Re: [Linuxwacom-discuss] How to find out what is really supported on 056a:00e6

[Jason G. <kil...@gm...>]

On Sat, Dec 1, 2012 at 6:40 PM, Maximilian Federle <
max...@go...> wrote:

> Am 02.12.2012 00:08, schrieb Favux ...:
> > On Sat, Dec 1, 2012 at 2:42 PM, Maximilian Federle
> > <max...@go...> wrote:
> >> Hello Favux,
> >>
> >> thanks for your extensive reply! Following your suggestion I edited
> >> /usr/share/X11/xorg.conf.d/50-wacom.conf in Ubuntu to change the
> following
> >> rule:
> >>
> >> Section "InputClass"
> >>      Identifier "Wacom class"
> >>      MatchProduct "Wacom|WACOM|Hanwang|PTK-540WL"
> >>      MatchDevicePath "/dev/input/event*"
> >>      Driver "wacom"
> >> EndSection
> >>
> >> to
> >>
> >> Section "InputClass"
> >>      Identifier "Wacom class"
> >>      MatchProduct "Wacom|WACOM|Hanwang|PTK-540WL"
> >>      MatchDevicePath "/dev/input/event*"
> >>      Driver "evdev"
> >> EndSection
> >
> > That will disable your stylus.  You just did it as a test?  For custom
> > .conf files see:
> >
> http://sourceforge.net/apps/mediawiki/linuxwacom/index.php?title=Xorg.conf.d
> >
> http://sourceforge.net/apps/mediawiki/linuxwacom/index.php?title=USB_Tablets_with_Touch
> >
>
> Yes of course, just as a test.
>
> >> I confirmed in Xorg.0.log that the evdev driver recognized the wacom
> device
> >> as a touch device. However testing this config revealed (with xinput
> >> test-xi2 11) that the driver reports being capable of using two fingers
> >> only. With synaptics I had no luck whatsoever, it refused to load for
> the
> >> wacom finger device.
> >
> > Well looking at wacom_wac.c (a wacom.ko component in the kernel
> > /drivers/input/tablet) it appears the E6 has 2FGT:
> > static const struct wacom_features wacom_features_0xE6 =
> >       { "Wacom ISDv4 E6",       WACOM_PKGLEN_TPC2FG,    27760, 15694,
>  255,
> >         0, TABLETPC2FG, WACOM_INTUOS_RES, WACOM_INTUOS_RES,
> >         .touch_max = 2 };
> >
> > I don't know whether the X230t/E6 touchscreen hardware does
> > theoretically support more fingers or not.
> >
> > Favux
> >
>
> That's pretty much what I am looking for. Naturally I'd be completely
> satisfied if I could use the the hardware to the full extent, even if
> that means "only" two finger touch. I am just irritated by the fact that
> I saw this "5 touches supported" message in xinput output.


I think Favux may have missed the fact that you've built the driver from
git. While the version that is provided with Ubuntu doesn't support > 2
fingers, the very latest release (0.18.0) introduced support. As you found
out, all you need to do is turn off in-driver gestures through xsetwacom to
pass all the hardware-tracked fingers up to the X server.

The "5 touches supported" message is a bug. We don't actually tell the
server how many fingers the hardware supports, since it says that it is
capable of handling it internally. I believe Ping is working on a patch
that will fix this, since she recently noticed this as well.


> I looked for
> info on the internet regarding the x220t and x230t capabilities. One
> claim that I found quite often was, that the x230t hardware supports two
> touches independently and 5 in some sort of "semi/fake" mode.
> As I haven't touched a windows OS in like 8 years I can't verify this
> fact, but I'd be interested in whether the linuxwacom project intends to
> support such "pseudo-touch" devices.
>
> Here is a youtube video of this particular touchscreen in action with up
> to 5 finger touch: http://www.youtube.com/watch?v=EoYx9QRgVDA&feature=fvsr
>
>
It may be possible to support "fake" touches without changing the
multitouch API, but I'd guess there would be unpredictable results. There's
no good X/Y value we can send for the location of any fake touches, so
anything we use (e.g. 0,0) is likely to trip up clients that expect each
touch to have "real" X/Y data.


> So, what's next? I guess if it just isn't there in the code, I got as
> far as possible. Programming hardware drivers is not in my current
> skillset though, so I'd have to wait/hope/wish for someone to do it. (I
> can only send you german beer for motivational purposes... might lead to
> interesting side effects...) :(
>
> Max
>
>
I doubt Ping would have had the kernel driver use the 2FGT code if the
hardware tracked the location of all five touches, so there isn't really
much of a next step. You could see if the people behind the XI multitouch
patches (e.g. Peter Hutterer) would be interested in having the multitouch
API deal with fake touches, but with more and more hardware tracking all
the touches (for Windows 8) there's little chance they'd feel it to be
worth the effort.

Jason

---
When you're rife with devastation / There's a simple explanation:
You're a toymaker's creation / Trapped inside a crystal ball.
And whichever way he tilts it / Know that we must be resilient
We won't let them break our spirits / As we sing our silly song.