, Gaussian Blur, Fisheye and Color Balance. Here the MacBook came in one second behind both the G5 and the MB Pro.
And the last project tested Final Cut Pro's rendering of SD footage rendring filters like Wind Blur, Color Offset and Unsharp Mask. Once again, the MacBook came out in the middle, beating the G5 by three seconds and losing out the the MB Pro by six seconds.
It should be noted that all tests involved rendering to a QuickTime file, which, of course, involves the performance of the hard drive into the equation. When rendering in the timeline, performance is much more impressive. (Test 1, for example, rendered in four seconds in the timeline, as opposed to 12 seconds when exporting a QuickTime file. We'll attempt to update these numbers to reflect a faster hard drive in the near future.)
Now Motion was a different story altogether. While it beat the G5 in three out of four tests, it didn't trounce the G5 in the way the 2.16 GHz MacBook Pro did. In none of the tests did the MacBook beat the MacBook Pro, as was to be expected.
Test 1 involved particles and nothing but particles in an NTSC DV-resolution project. Included in the project were the following systems using default values from Motion's library of presets: Cloud Transport, Clockwork, Heavy Sparks, Spiral and Star Tunnel.
The second test involved even more particle systems, six in total, and this time in DVCPro 720p. They included Magic Dust, Meta Wash, Smoke Cloud, Shell, Rocket and Spiral.
For the third test, we focused on effects rendered over multiple video layers in a D1-format project. Effects were applied to individual clips and to layers as a whole and included Radial Blur, Bevel, Refraction, Echo and Fun House.
And the final project involved multiple layers of video, text, shapes and replicators with behaviors and effects. Behaviors included Randomize, Grunge 4, Spring, Wind and Sequence Replicator. Effects included Echo, Fun House, Black Hole, Refraction and Radial Blur. The project was at D1 resolution.
Finally, we come to Compressor, with the weirdest results of all. In these tests, the MacBook beat the G5 in four out of five encoding tasks. The one test it lost to the G5 it lost by only a second. It tied the MacBook Pro in two tests. And it beat the pants off the MacBook Pro in two others, which can only be explained by the system component updates I mentioned earlier. (Unfortunately, we're unable to retest the MacBook Pro with the latest software updates, as that review unit is no longer in DMN's possession.) It lost to the MacBook Pro in only one test, and that by just a second.
Here are the results of the tests.
Test 1 encoded SD footage to MPEG-2 using a two-pass VBR with an average bitrate of 6.2 Mbps. The systems were all fairly even.
Test 2 took a 720p source file and encoded it as HD MPEG-2 (23.98 fps) at 19.0 Mbps (maximum 27.0 Mbps) with a two-pass VBR using "Best" motion estimation. Again the MacBook and MacBook Pro came out just a little bit ahead of the G5.
For the third test, though, we see a dramatic difference in encoding time. This was an H.264 HD encode involving 720p footage with a frame rate of 59.94 and an average bitrate of 10.25 Mbps (maximum of 23.0 Mbps). Here the MacBook Pro lost out badly to the G5, while the consumer-level MacBook beat them both out by a considerable margin.
For the fourth test, we returned to an MPEG-2 SD encode, this time using a one-pass VBR at an encode rate of 6.2 Mbps. The results were once again close.
The fifth test, though, came out again in favor of the lowly MacBook, beating the G5 by two seconds and the MacBook Pro by 11 seconds (inexplicably). This time we took some standard-definition footage and encoded it in H.264 for iPod video at 320 x 240 with a frame rate of 29.97 and a bitrate of 600 Kbps.
When I first got ahold of a MacBook Pro last month, I was shocked by its capability to beat out desktop G5 systems consistently. But I was doubly shocked to see such dramatic results from the newer and lower-end 2.0 GHz MacBook as well. After all, this is not just a notebook competing with a fairly current desktop system, but a consumer-level notebook at that.
Normally you don't even think about running benchmarks of professional-level creative software on a laptop against desktops, and certainly not a laptop targeted toward consumers. But in the vast majority of tests, the MacBook beat out the desktop system and certainly proved itself competition for all but the highest-end (G5 Quad) Mac systems on the market today.
Again, like the MacBook Pro, the 2.0 GHz 13-inch MacBook proves a thoroughly viable machine for users of Final Cut Studio.
We will continue this benchmark series next time around with a look at Adobe applications running in Rosetta under Mac OS X on the MacBook. Then we'll follow it up with benchmarks of 3D software, plus various tests involving Windows software on the MacBook.
In the meantime, if you have any questions, be sure to drop me a line or visit me in one of the forums listed below.
Related Keywords:macbook, benchmarks, final cut studio, 13-inch macbook, final cut pro, motion, compressor
Source:Digital Media Online. All Rights Reserved