Touchpad stylus mode

[Bert Freudenberg]

Hi Zeph,

is there a way to get at the raw values for both sensors?

Also, if you indeed scale the values to the screen height, it's  
pretty much useless for painting or writing, which was the whole  
point of having it. Besides, getting up/down events from the stylus  
is essential for any writing/painting, so it would be nice to get  
this in first.

- Bert -

On Mar 13, 2007, at 21:42 , Zephaniah E. Hull wrote:

> Sorry for the delay.
>
> 1 should work fine, grabbing the device should get you the coordinates
> properly and everything should more or less Just Work.
>
> There seems to be a bug somewhere with the scaling code, in that  
> the PT
> device isn't getting scaled to the screen size, that's on my todo list
> to figure out the cause of.
>
> 2 is a bit harder.
>
> Firstly, the PT sensor is currently setup as an absolute device
> independent of the GS sensor, and it should be scaled to the screen so
> that there is a roughly 1 to 1 mapping between a point on the PT  
> sensor,
> and a point on the screen.
>
> Now, that's not the final state.  There was a lot of work spent in the
> latter part of last year on being able to do calibration, such as
> setting the scaling min/max values, flipping X and Y, rotation, and  
> the
> Z axis button threshold (A Z value >= that value causes a button  
> press).
>
> There is also protocol support for setting the area of the absolute
> device, the width of it on the screen, the hight on the screen, the  
> X/Y
> offset on the screen to move that area around, what screen in a
> multi-screen setup, and the ability to specify that the area offset is
> being pushed around by another input device.
>
> Now, note that I said that there is _protocol_ support.  Work on this
> more or less stopped when we hit the B1 touchpad issues that caused  
> the
> change from being able to use both the GS and PT and the same time, to
> having to switch between them.
>
> Further, that protocol support is not in any current release, but  
> is in
> the mainline git tree, and should hit maybe the next xorg release.
>
> Some work is going to be required to get xf86-input-evdev to do
> something with the new fields, and some server work is going to be
> needed to hook in the following support.
>
> At some point I need to move a machine over to a git X server and  
> start
> work on adapting xf86-input-evdev to it, but exactly when and how
> quickly is still up in the air.
>
> Zephaniah E. Hull.
>
> On Wed, Mar 07, 2007 at 10:14:23AM -0500, Jim Gettys wrote:
>> I think there are two logical operations, given the though I've  
>> had on
>> the topic:
>>
>> 1) if the client explicitly asks for a absolute pointing device as a
>> secondary input device in the XInput , we give them exactly what they
>> ask for, telling them they have a tablet of the the appropriate size,
>> etc.
>>
>> 2) if not, and the tablet device is being "ored" into the default
>> pointing device with the capacitive sensor, then the area on the  
>> screen
>> that is reported as coordinates from the tablet sensor might want  
>> to be
>> pushed around on the screen, as the relative pointing device  
>> (touchpad)
>> pushes to the edge of the tablet.
>>
>> How far we are from things working this way, I'll let Zephaniah  
>> answer.
>>
>>                                        - Jim
>>
>>
>> On Wed, 2007-03-07 at 10:19 +0100, Bert Freudenberg wrote:
>>> *bump* Anyone?
>>>
>>> On Mar 2, 2007, at 15:40 , Bert Freudenberg wrote:
>>>
>>>> The B2 now can be used as a graphics tablet, hardware-wise at
>>>> least. Great!
>>>>
>>>> Now there needs to be some driver adjustment IMHO. The resistive-
>>>> mode movement is simply mapped to the core pointer right now,
>>>> moving about the upper third of the display, but without generating
>>>> any down/up events. It's pretty much useless for writing or  
>>>> painting.
>>>>
>>>> So, what is the plan here?
>>>>
>>>> Will it be mapped to a secondary xinput device, so activities can
>>>> handle it themselves? Or do all activities need to put a writable
>>>> area on top for the pad to be useful (provided up/down events are
>>>> delivered for the core pointer)?
>>>
>>> - Bert -
>>>
>>>
>>> _______________________________________________
>>> Devel mailing list
>>> Devel at laptop.org
>>> http://mailman.laptop.org/mailman/listinfo/devel
>> -- 
>> Jim Gettys
>> One Laptop Per Child
>>
>>
>
> -- 
> 	  1024D/E65A7801 Zephaniah E. Hull <warp at aehallh.com>
> 	   92ED 94E4 B1E6 3624 226D  5727 4453 008B E65A 7801
> 	    CCs of replies from mailing lists are requested.
>
> If books were designed by Microsoft, the Anarchist's Cookbook would
> explode when you read it.
>   -- Mark W. Schumann in the Scary Devil Monastery.