Re: Scrolling performance

[Roo <oscosc gmail com>]

On Thu, 06 Jul 2006 14:11:18 -0500, Michael Ekstrand wrote:

> On Jul 6, 2006, at 1:28 PM, Roo wrote:
>> On Thu, 06 Jul 2006 08:21:56 -0500, Michael Ekstrand wrote:
>>> Part of the problem/reason: for some reason, it seems to work (or at
>>> least give acceptable results) for many of us.
>>
>> No doubt, when you throw enough hardware at problems it does tend to
>> hide
>> them.
> 
> Did you read the list of hardware I mentioned not having problems on?

Yes... I'm guessing they have better graphics cards. BTW: there have been
many, many complaints about GTK's performance on a wide variety of cards.

>>> So basically: there are more variables than just GTK.
>>
>> Ok, so presumably there's a way to take gtk2-2.8.19-2.src.rpm (the
>> currently installed GTK2 on my FC5 system) and recompile it (or is
>> there an env var?) telling it not to use Cairo so I can see if there's
>> a difference and hence narrow things down.
> 
> Looking at configure.in in the GTK2.8 sources, it appears that there is
> no way to disable cairo.

Right... so the GTK developers left normal users without the ability to
compare it directly. That makes a lot of sense if they wanted to find out
the real facts and figures behind Cairo...

>> I could also use a recommendation
>> for a benchmark app to generate some hard numbers under the different
>> setups.
> 
> Check out gtkperf.  I haven't used it, but it claims to to benchmark
> GTK.

I know about that one. I just figure the GTK developers had something of
their own. I mean, I would expect them to have their own performance
testing tools... or has that been the major problem all along? They don't
have their own tools and just guess.

Even with gtk-perf I'm still stuck being unable to make a comparison,
since they took away that option.

>> I'm seeing these issues, and I'm fed up of them not being fixed. I'd
>> like
>> to know what tools the GTK developers use for measuring performance
>> changes, and how I can cut out Cairo in order to get a grip on exactly
>> what is going on.
> 
> Are you using the GTK default theme?

I've tried all the different themes. I've tried tweaking everything at one
time or another (and restoring the original settings). I have about 50
different varieties of xorg.conf with different options

> I'd try cutting any non-default themes out first.  And it looks like
> Cairo can't be removed, for better or worse.

I don't think that's acceptable at all. Cairo pretty obviously has major
performance problems (even if you accept that it is only on some cards),
and the GTK developers forced everyone to use it without even a fallback.

>> I know you are trying to cover the bases here, but that just doesn't
>> cut it as an excuse. GTK2 has been slow from the beginning, and it's
>> getting slower. Other toolkits, like Qt (which I don't like I must
>> stress), manage
>> to be correct and avoid the major performance problems that have
>> plagued GTK for years now.
> 
> Depends on your perspective.

Not really. Qt is nice to code for too.

> Profile.  What happens if you build Cairo, GTK and GTKperf or gtk-demo
> or something with profiling enabled, and look at the profiling data?
> Where are GTK and Cairo spending their time?

But with what am I comparing it? I *know* it wasn't this bad with FC4
(pre-Cairo), but the GTK developers have removed the option of testing
that without wiping the box. I'm sure you can see the utter frustration
involved here. No wonder performance problems never get fixed. The GTK
developers seem to go to great lengths to ensure that reports are very
ignorable.



For what it's worth, here are the results from three Gtkperf runs with
different control themes:

Controls: Bluecurve
Borders: Atlanta

GtkPerf 0.40 - Starting testing: Thu Jul  6 21:31:41 2006

GtkEntry - time:  0.13
GtkComboBox - time:  1.51
GtkComboBoxEntry - time:  1.26
GtkSpinButton - time:  0.16
GtkProgressBar - time:  0.14
GtkToggleButton - time:  0.38
GtkCheckButton - time:  0.36
GtkRadioButton - time:  0.43
GtkTextView - Add text - time:  1.35
GtkTextView - Scroll - time:  0.58
GtkDrawingArea - Lines - time:  0.52
GtkDrawingArea - Circles - time:  1.47 GtkDrawingArea - Text - time:  3.56
GtkDrawingArea - Pixbufs - time:  0.27
 ---
Total time: 12.12




Controls: Clearlooks

GtkPerf 0.40 - Starting testing: Thu Jul  6 21:33:37 2006

GtkEntry - time:  0.11
GtkComboBox - time:  4.44
GtkComboBoxEntry - time:  3.53
GtkSpinButton - time:  0.57
GtkProgressBar - time:  0.33
GtkToggleButton - time:  1.74
GtkCheckButton - time:  1.77
GtkRadioButton - time:  2.19
GtkTextView - Add text - time:  1.75
GtkTextView - Scroll - time:  0.93
GtkDrawingArea - Lines - time:  0.70
GtkDrawingArea - Circles - time:  1.61 GtkDrawingArea - Text - time:  3.66
GtkDrawingArea - Pixbufs - time:  0.18
 ---
Total time: 23.54




Controls:Thinice

GtkPerf 0.40 - Starting testing: Thu Jul  6 21:31:41 2006

GtkEntry - time:  0.13
GtkComboBox - time:  1.51
GtkComboBoxEntry - time:  1.26
GtkSpinButton - time:  0.16
GtkProgressBar - time:  0.14
GtkToggleButton - time:  0.38
GtkCheckButton - time:  0.36
GtkRadioButton - time:  0.43
GtkTextView - Add text - time:  1.35
GtkTextView - Scroll - time:  0.58
GtkDrawingArea - Lines - time:  0.52
GtkDrawingArea - Circles - time:  1.47 GtkDrawingArea - Text - time:  3.56
GtkDrawingArea - Pixbufs - time:  0.27
 ---
Total time: 12.12



BTW: I *don't* use clearlooks normally. I didn't need gtkperf to tell me
that its performance is pitiful.