Phone performance

[Daniel van Vugt]

It's not just frequency either. On arale (Meizu) for example, smoothness 
correlates directly with whether multiple CPU cores are online or not:

   cat /sys/devices/system/cpu/online

Usually the kernel only keeps one core online, which makes Unity8 
stutter. But if you touch it enough then the second core (out of eight) 
comes online and everything is smooth. I wonder if more aggressive use 
of threads might help...


On 14/08/15 16:48, Daniel van Vugt wrote:
> Hi all,
>
> In testing performance optimisations on various phones, I keep running
> into an annoying hurdle.
>
> Although you can optimise your Mir server/clients in such a way that
> they're smoother more often, there's an additional variable outside of
> Mir and Unity that gets in the way. That seems to be frequency scaling
> done by the kernel. Sometimes on desktops too, but I'm mostly concerned
> about phones here.
>
> I find it suspicious that on some devices you can turn stuttering into
> smoothness just but touching the screen a lot. But the smoothness soon
> goes away when you're not touching the screen. In the extreme case, if
> you're logged into the phone remotely you will also notice the system
> can become unusably slow when the screen has turned off. That's useful
> for a real phone's battery life, but it serves to illustrate that the
> kernel is doing a lot behind the scenes. I'm more concerned about how
> can we keep phone graphics performing as well as they do during touches,
> even when we're not touching them?
>
> - Daniel
>