synaptics bug 31636: non-uniform horizontal and vertical acceleration

Chase Douglas chase.douglas at canonical.com
Tue Apr 3 14:24:27 PDT 2012


On 04/03/2012 01:53 PM, Martin Spacek wrote:
> On 12-04-03 09:29 AM, Chase Douglas wrote:
>> On 04/02/2012 08:28 PM, Martin Spacek wrote:
>>> Hello,
>>>
>>> About a year ago, Chase Douglas suggested I post about this here. I've only
>>> recently become annoyed enough to finally do so.
>>>
>>> When I add a secondary monitor horizontally to my primary one, my synaptics
>>> touchpad (on a Thinkpad W510) becomes much more sensitive horizontally than
>>> vertically. If I add the secondary monitor vertically, it becomes more sensitive
>>> vertically than horizontally. Others have reported the same issue on different
>>> machines. Here are the relevant bug reports:
>>>
>>> https://bugs.freedesktop.org/show_bug.cgi?id=31636
>>>
>>> https://bugs.launchpad.net/xorg-driver-synaptics/+bug/726832
>>>
>>> Surely this can't be a design feature, can it? No other input devices (external
>>> mouse or trackpoint, both via evdev apparently) behave this way, and neither
>>> does the touchpad in Windows. This essentially breaks the touchpad for me on a
>>> multi-monitor setup. Fortunately, when plugged into an external monitor, there's
>>> usually access to an external mouse, which is perhaps why this one has flown
>>> under the radar for so long. This bug has been around for coming up to a year
>>> and a half. Could someone maybe have a look at it?
>>
>> I have heard from a few people that they actually prefer this behavior.
>> I think we either need to stick with what we have or make it
>> configurable (and maybe change the default).
>>
>> Peter, any thoughts?
>>
>> -- Chase
> 
> Chase, I find it incomprehensible how anyone could want this behaviour. Did any 
> of them explain why they prefer it? Could you maybe get them to comment on this 
> thread?

The reason they gave was that you would need to drag your finger much
farther with every additional monitor you add, when you need to go from
one side of the X screen to the other.

> Again, the inconsistency with the multi-monitor behaviour of other pointing 
> devices within Linux is glaring, as is the inconsistency with touchpads on other 
> OSes. It would seem to me that going against the consensus across pointing 
> devices and OSes (at least, as perceived by myself) should require extraordinary 
> justification.

I personally agree with you, but such a change would affect everyone who
uses a trackpad with multiple monitors on X.org today. I think we can
make the change and tell people to get over it :), but it's not just my
call.

-- Chase


More information about the xorg-devel mailing list