Intel SNA With Unity, Unity 2D & GNOME Shell

[phoronix]

Phoronix: Intel SNA With Unity, Unity 2D & GNOME Shell

After the benchmarks a few days back of Intel Sandy Bridge Acceleration On Non-SNB Hardware, Chris Wilson of Intel who has been responsible for much of the "Sandy Bridge New Acceleration" work requested more tests, but this time to see the effect that the compositing window manager has on this new acceleration architecture. As a result, here is some quick tests of Intel's Sandy Bridge graphics under the Unity, Unity 2D, and GNOME Shell desktops.

http://www.phoronix.com/vr.php?view=16407

[przemoli]

I wish that there where someone so eager to listen to power regression tests

Its good to see how test results where handled on both sides! Good work.

PS Can you (as Phoronix) make an offer of preparing special LiveCD linux with linux drivers for Ivy Bridge release?
This is another space where cooperation might be very productive!

[kraftman]

According to some previous Phoronix benchmarks Unity 2D is faster than Unity, because of this:

When a system call is called by the Linux binary, the trap code dereferences the system call function pointer off the proc structure, and gets the Linux, not the FreeBSD, system call entry points...

*

Kwin suspends compositions and they're disabled when you're running Unity 2D, so they're on pair with each other. If PC BSD with kwin was faster than Ubuntu with Unity (3D) and if that was a proof of 3D being faster on PC BSD than on [entire] Linux, then Unity 2D must be faster than Unity (3D) because of the quoted sentence.* It's damn disappointing Phoronix does idiotic yellow journalism. It's a shame you keep basing on tests which were proven wrong, so you're simply misinforming your readers.

[gilboa]

You know... this paints Phoronix at very vivid and unflattering colors.

On one hand, Poronix publishes a sensational story claiming that Linux is slower than BSD at running Linux games, using questionable benchmarking methodology (Unity on Linux, kwin [with or w/o effects] on BSD, etc), completely ignoring calls to redo the benchmarks with a *sane* benchmarking methodology.
On the other hand, within days Phoronix publishes a second story comparing Unity, Unity2D and GNOME shell (I could only wonder why kwin is not benchmarked) that more or less proves that Unity has a *very* negative effect on gaming performance completely ignoring the previous story. Somehow I doubt that it was a huge surprise to the poeple @Phoronix.

The irony is by doing this Phoronix is shooting itself in the foot; Each time a sensational story is being published for the sake of publishing a sensational story, adequate benchmarking standards be damned, one more kernel/distribution developer adds Phoronix to his black list. From there, the road to being completely ignored by the community tends to be fairly short.

- Gilboa

[drag]

You guys are just repeating what people know. There isn't any point to it and it's just not on topic. There is no need to degenerate this any further.