The Cost Of Running Compiz

[nanonyme]

Or rather suspend at all if games are run full-screen.

[Zhick]

Yes it does because according to benchmark compositing is faster in DRI1 than DRI2.

I'm pretty sure that's what brent meant. It's obvious that there's not going to be a big performance hit on DRI1 drivers when running compiz, because the app will just ignore the compositor and draw right on the screen. With DRI2 the app gets composited correctly, which is why it's slower.

[reavertm]

It happend so many times before but this time I really had to register and write this post. Phoronix should really pay more attention to numbers and their math. For example, in the "Unigine Heaven v2.0 1980x1080" we see that the Metacity result is 74% faster than Compiz, not 63%. I haven't found a rounding of the numbers that would actually lead to 63%. If Metacity leads to 25.34 fps and Compiz to 14.56, then Metacity is (25.34-14.56)/14.56=0.74... times faster than Compiz. On the contrary, Compiz reaches just (25.34-14.56)/25.34=0.425=42.5% of the performance of Metacity. Now question to the reader: By what percentage is Compiz slower than Metacity?

You're quite wrong here, and 42.5% is the number you're asking for.
It's really simple.
Given numbers 25.34 fps for Metacity and 14.56 fps for Compiz.
It's a matter what you're comparing to and you assign 100% there and then just solve simple proportion.
So how is Compiz performance related to Metacity?

Metacity 25.34 - 100%
Compiz 14.56 - x

x = ( 14.56 * 100% ) / 25.34 ~= 57.5%

What does it mean?
That Compiz performance is 57.5% of Metacity performance so Compiz is slower than Metacity by 100% - 57.5% = 42.5%

I - like you claim - Compiz reached 42.5% of Metacity performance it would mean that Compiz would need to have 25.34 * 42.5% ~= 11 fps

Now how does Metacity performance relate to Compiz performance?

Compiz 14.56 - 100%
Metacity 25.34 - x

x = ( 25.34 * 100% ) / 14.56 ~= 174%

What does it mean?
Again, Metacity performance is 174% of Compiz performance which means Metacity is faster by 74% than Compiz.

Overall, Phoronix should definitly come in contact with someone with a more academic background.

Like yourself?

cheers

[Svartalf]

What good does it do to push 400 FPS when your monitor only refreshes itself at 60hz?

The bogus levels serve to show something resembling peak performance out of the GPU- not that you'd be able to SEE it.

However, it's a decent enough metric to know you won't run out of headroom for framerate with a complicated scene if you're pushing 400FPS, whereas unless you're averaging 30-60, you might have some areas where the FPS fall below the usable level- it still averages out to the 30-60, but...

[deanjo]

Enabling composite windows in KDE4.x apparently also gives a drop. From http://forums.gentoo.org/viewtopic-p...931bd2#6221554

You notice those are all while running in windowed mode. In full screen mode the results are identical.

[Svartalf]

I find it offensive that more people don't understand this.

I find it offensive that you made a remark like this. While your explanation is fairly good for why high FPS' might be useful- it's not WHOLLY the case as not all titles will render well with VSYNC turned off and not all input loops are tied to the rendering loop.

[MathGuy]

Like yourself?

cheers

Touche. Yes, like myself

Really embarrasing. I preparred the text in an external text editor and screwed up copying it into the form. Of course, I didn't want to post the solution to the question. Embarrasing. But the rest of my opinion still holds.

Cheers.

[etnlWings]

Yeah. But that wasn't the issue.

No shit. It's called a general comment.

[V!NCENT]

I haven't found a rounding of the numbers that would actually lead to 63%. If Metacity leads to 25.34 fps and Compiz to 14.56, then Metacity is (25.34-14.56)/14.56=0.74... times faster than Compiz. On the contrary, Compiz reaches just (25.34-14.56)/25.34=0.425=42.5% of the performance of Metacity. Now question to the reader: By what percentage is Compiz slower than Metacity?

Sup dawg. Awsum meth dur, yo.
25.34 / 14.56 = 1.74 = 74% faster.
They teach this shit at first school... with your 'academic' credzzz... Maybe Michael just made a type arror on his calculator. Seriously...
So how many times slower is compiz?
14.56 / 25.34 = 0.57% the speed of Compiz, so 100 - 57 = 43% slower... Wow that was hard as fsck!!!1111

The test scenarios for the article should be designed more according to the "scientific approach" to do evaluation. Otherwise, Phoronix just targets readers who are just able to understand black/white messages. "Scientific depth" is the keyphrase here.

I never before said this in my entire life, but dude... you need to get laid...

[V!NCENT]

Correction time...

So how many times slower is compiz?

So by what percentage is Compiz slower?

14.56 / 25.34 = 0.57% the speed of Compiz, so 100 - 57 = 43% slower...

14.56 / 25.34 = 0.57 times the speed of Metacity, so 100 - 57 = 43% slower...