Slashdot Mirror
Ubuntu 8.10 vs. Mac OS X 10.5.5 Benchmarks
An anonymous reader writes "As a sequel to their Is Ubuntu Getting Slower? Phoronix now has out an article that compares the performance of Ubuntu 8.10 to Apple's Mac OS X 10.5.5. They tested both the x86 and x86_64 spins of Ubuntu and threw at both operating systems a number of graphics, disk, computational, and Java benchmarks, among others. With the Mac Mini used in some of the comparisons, 'Leopard' was faster, while in others it was a tight battle."
Surely we should be united against the common enemy.
It's not fight, it's play. And when one system wins in terms of speed or usability, both systems win in terms of a weaker common enemy.
This thread will end up getting moderated flame-bait, but what would that common enemy be? Personally I think Windows is rather ok now, Windows 7 will probably be even better, who knows, maybe even better than snow leopard.
The only thing I see as an enemy is ideas which are pushed down my throat no matter what if I want them or not. I want to use my data and my applications in the way I feel like, not be forced to a single method just because someone else thought it was the best one. But that is true for all operating systems and no special "enemy."
I like many things in OS X and in applications for it because it makes sense and makes using the computer more comfortable, I don't like some other things because they don't let me do the things I want to do.
The huge amount of applications for Windows makes it rather likely that you can find one which fits your purpose, some for the window managers and such in the free unix-like oses.
... for those that can't be bothered to read this lengthy yet information sparse piece.
1. MacOS X is faster in graphics intensive benchmarks.
2. The other benchmarks are fairly even with Ubuntu coming out on top more often than OS X (one notable exception is SQLite).
This is hardly anything new. OS X has a well optimised graphics system with good drivers for the intel chips (which up until now was used in both Macbooks and Mac Minis).
Also SQLite is AFAIK integral to many features of OS X, and for this reason it makes sense for Apple to have optimised for it.
Overall the benchmarks suggests that Linux (not just Ubuntu) needs some work on the graphics system and the Intel drivers. What a shock.
For things like compilations, there's a bunch of file opens, caching the compiler and loader, gobs of Mallocs, and so forth that probably intersect the OS. Then there's the driver and video layer tests that look at frames per second. Leopard had 2 to 4 times faster frames per second. then there's the supporting distro services. Tests of My SQL were 4 times faster on the mac. And then there's things like the optimzation of VMs like JAVA where again Leopard excels. THese are clearly optimization problem and can be improved. the purpose of comparing it against a mac is not simply to say "oh yeah mac is faster than unbuntu", but rather to give a bench that shows how much room for optimization ubuntu has. Conclusion is that in almost every aspect Ubuntu is severely unotimized. Since older Linux seemed to be more optimized it suggests that feature bloat is probably either screwing up the design of linux or no one is paying attention to optimizing those features.
Ubunu isn't getting slower, Mac OSX is getting faster.
Do any of you recall Mac OSX 10.0?
The day I installed Apple's first "modern" OS, I thought X marked the spot of Apple's demise.
Apple has done an admirable job bringing MacOS into the 21st century, and their future looks promising.
Next year, we will be seeing how much the extreme emphasis Apple is placing on performance will affect comparisons like these. Apple has figured out that since they can no longer hope to use differences in the CPU to differentiate themselves with generic Windows boxen, they will be using Microsoft's extreme backwards compatibility needs against them when it comes to fully using all the cores--whether they be in a CPU or a GPU--in a computer, and making full use of the 64-bit instruction set. GPGPU programming can give a huge performance boost to certain algorithms and the cleaner, more register rich, 64-bit instruction set is intrinsically faster in addition to allowing larger data sets.
That's why they stopped selling non 64-bit capable computers a couple years ago, and why the new MacBooks have much improved integrated graphics. That's why they are moving their developers to include 64-bit compiles as part of newly shipped universal binaries. Next year is when all this latent potential gets switched on.
Linux has the opportunity to do the same; perhaps more opportunity as it has less of a legacy binary issue, although Linux has to deal with a multitude of graphics chips, Apple only has to optimize for a handful.
It came from nothing to something in a very short period of time.
I wouldn't call Debian "nothing".
Mac OS X doesn't have to accommodate variances in the hardware it is running on in the same way that Linux or Windows has to do. Therefore, it can exploit the hardware better. It's the same principle that applied to game developers targeting the XBox rather than a standard PC. Standard PCs might be more powerful, but the XBox is a non-moving target, so you don't need to write to the lowest common denominator, and can exploit the particular strengths of the hardware better. So, it's unreasonable to expect an OS that is written to work on multiple platforms to compete in this fashion.
-1 Uncomfortable Truth
Also worth mentioning are the collection of posts from the last thread that convincingly argued various problems with the Phoronix Benchmarks. Example 1 Example 2 Example 3
Speed tests are good, let's make sure we're doing them right
Every one of those examples are fail at reasoning weaknesses in the Phoronix Test Suite and this is why:
Example 1
If you look closely you'll notice that (a) the benchmarks were run on a Thinkpad T60 laptop, and (b) there were significant differences on some benchmarks like RAM bandwidth that should have little or no OS components.
If you look closely you'll notice that (a) the laptop the benchmarks are run on effects in no way, the validity of the benchmark as long as they are run consistently on the same laptop and (b) some benchmarks like RAM bandwidth have theoretical limits that are not effected at all by the Operating System but in actual practice, is entirely limited by the operating system you are using.
Example 2
Some of the benchmarks were hardware testing, and those showed variation. They should not, unless the compiler changed the algorithms used to compile the code between distros.
All of the benchmarks were testing the hardware and should have showed variation. The compilers used on all the benchmarking applications are all the same. But the compilers used to build the Operating Systems are all completely different versions. Therefore the compiler on each distro will compile the same "algorithm" slightly different way. That is assuming there were no changes between implementation of packages between distros (of which there were actually hundreds of thousands of changes in the code itself, build options, and runtime configurations)
Example 3
The test suite itself: The Phoronix test suite runs on PHP. That in itself is a problem-- the slowdowns measured could most likely be *because* of differences in the distributed PHP runtimes.
The Phoronix-Test-Suite Only uses its PHP back-end to aggregate benchmarking information. If a compilation with GCC took 5 seconds, its going to take 5 seconds no matter what version of the PHP runtime is used to to start the sub-shell that GCC runs in. It's take the same amount of time if you invoked GCC from bash, from perl, python, java, tcl, C, or C++. It doesn't matter because GCC is its own process just like every other benchmark.
What exactly are they testing? The whole distro?
Yes.
The kernel?
Yes again, since that is a part of the distro
If they're testing the released kernel, then they should run static binaries that *test* the above, comparing kernel differences.
No, what wouldn't prove anything as most of the binaries with ea