Hackintosh Performance

The extensive array of benchmarking software we take for granted in the Windows world doesn't really exist in the Mac world, so I'll be using the same three programs I used in my original Hackintosh article. This time I'll be comparing four systems:

• My original Core i7 920 Hackintosh, with the processor base clock at 172MHz and the multiplier at 21 for a speed of 3.6gHz
• A 15" MacBook Pro laptop with Intel's Core i7 2820QM processor. This is the fastest laptop configuration Apple offers, with a base CPU speed of 2.3gHz and a max turbo speed of 3.4gHz.
• The Sandy Bridge Hackintosh with a Core i7 2600K processor at the stock settings, 3.4gHz with a max turbo speed of 3.8gHz.
• The Sandy Bridge Hackintosh with the maximum turbo boost multiplier set to 49, resulting in a max turbo speed of 4.9gHz.

Geekbench Results

Primate Labs' GeekBench is a "one-click" benchmark utility that performs a number of processor and memory tests; it does not test video cards or disk I/O. There are 12 integer CPU tests, 14 floating-point CPU tests, 5 basic memory tests, and 8 memory bandwidth tests. All tests comprise a mixture of single-threaded and multi-threaded versions. It produces a weighted composite score based on the individual scores.

Interestingly, the original 920-based system produces better scores in the integer and floating point benchmarks than does the stock Sandy Bridge system. The overclocked Sandy Bridge system returns 28% better integer and 26% better floating point scores than the 920. Althought the Sandy Bridge memory system is dual- rather than triple-channel, it's running its memory at 1600mHz while the 920 was running at just over 1400mHz. The overclocked configuration has 39% better memory performance and 54% better bandwidth than the 920 system. These memory differences are much higher than I'd expect and I have no real explanation for them. Latencies were 9-9-9-24 in both cases.

It's interesting to see how close the scores of the MacBook Pro are to the overclocked 920. The 2820QM mobile processor is virtually even with the stock-clocked 2600K in these tests.

CINEBENCH 11.5

Maxon Cinebench is a real-world test suite that assesses the computer's performance capabilities. Cinebench is based on Maxon's award-winning animation software, Cinema 4D, which is used extensively by studios and production houses worldwide for 3D content creation. Maxon software has been used in blockbuster movies such as Spider-Man, Star Wars, The Chronicles of Narnia and many more. Cinebench Release 11.5 includes the ability to more accurately test the industry's latest hardware, including systems with up to 64 processor threads and the testing environment better reflects the expectations of today's production demands. A more streamlined interface makes testing systems and reading results incredibly straightforward.

The Cinebench R11.5 test scenario uses all of a system's processing power to render a photorealistic 3D scene, "No Keyframes", the viral animation by AixSponza. This scene makes use of various algorithms to stress all available processor cores.

The stock-clocked 2600K is just over 7% faster than the overclocked 920 in the single-core rendering test, but 53% faster when overclocked. In multi-core rendering, there's only a 5% advantage to the 2600K running stock, but the difference widens to 36% when overclocked. Again, the MacBook Pro is turning in scores that would have been unthinkable for a laptop computer even a year ago. Intel's mobile Sandy Bridge parts are quite impressive.

Handbrake 0.95

Few consumer applications will make good use of a six-core processor, or even a four-core processor. Extra cores can give you a system that remains responsive when performing a computationally-intensive background task, but will rarely accelerate the execution of an individual program. There are several reasons for this:

• Most users' systems still have two cores. According to Steam's July 2011 statistics, dual-core systems comprise 49.28% of its users, quad-core systems 41.73%, and six-core systems a miniscule 1.29%. This shows a huge jump in the number of quad-core systems out there, as the percentage has almost doubled from the 27.49% in August 2010.
• Relatively few computational problems scale well with the number of threads available. (One common task that does is rendering, which is why modern video cards have dozens or hundreds of cores).
• Writing good multithreaded code is difficult and time-consuming. Developers generally realize a better return on their effort for other code optimizations.

All that said, media transcoding (converting to a different format) is something that does scale well with the number of available cores, and the free and open-source Handbrake 0.95 video transcoder is an example of a program that makes full use of the computational resources available. For this test I used Handbrake 0.95 to transcode a standard-definition episode of Family Guy to the "iPhone & iPod Touch" presets. The encoding times are in seconds; lower is better.

The stock Sandy Bridge system is only about 4% faster than the overclocked 920, but the overclocked 2600K is 24% faster. The MacBook Pro trails here, but still turns in a credible performance for a laptop.

I'll discuss my thoughts and present my conclusion in the next section.

<< Previous Page - Next Page >>