Page 2 of 8 FirstFirst 1234 ... LastLast
Results 11 to 20 of 73

Thread: KDE Vivaldi Tablet Finally Shipping For QA Testing

  1. #11
    Join Date
    Oct 2011
    Posts
    224

    Default

    Quote Originally Posted by e8hffff View Post
    If you google, I was an early supporter of PlasmaActive on the Nexus7.

    I come to the realisation it had little chance of getting successful as it was crippled by the ability to move to other devices easily. Mir solves this problem since it uses Android drivers, but Kubuntu teams have demonised Mir and have locked down to Wayland, therefore what hope is there. Some claim Wayland and the Sailfish project show promise of Android leanings, but no where near the visionn of Canonical's Mir which is a fundamental to them to use Android drivers.
    Some of the biggest bullshit posted on phoronix in a while. If you knew anything you'd know jolla employs the creator of libhybris. "Some claim wayland and sailfish show promise of android leanings" is some huge FUD.
    Stop the fanboism.

  2. #12
    Join Date
    Jun 2011
    Posts
    861

    Default

    Quote Originally Posted by caryhartline View Post
    Aaron Seigo is basically making a tablet for himself and a few friends.
    Wow I knew Aaron Seigo was popular but I didn't know he had that many friends that all the people that Make.Play.Live are making this for are only a few of them.

  3. #13
    Join Date
    Jan 2013
    Posts
    525

    Default

    Quote Originally Posted by n3wu53r View Post
    Some of the biggest bullshit posted on phoronix in a while. If you knew anything you'd know jolla employs the creator of libhybris. "Some claim wayland and sailfish show promise of android leanings" is some huge FUD.
    Stop the fanboism.
    Correct me if I'm wrong, but there was some talk about Wayland requiring EGL drivers that had additional information than what's usually in such drivers, so they couldn't use them as-is.

  4. #14
    Join Date
    Dec 2009
    Posts
    336

    Default

    I hope Plasma Active will start using Wayland soon, and also there will be native EGL drivers without need for libhybris.

    About libhybris read:


  5. #15
    Join Date
    Jul 2010
    Posts
    593

    Default

    Quote Originally Posted by intellivision View Post
    Correct me if I'm wrong, but there was some talk about Wayland requiring EGL drivers that had additional information than what's usually in such drivers, so they couldn't use them as-is.
    Weston doesn't require EGL at all (with Pixman backend). There's also a Raspberry Pi backend for Weston that's quite different from usual OpenGL ES stuff. So I don't think that's the case.

  6. #16
    Join Date
    Jan 2013
    Posts
    525

    Default

    Quote Originally Posted by Teho View Post
    Weston doesn't require EGL at all (with Pixman backend). There's also a Raspberry Pi backend for Weston that's quite different from usual OpenGL ES stuff. So I don't think that's the case.
    Software rendering, urgh.
    I had my fare share of LLVMpipe's <s>blisteringly fast performance</s>, I'm hoping they put some more optimisation into Wayland's software renderer.

    However, what differentiates 'native Wayland' EGL drivers from those that require libhybris?
    Last edited by intellivision; 07-22-2013 at 09:57 PM.

  7. #17
    Join Date
    Jun 2009
    Posts
    1,118

    Default

    as always it seems that is too mainstream to check facts?

    1.) Vivaldi has suffered many shortcomings because they are focused in open source friendly hardware and many providers fail them until they found the right one [aaron seijo has posted like zillion times including hangouts explaining why he rejected hardware that cannot be used using open source code]

    2.) Plasma-Active is integral part of KDE SC and is updated every release, maybe you are checking some freaking old code from before it got integrated into KDE git?

    3.) you can switch between plasma and plasma active realtime and even use both 1 per activity, kde 4.0 was years ago you know? is 4.11 this days

    4.) Wayland can run on anything that have GPU on it, close enough to the hardware that is actually impressive and it can use android drivers perfectly fine[lot better than Mir, after all libhybris was created FOR Wayland], it can run using software and it can run using SoC specific acceleration path[see collabora and raspberry PI example] too.

    5.) Qt5.1 support wayland almost perfectly[and Native not through some half done layer] and Qt-5.2[KDE SC 5 version] will fix all the corner cases remaining

    6.) Wayland software render is not your regular X11 software renderer, it basically uses pixman for very fast CPU operations and SHM for buffering Surfaces into RAM nothing else and is freaking fast even on ARM[youtube is your friend not your foe]

    7.) Vivaldi lastly is not a world domination device nor is your next google killer, is just a project focused of generate a portable device for developers[or KDE fan] FOSS friendly and vendor unlocked to enjoy freedom from device to software[again is everywhere kdeplanet, google, G+, youtube, phoronix, the H, forums, etc], so for the target audience we don't care if it have 30000dpi screen or if it have 32 ARM cores as long is offer openness and good FOSS support, is not a consumer device

    8.) Jolla is actually using wayland and is actually finished[unlike ubuntu phone it actually exist] and is in the process of getting partners[pretty much close to FirefoxOS work with ZTE] carriers and manufacturers, ofc i don't expect it on the next samsung GS5 but more in mid HTC/huawei/ZTE phones[like firefox it is very resource friendly and scalate quite well]

    9.) Mir is a project for 2014-2015, today is just vaporware with demos, so wait until it actually can do more than flip surfaces, so it can be compared properly against wayland

  8. #18
    Join Date
    Feb 2011
    Posts
    1,084

    Default

    Quote Originally Posted by intellivision View Post
    Correct me if I'm wrong, but there was some talk about Wayland requiring EGL drivers that had additional information than what's usually in such drivers, so they couldn't use them as-is.
    Mir has all the same requirements in this regard.

  9. #19
    Join Date
    Jun 2008
    Posts
    164

    Default Source of information about resolution.

    Quote Originally Posted by anybody View Post
    If they really still plan to launch a tablet with 800x600 resolution... well good luck finding people who are into cheap, crappy hardware.
    Also, it's probably still a TN display with bad viewing angles.
    Guys! Michael just copied over the old Vivaldi (formerly Spark) specs:
    http://makeplaylive.com/
    The new (rough) ones are here:
    http://rhombus-tech.net/community_ideas/kde_tablet/
    Blogs about hardware:
    http://rhombus-tech.net/community_id...e_tablet/news/
    The main card will be swappable, meaning you can upgrade SOC, RAM, etc:
    http://rhombus-tech.net/

  10. #20
    Join Date
    Jan 2013
    Posts
    1,446

    Default

    Quote Originally Posted by e8hffff View Post
    If you google, I was an early supporter of PlasmaActive on the Nexus7.

    I come to the realisation it had little chance of getting successful as it was crippled by the ability to move to other devices easily. Mir solves this problem since it uses Android drivers, but Kubuntu teams have demonised Mir and have locked down to Wayland, therefore what hope is there. Some claim Wayland and the Sailfish project show promise of Android leanings, but no where near the vision of Canonical's Mir which is a fundamental to them to use Android drivers.
    Do you realize that libhybris, which enables Mir to use Android drivers, was originally developed by Jolla, for Wayland, in order to use Android drivers in Wayland, and is still used by Jolla in order to use Android drivers in Wayland? So not only is it untrue that Mir has an advantage in this regard, the only reason Mir can even use Android drivers is because of Wayland. But then, the entire Mir couldn't exist without all the work done by Wayland developers.

    Wayland defines a stable protocol, which allows everyone to use it, and it is designed by the community, with the needs of the entire community in mind. Whereas Mir is a one trick pony, Canonical's in-house solution built solely for supporting their other in-house solutions (Unity), and to top that off Canonical makes no promises to consider anyone else's needs or even to maintain a stable API, which means that there's simply no way for KDE or anyone else to even start considering using Mir as their display server.

    You need to understand a difference between Wayland and Mir: Wayland is a protocol, which allows everyone to write their own compositors to suit their particular needs (which is what everyone is doing) while still maintaining compatibility and using shared resources at the lower levels of the stack (EGL, Mesa, etc). Mir is not, it's just a program, a single, monolithic program that is controlled by one corporation. If anyone wants to customize it to make it suit their needs, they can't, not without making it incompatible with the "real" Mir down the line, because Canonical doesn't promise to keep anything stable and doesn't promise not to break things. Whereas Wayland already has a guaranteed stable protocol, which allows the developers of display systems to work on their solutions right now.

    So tell me, why exactly should KDE, Jolla or anyone other than Canonical bother with Mir? Why, when they already have a much better alternative available (Wayland) that suits their needs much better? What possible reasoning could you have for wanting others to use Mir besides blind fanboyism? Maybe you have some new inside info that no one else knows? Please share, I'm listening.

Tags for this Thread

Posting Permissions

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