0

I use a third party extension to Windows Explorer (in Windows 8.1) called QTTabBar. In this screenshot you can see how by default there is a huge number of custom keyboard shortcuts activated by it, and I assume these are 'loaded' into Windows Explorer, the registry, or the system memory in some sort of way as a result.

alt text

There's already dozens of other hotkeys for the system to watch out for such as my many AutoHotkey scripts, on top of Windows' default shortcuts themselves.

Would turning off these excessive shortcuts (which I don't use), actually speed things up at all? Or, to put it another way - does having them 'on' slow down Windows at all? Is it like browser extensions slowing down browsers in any way?

Or would there not make a difference between 1 or 200 shortcuts ticked in QTTabBar, and to increase performance I would have to turn off the app altogether?

If it's relevant to the question, it would be helpful to know how (third-party) keyboard shortcuts actually use up system resources in Windows - whether it's a constant on-the-fly memory allocation depending on what programs are open or what window is focused on and what hotkeys it's telling Windows to honour, or whether it's more of a static interaction with Windows such as registry entries recorded by the program in question.

Improve this question
1
  • I can tell you the skinny about in-program shortcuts, but once you throw in programs such as AutoHotkey things get a bit unpredictable.
    – Ignacio Vazquez-Abrams
    Commented Nov 12, 2013 at 7:44

1 Answer 1

Reset to default
0

and I assume these are 'loaded' into Windows Explore

More likely the tool loads a (session scoped) hook on to the applicable keyboard events.

Would turning off these excessive shortcuts (which I don't use), actually speed things up at all?

Clearly having fewer customisations to check on each keyboard event would require less processing.

But the amount of processing all of those, on even a few years old CPU is tiny. Say it takes 50 cycles to check each one, so that's 5000 for 100 shortcuts.1 But CPUs can perform of the order of a billion cycles per second. Ie. that 100 shortcuts is using 0.0005% of the CPU.

Of course on a sufficient number of shortcuts added on a sufficiently slow system with a sufficiently large number of processes neeeding CPU it might be an issue, but even then it is likely that cutting down on other tasks will have a bigger impact.

Summary: most CPUs spend most of their time idling waiting for user input. A small (relatively) increasing of processing on the input queue is most unlikely to have any user measurable difference.2

1 I'm ignoring all the fixed—independent of the number of shortcuts—costs here. Eg. there will be a cost of Windows calling the hook, this could easily be larger than the checking time, but it is still tiny...

2 Of course there are tools (eg. Windows Performance Toolkit) that can look at this level of detail, but I'm talking about user perception.

Improve this answer

You must log in to answer this question.