Apple OS X 10.10 vs. Ubuntu 14.10 Performance Test | Hacker News Reader

Apple OS X 10.10 vs. Ubuntu 14.10 Performance Test

phoronix.com

16 pointsniutech8 comments

GPU drivers have always been a nightmare on OS X, but I didn't know just how bad they really are. Keep in mind that we're comparing against Mesa drivers here. Try to imagine what the benchmarks will look like when Valve's Steam Machines have shipped.

The same is true for HFS+. No real surprises here, even though I didn't know it was that slow. Most people are probably more interested in ZFS vs. Btrfs benchmarks.

I would have loved to see some kernel oriented benchmarks like network stack performance, etc., but then again, performance benchmarks don't really matter anyway. I'd like OS vendors to focus a bit more on perceived performance. Just the other day I plugged an external USB HDD into my MacBook Pro and a YouTube video stopped playing in Chrome (because of some blocking IO call, I presume). The same happens quite frequently with network IO and freezing UIs.

I have to admit I had not given the performance differences between Linux and OSX much thought, but I am quite surprised. One might think that Apple, building both the hardware and the operating system should be able to make the two work together rather well.

(There are of course issues the benchmarks do not measure, specifically battery life, it would be interesting to see some numbers on that.)

I switched from a 2009 Asus EeeBox running Ubuntu 12.04 to a MacMini with an i7 about a year ago, and I had planned to install Debian or Ubuntu on it, but so far I have postponed that. Maybe I should reconsider. There are many things I have come to like about OSX, but some things I do miss about Linux, especially the poor integration of XQuartz into the desktop is disappointing. Cygwin's X Server on Windows integrates into the system a lot more smoothly than XQuartz does on OSX.

I'm a bit surprised at the "computational benchmarks" [1] though. Can anyone confirm these are indeed CPU-bound, and not somehow GPU-assisted? If so, how can the underlying kernel possibly make such a difference? Scheduling? (if so, what kind of load was the machine running?) Clearly there must be something more to it.

[1] http://www.phoronix.com/scan.php?page=article&item=osx10_ubu...

PolyBench/C is open source. Straight forward matrix multiplications in C. The code looks exactly like you'd expect it to look.

See the description for implementation details: http://web.cse.ohio-state.edu/~pouchet/software/polybench/#d...

Perhaps power management is somewhat to blame here? I would be curious to see how power consumption compared while running the benchmarks. It's well known that OSX gets much better battery life than Linux on Apple hardware. I wonder if the CPU was being throttled to conserve power or thermal output.

> It's well known that OS X gets much better battery life than Linux on Apple hardware.

I couldn't find any information to back that up. Do you have a link to a benchmark?

I don't think kernels throttle CPUs — CPUs do that themselves already. I could imagine Apple doing something like OpenBSD in that they favour security over performance for some operations, but that seems unlikely. IOKit might be to blame as well, given that it's written in embedded C++ instead of C.

All things considered, I think it's highly likely that, while OS X is quite shiny on the outside, the underlying OS isn't as modern as some of the others. Windows and OpenBSD have better security features and now Linux apparently has better performance.

Well, but I guess theoretically it could be doing stuff like setting lower thermal thresholds for throttling. So perhaps not exactly a function of scheduling, but getting the CPU to make itself slower.

Still, this to me is the most interesting result from that set, and I'd really love it for someone to dig into it.

I'm curious to what would be the Windows 8.1 scores to in this test.

Reply on news.ycombinator.com