Page 8 of 39 FirstFirst ... 67891018 ... LastLast
Results 71 to 80 of 385

Thread: new xf86-input for macbooks and other multitouch-touchpads.

  1. #71
    Join Date
    Feb 2006
    Location
    Viña del mar, Chile
    Beans
    41
    Distro
    Ubuntu 10.04 Lucid Lynx

    Re: new xf86-input for macbooks and other multitouch-touchpads.

    Code:
    error: ‘MT_BUTTON_SWEEP_****' undeclared
    What happened? :O
    NiñoScript.age = new Number(16); //(>^;^<)
    (Wow, I did that signature when I was 16… I'm 20 now… it's sad to see how I'm slowly getting old u.u)

  2. #72
    Join Date
    May 2008
    Beans
    745

    Re: new xf86-input for macbooks and other multitouch-touchpads.

    Quote Originally Posted by NiñoScript View Post
    Code:
    error: ‘MT_BUTTON_SWEEP_****' undeclared
    What happened? :O
    I see, you are running xinput7... never got to test that. If you change the "SWEEP" lines in src/multitouch.c to "SWIPE", it should work better. Also, the swipe button mappings for xinput7 are probably wrong, as the proper buttons seem to be undefined in X at this point. Hopefully it will still work; after all it is the integrated touchpad that is most interesting right now.

    *goes to roll out a corrected patch sequence*

  3. #73
    Join Date
    Jun 2007
    Location
    Indian Trail, NC
    Beans
    96

    Re: new xf86-input for macbooks and other multitouch-touchpads.

    Any hints for those of us not running macbooks, but have multitouch pads? Is this possible? There are a load of us that would love to try this out. The default synaptics driver is giving us fits.

  4. #74
    Join Date
    Feb 2006
    Location
    Viña del mar, Chile
    Beans
    41
    Distro
    Ubuntu 10.04 Lucid Lynx

    Re: new xf86-input for macbooks and other multitouch-touchpads.

    Is there still a threshold for the "clicking finger" movement?
    In MacOSX that doesn't happen, you can move the lower finger a lot,
    as long as it stays lower than the other finger it doesn't move the cursor/scroll.

    Maybe it's a good idea to have a threshold, as if you really move it a lot,
    it may mean that you actually want to scroll or something with the lower finger;
    but then, it should be a big threshold;
    right now, the very small movement that my upper finger does on my whole hand
    (probably not bigger than 1 milimeter in the direction I'm moving my finger)
    is being enough to sometimes give a false scrolling.

    On the other side, I can swipe (sweep?) up and down to go back and forth in my browsers current tab history
    (In MacOSX it's swipe left and right, but who cares, this isn't MacOSX )


    BTW, horizontal scrolling isn't working too great here, I'm not sure how to explain the isue, but it's like very few huge steps instead of lots of small ones like the vertical one :\

    Oh, and thanks for fixing the "scroll at the top restarts X" bug, it was really anoying,
    mostly because a friend saw it and was teasing me by logging me out of my session, just putting 2 fingers on the top part of the trackpad (I hate you Yanko)
    Last edited by NiñoScript; April 14th, 2010 at 08:48 AM. Reason: hard to read if it was in just one line
    NiñoScript.age = new Number(16); //(>^;^<)
    (Wow, I did that signature when I was 16… I'm 20 now… it's sad to see how I'm slowly getting old u.u)

  5. #75
    Join Date
    May 2008
    Beans
    745

    Re: new xf86-input for macbooks and other multitouch-touchpads.

    Quote Originally Posted by NiñoScript View Post
    Is there still a threshold for the "clicking finger" movement?
    In MacOSX that doesn't happen, you can move the lower finger a lot,
    as long as it stays lower than the other finger it doesn't move the cursor/scroll.
    No special threshold for the clicking finger, but a finger can participate in a movement gesture regardless of where it on the pad. It was made that way purposely, but apparently it is still not how it works in OSX. Bummer.

    Maybe it's a good idea to have a threshold, as if you really move it a lot,
    it may mean that you actually want to scroll or something with the lower finger;
    but then, it should be a big threshold;
    right now, the very small movement that my upper finger does on my whole hand
    (probably not bigger than 1 milimeter in the direction I'm moving my finger)
    is being enough to sometimes give a false scrolling.
    This is how it works currently: The fingers participating in a move gesture are called the active fingers. The number of active fingers is the maximum of the number of pointing fingers and moving fingers. The pointing fingers are generally the ones above the clicking area, but a single finger can point anywhere on the pad, and the clicking fingers may rest above the clicking area as long as the bottom finger stays the same. The moving fingers are those moving jointly on the pad.

    From what I gather, this is almost correct, except movement should only be extracted from fingers strictly above the clicking area. How does that sound?

    On the other side, I can swipe (sweep?) up and down to go back and forth in my browsers current tab history
    (In MacOSX it's swipe left and right, but who cares, this isn't MacOSX )
    Nice

    BTW, horizontal scrolling isn't working too great here, I'm not sure how to explain the isue, but it's like very few huge steps instead of lots of small ones like the vertical one :\
    That is precisely how it is implemented. The scales for horizontal scroll and horizontal swipe are currently the same. I suspect separate scales for scrolling and swiping will fix this.

    Oh, and thanks for fixing the "scroll at the top restarts X" bug, it was really anoying,
    mostly because a friend saw it and was teasing me by logging me out of my session, just putting 2 fingers on the top part of the trackpad (I hate you Yanko)
    I intend to make it very hard for you to find problems like that in the mainline driver

    Cheers!
    Last edited by kosumi68; April 14th, 2010 at 09:33 AM.

  6. #76
    Join Date
    Feb 2006
    Location
    Viña del mar, Chile
    Beans
    41
    Distro
    Ubuntu 10.04 Lucid Lynx

    Re: new xf86-input for macbooks and other multitouch-touchpads.

    If you want, I can make some testing/asking in MacOSX and give a detailed report on how it exactly works
    Maybe even with video included
    I guess this weekend I may have enough time
    NiñoScript.age = new Number(16); //(>^;^<)
    (Wow, I did that signature when I was 16… I'm 20 now… it's sad to see how I'm slowly getting old u.u)

  7. #77
    Join Date
    May 2008
    Beans
    745

    Re: new xf86-input for macbooks and other multitouch-touchpads.

    Quote Originally Posted by NiñoScript View Post
    If you want, I can make some testing/asking in MacOSX and give a detailed report on how it exactly works
    Maybe even with video included
    I guess this weekend I may have enough time
    That would be cool, something for youtube (unless its already there :-O)

    One more question: What about two-finger click? Is that also unaffected by moving the clicking finger above the clicking area?

  8. #78
    Join Date
    May 2008
    Beans
    745

    Re: new xf86-input for macbooks and other multitouch-touchpads.

    Ok, new version in next. This might even become alpha-2 in master if tests are positive. The horizontal scrolling should work better now, and I believe the integrated button is pretty well covered; nothing happens when the clicking finger moves, up to the point where the other (bottom pointing finger) was when the fingers were put there. Scrolling with two fingers when either or both start off in the clicking area works smoothly, too.

    Cheers!

  9. #79
    Join Date
    May 2008
    Beans
    745

    Re: new xf86-input for macbooks and other multitouch-touchpads.

    Quote Originally Posted by jfrorie View Post
    Any hints for those of us not running macbooks, but have multitouch pads? Is this possible? There are a load of us that would love to try this out. The default synaptics driver is giving us fits.
    It all boils down to what hardware currently has kernel support for MT events. In the latest and greatest kernel, these drivers have MT support:

    drivers/hid/hid-3m-pct.c
    drivers/hid/hid-debug.c
    drivers/hid/hid-magicmouse.c
    drivers/hid/hid-mosart.c
    drivers/hid/hid-ntrig.c
    drivers/hid/hid-quanta.c
    drivers/hid/hid-stantum.c
    drivers/input/input.c
    drivers/input/mouse/bcm5974.c
    drivers/input/touchscreen/usbtouchscreen.c

    Some of them are available in 2.6.32, which lucid is based on. Some of them have been applied as extra patches to lucid. Some of them might be available as dkms packages. Which driver would apply to you?

  10. #80
    Join Date
    Nov 2006
    Beans
    22

    Re: new xf86-input for macbooks and other multitouch-touchpads.

    Thankyou!

    Had a couple of issues with the bcm5974 driver, and it unloading during X startup. If anyone else has the same issue:

    Add usbhid to the blacklist firstly, so it doesn't load - /etc/modprobe.d/blacklist.conf

    usbhid

    Add bcm5974, then usbhid to /etc/modules so that it loads as expected and then usbhid continues as normal afterwards

    bcm5974
    usbhid



    Anyway, it seems to be working well now, after using the Macbook for so long in MacOSX on the MBP for so long, the touchpad is key, and being without the drag/drop was an issue to say the least. Kudos to the developer(s)!

    One thing: I'm not sure of the correct terminology, but when you attempt to drag and drop (two fingers down, then dragging by moving one finger (normal). Is there any way to speed up the drag speed??

Page 8 of 39 FirstFirst ... 67891018 ... LastLast

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •