Benchmarks: Page (1) of 1 - 07/31/07

email article

print page

Mac OS or Windows? The Great After Effects CS3 Smackdown Attempting to determine the faster OS on identical hardware By Kevin Schmitt

There are many, many reasons to love Intel Macs, but the one I'm going to focus on today is Boot Camp. Being able to run a full copy of Windows natively on the exact same hardware as Mac OS X not only represents a 2-for-1 bang for your buck when it comes to buying a production rig, but also gives average Joes such as yours truly the opportunity to compare apples to apples (so to speak) when running cross-platform software. And with After Effects CS3 finally out in the wild (and in Intel Mac-native form, no less), it's time to see how well the Mac and Windows versions of AE do on identical hardware.

Methodology, such as it is

I'll readily admit that I'm new to the benchmarking game, as I really never had reason until now to play in that particular sandbox. But curiosity is driving me to finally jump in, so away we go. Fortunately, DMN has a standard After Effects benchmarking suite that we've used for a while now, so that takes some of the guesswork out of it. Now, my first inkling was to put AECS3 through its paces on my Mac Pro; alas, the copy of Windows Vista installed on my Boot Camp partition only addresses the first 2 GB of the full 3 GB complement on that machine (which seems to be a general problem for both XP and Vista running in Boot Camp, if the copious forum postings Google brought to my attention are any indication), so I knew I wasn't going to get a fair fight there. However, I did want to get the Mac Pro into the mix regardless, just for giggles, as we'll see shortly.

I ended up using a first generation 2.0 gHz Core Duo MacBook for the full triple-OS test. I know, I know?not exactly primo hardware, but with "only" 2 GB of RAM on the machine, all OSes had the same resources to play with. Each test was performed with fresh installs of Mac OS X 10.4.10 (Tiger), Windows Vista Business, and Windows XP Pro SP2. All available patches were applied to each OS, and in all cases, the tests were run with After Effects CS3 as the only active program (screensavers disabled, no third-party background processes, etc.). Each test was run twice, with the scores averaged together (which is reflected in each result table). Lastly, regardless of platform, I had each test render out to QuickTime 7's Animation codec at Best quality. Hopefully, that's about about as equal as you can get.

The Tests

We've run all these tests before in various benchmarks articles, and while they were optimized for After Effects 5.5 and 6, they're still good to go on AECS3. Here's a quick recap of what's what:

1. The first test (Simple Animation) involves a basic sequence consisting of a PICT file with tracing paths.
2. Test two (Video Composite) is a relatively standard compositing setup, comprised of green screen layer, a CGI layer, and a background plate originally shot on film.
3. The third test (Data Project) mixes bitmap and vector layers, animating randomly sequenced zeroes and ones.
4. Test four (Gambler) composites vector-based Illustrator files in 2D.
5. The fifth test (Source Shapes) also uses vector images, though moved around in After Effects pseudo-3D environment and "shot" with a 3D camera.
6. Test six (Virtual Set) uses bitmap images constructed into a 3D environment, also making use with a 3D camera.
7. The final test uses Brian Maffit's Total Benchmark for AE6, which is described and available from this link.

If you would like to replicate the first six tests for yourself, pick up the book After Effects 5.5 Magic that includes a CD containing these AE project files (and many more) along with all the media you'll need to reproduce our results. Special thanks to After Effects 5.5 Magic's author Mark Christiansen and the book's editor, Nathan Moody, as well as New Riders Publishing for giving us permission to use materials from this excellent book.

The Results

Enough setup: let's get to the tables with the stuff in them, shall we? This first table is the MacBook results; all times are in seconds, with the best average time highlighted in red:

MacBook - Single Process

	XP 1	XP 2	XP Avg.	Vista 1	Vista 2	Vista Avg.	Tiger 1	Tiger 2	Tiger Avg.
1. Simple Animation	3	3	3	3	3	3	4	4	4
2. Video Composite	26	26	26	26	27	26.5	31	32	31.5
3. Data Project	55	54	54.5	56	59	57.5	70	67	68.5
4. Gambler	17	16	16.5	17	18	17.5	22	22	22
5. Source Shapes	89	89	89	90	90	90	104	104	104
6. Virtual Set	77	78	77.5	88	88	88	95	96	95.5
7a. Total Benchmark Part 1	60	60	60	69	69	69	72	72	72
7b. Total Benchmark Part 2	866	844	855	875	874	874.5	951	937	944

This second set of tests was run on a Mac Pro (Quad 2.0 gHz Intel Xeon with 3 GB of RAM). Despite the fact that Mac OS X was able to address the full 3 GB of RAM, Vista still took all but one of the tests (and tied in another) even though it "saw" only the first 2 GB. So take these results with a grain of salt if you like, but it's more of the same. The benchmarks in this case were run on Mac OS X 10.4 (Tiger) and Windows Vista Business (no XP this time), and were run both as a single process and with After Effects CS3's new multiprocessing preference enabled (hence the two tables). Again, all times are in seconds, and best average times are in red:

Mac Pro - Single Process

	Vista 1	Vista 2	Vista Avg.	Tiger 1	Tiger 2	Tiger Avg.
1. Simple Animation	2	2	2	3	3	3
2. Video Composite	16	16	16	19	19	19
3. Data Project	36	37	36.5	55	46	50.5
4. Gambler	14	14	14	17	17	17
5. Source Shapes	49	49	49	67	63	65
6. Virtual Set	54	53	53.5	64	63	63.5
7a. Total Benchmark Part 1	50	50	50	56	54	55
7b. Total Benchmark Part 2	518	524	521	608	603	605.5

Mac Pro - Multiprocessing Enabled

	Vista 1	Vista 2	Vista Avg.	Tiger 1	Tiger 2	Tiger Avg.
1. Simple Animation	2	2	2	2	2	2
2. Video Composite	15	16	15.5	18	18	18
3. Data Project	32	32	32	43	43	43
4. Gambler	9	9	9	11	11	11
5. Source Shapes	38	38	38	48	49	48.5
6. Virtual Set	55	55	55	50	50	50
7a. Total Benchmark Part 1	27	27	27	33	33	33
7b. Total Benchmark Part 2	543	559	551	577	578	577.5

The Conclusion

Well, that was bracing, wasn't it? While I'm not at all surprised that the Mac OS X version of After Effects CS3 didn't keep up with either version of Windows, I am slightly disappointed that the results weren't closer, what with Adobe's re-embracing of the Mac platform in the CS3 Production Premium bundle and some shiny new Intel chips to play with.

If you want the fastest possible speed, it appears that Windows XP is still the way to go, though Vista didn't do too shabbily either. In all cases, if you have the system resources available, enabling multiprocessing will generally net you faster times (though part 2 of the Total Benchmark on Vista was slower with MP enabled, strangely enough).

In any event, I look forward to adding Mac OS X 10.5 (Leopard) numbers into the mix once Apple deems it fit for release. In the meantime, do with these results what you will, and regardless of platform, enjoy playing with all the new and cool stuff in After Effects CS3.

Page: 1

Though the fame, riches, and notoriety of being a DMN contributor are both tantalizing and substantial, Kevin Schmitt still stubbornly insists on continuing his work as the Director of Interactive Services at EFX Media, a production house located just outside of Washington, D.C. Feel free to follow his updates and contact him through Twitter if you have something to share - he's ready to believe you!

---

Related Sites: Creative Mac ,   Digital Producer ,   Media Workstation ,   IBC News ,   Digital Animators ,   Animation Artist ,   Siggraph News ,   Audio Video Producer ,   BN - Broadcast Newsroom ,   Digital Post Production ,   Oceania ,   MacAnimationPro ,   MacVideoPro ,   MacDesignPro ,   CEN - Desktops ,   BN - Hardware
Related Newsletter: DMN Newsletter ,   Timeline Newsletter ,   Pixels Newsletter ,   Mac Alert Newsletter ,   Renders ,   Digital Media Net ,   DMNForums

---