[Wayland-bugs] [Bug 90172] Double tap not always recognized

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=90172

--- Comment #15 from Velimir Lisec <lisec.velimir at gmail.com> ---
(In reply to Peter Hutterer from comment #14)
> I don't think we can actually reduce the first tap timeout by much. If you
> hit a ~200ms delay on a "normal tap" (comment #9) then reducing it further
> risks missing it altogether. 

Yes, but if I'm not mistaken in the current implementation timer is reset after
the first release event arrives. That means that delay between first touch down
event and first release event is DEFAULT_TAP_TIMEOUT_PERIOD, and delay between
first release event and second touch event is also DEFAULT_TAP_TIMEOUT_PERIOD.
200 ms from "normal tap" is delay between first and second touch down event. So
with current implementation, if DEFAULT_TAP_TIMEOUT_PERIOD is set at 180, delay
between first and second touch down events can theoretically go up to 360 ms
(180 + 180) and the tap would still be recognized.

> Here's a patch that reduces the second tap timeout, I think that's better to
> rely on. That should then also reduce the total time so we're more likely to
> be interpreted as a click in the default settings of the client.

I get the best results with the first patch applied and
DEFAULT_TAP_TIMEOUT_PERIOD set to 150 ms so I think you should use that one.

Also, DEFAULT_TAP_TIMEOUT_PERIOD is used as a timeout between finger lifted
during drag and finger lowered to continue drag. That time should be increased,
and I think 180 ms is to low, 300 (or 350?) should be good value for that
timeout (that's a guess, i haven't tried it. Under synaptics driver I use a
timeout of 700 ms, but that is to long as a default. Ideally, that value should
be configurable).

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/wayland-bugs/attachments/20150429/5415a828/attachment.html>