Posts by benji33

    How is the default view of a panorama taken in account during spot naviguation of non oriented scenes ?

    For instance with this view tag the gyroscope returns the hlookat 0 & vlookat 0 position + gyroscope orientation as basis and in desktop the default view is respected with real hlookat and vlookat.
    If I remember the gyro1 plugin wasn't do that into the 1.19-pr3 version and was using real hlookat and vlookat sets positions.

    Something has changed with gyro2 if yes how to do? Thanks

    Hello,

    I've tested Gyro2 with partial panoramas and it seems there are some strange behaviors, see below. Tested on iOS 11.4 Safari with an iPhone 6S.

    For instance I've a partial panorama with this view tag and the gyroscope is enabled only after the onavailable event check action which is done after a 1 second delay, see below my settings :

    1. If I deactivate the gyroscope and activate it again, with the switch() action, on the same partial scene, the partial limits are lost and I can see the black part of the sphere.
    2. If I deactivate the camroll on the "onnewpano" event when the new scene is a partial one, to avoid the display of black parts in corners, I've noticed that the gyroscope can be freezed. It happens frequently.
    3. If the partial scene is the first one of the tour the partial limits are not respected and I can see the black parts of the sphere when the gyroscope is activated.
    4. Navigating to a new scene resolve all issues but not the freeze of the gyroscope. In this case only a switch of the enabled property is working.

    Thanks for the check.

    Hello,

    I've tested this code with Safari 11.1.1 on macOS.

    Code
    if(device.safariversion GE 11,
      trace('VERIFED');
    ,
      trace('NOT VERIFIED');
    );

    It returns NOT VERIFIED. The check is VERIFED with single "." version as iOS ones but not with macOS ones due to the string comparison.
    There is a way to do it easily without a string parse?

    Better solution at this time :

    Code
    indexoftxt(indexsafariversion, get(device.safariversion), ".");
    subtxt(majorsafariversion, get(device.safariversion), 0, get(indexsafariversion));
    if(majorsafariversion GE 11,
      trace('VERIFED');
    ,
      trace('NOT VERIFIED');
    );

    Hi,

    I've tried to setup a picture hotspot with a flying attibute. It works.
    I've added a hotspot which is covered by the flying picture, with the add of zorder/zorder2 values (flying picture receive a zorder of 10 and hotspot a zorder value of 3). It works.

    But

    If I add a text layer created with the addlayer method as child of the previous hotspot, the hotspot itself and the layer are always over the flying picture even with all zorder changes.

    Is it a known issue with children layers ?

    Thanks,
    Benjamin

    Hi,

    Of course the sample code above was designed for test only ;)
    Yes it's a workaround and not a fix ... sorry for the mistake.
    Thanks for the lesson about device motion/device orientation.

    I've tested the 2 libs for Chrome 65 on 3 devices and it works with them. But your previous workaround post with only the XML edit wasn't working.

    Thanks for this quick edit to partially solve the customers problem linked to the Chrome issue.

    Hi,

    With this fix mobilevr_sensor="0" I've no result.
    No move on vertical and horizontal axis with the fix and vertical only without the fix ! (tested with Huawei P8, LG G4 and Samsung S6)


    With krpano 1.19pr14 webvr sample it doesn't work also.

    My own code

    I've an issue with z-index and fullscreen.

    In fact I use external div element out of the krpano div and I apply my own fullscreen treatment.
    But since latests krpano version I've noticed that the z-index of the krpano element is applied to the highest zindex available (2147483647) and the others div cannot be displayed anymore with Android Chrome browser when rotating the screen (which apply the fullscreen event).

    Is it a krpano or Chrome issue/feature ?

    When Simulated VR mode is activated on Chrome desktop (PC) when rotating with the mouse on Yaw axis the view is refreshed at a previous position during "quick" rotations. Works good on Firefox.

    Many users report a difficulty to access a tour through the WebVR API with the Samsung Internet Browser for GearVR. It works great with the Oculus Browser.
    Do you know the difference between the 2 differents browsers? It seems that the Oculus Browser has a better use of the available resolution. John Carmack FB Post