Beiträge von RevolutionViewing

    Sorry if I'm missing something obvious here, but I'm trying to unload/destroy a krpano instance using the removepano function, but "removepano" appears to be undefined, even after krpano has set itself up.

    I wondered if it was supposed to accessed on the element itself (eg, document.getElementById("krpanorv").removepano()) but that doesn't seem to work either.

    Is the documentation on this outdated, or am I doing something stupid?

    See here:
    Mac Safari 10.13.2 360 video playback ISSUE

    Here the Apple bug-tracker:
    https://bugs.webkit.org/show_bug.cgi?id=181445

    In Safari 11.1 (available as Safari Preview) this is already fixed.

    The fix in that thread worked perfectly, thanks!

    This whole thing just feels like Apple were in such a rush to get their Spectre fix out that they didn't test it properly.

    On January 8th Apple released an update for Safari as part of its Spectre mitigation. Unfortunately, this new version seems to prevent 360 videos from playing.

    This example just displays a black screen:

    https://d8d913s460fub.cloudfront.net/krpanocloud/vi….html?v=119pr14

    While others show that they're playing (and the timeline scrubber progresses as expected), but the pano is stuck on the first frame of the video:

    https://d8d913s460fub.cloudfront.net/krpanocloud/vi….html?v=119pr14

    https://d8d913s460fub.cloudfront.net/krpanocloud/vi….html?v=119pr14

    You can still pan around the static image, and in all examples with audio, the audio plays correctly despite the video itself being frozen.

    Is this something you're aware of, and is it fixable?

    -Dan

    Hi Klaus,

    Thanks for replying. Good to know that it'll be addressed in the next version, really appreciate you taking the time to adjust things for something that's so far outside krpano's normal usage!

    Edit - Cordova lets you forcibly set the user agent, so thanks to your help I've been able to get it to work by simply stripping out the '; wv' bit and manually pushing the user agent back to the browser. It's a bit messy but it works, so thanks again!

    We've been embedding krpano in our apps using Cordova for a while now, and never had any issues with it. However, we're now struggling to get VR working on Android.

    If we upload the content to web, it's correctly detecting the device. However, as soon as it's in the app it's failing and reporting an unknown device.

    I can't find much documentation about how the device detection feature works, other than a mention that it works off user agent on Android. I've done some testing and the user agent is slightly different. Not by much, but perhaps enough to fool whatever check is in place? (There's an extra "; wv" in the build and and extra "Version/4.0").

    Code
    Chrome browser:
    Mozilla/5.0 (Linux; Android 7.0; SM-G920F Build/NRD90M) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/62.0.3202.84 Mobile Safari/537.36
    Chrome in Cordova:
    Mozilla/5.0 (Linux; Android 7.0; SM-G920F Build/NRD90M; wv) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Chrome/62.0.3202.84 Mobile Safari/537.36

    And I'm not sure if it counts, but krpano also returns a different screen size depending on the method used:

    Screen Size

    • Krpano running in browser: 5.1
    • Krpano running in app: 5
    • Cordova screen detection plugin: 5.06

    I'm guessing this is a complicated question that's outside the scope of what you'd usually deal with here, but any suggestions as to how we might be able to make this work (possibly by fooling the device detection?) would be greatly appreciated.

    Hi,

    We're building a Virtual Tour using 360 video. I have an audio track and the current progress of this track defines which video source to show. These two need to be separate, as when the audio is paused, we need the video to continue looping. There are multiple ways to change the progress of the video, through audio progressing normally and hitting milestones, through the scrubber on the bottom, or via the backwards and forwards arrows.
    I have populated the current version with some videos we already have, so it may all seem a little random.

    On desktop, these video sources load seamlessly, with all of the methods of changing soures. However support on mobile is inconsistent. On iOS, strangely, it works how it should. On Android, the only way the video source will update is if you allow the audio to progress normally. Scrubbing a position on the timeline, or using the backwards and forwards arrows, requests the new video source with KRP, but this request isn't fulfilled and never returns as ready.

    I set up trace() calls when the video is requested, and when the video is ready. As you can see with the example link below.

    I have been testing on an iPhone 7 with Safari 10.3.3, a Google Pixel with Chrome 60, and A Samsung Galaxy S6 with Chrome 61.

    Test Link

    If you have any questions, please let me know.

    Thanks!

    EDIT:
    I'm aware it's probably a mobile user gesture issue, how mobile gernarally needs a click event to play audio. But it's weird on this with the natural timeline changing the source working fine, and the click events not.

    Hi,

    We're developing an app that will be locked into portrait view, and we want to have VR-enabled 360s. Obviously currently this just results in extreme letterboxing (below) as the device never rotates and orientation lock isn't supported on iOS.

    I appreciate that this is a very niche use case, but is there any way to force krpano to change the direction of the screen split, so that it would display correctly in our app when held in portrait?

    Thanks,
    Dan

    We've received emails from three separate clients in the past 6 months where their cube faces have been scrambled. This has completely baffled us, we cannot replicate it on our machines (or browserstack.com), and their browsers specs are well within reasonable parameters. There are no errors in Krpano's console, or Chrome's developer console.

    I've seen previous threads where this was an issue on Android and iPad devices, but I have yet to see this as an issue on desktop. Potential explainations mention GPU memory issues, which would explain why we can't replicate this, even if we emulate Chrome 20.

    Does this have anything to do with Chrome's indecisive reporting of CSS 3D support? If anyone can share any insight into the issue, it would be greatly appreciated.

    Hi everyone,

    So the project I'm working on has the WebVR button, and for a few reasons, we need a separate button to toggle only the gyro. The problem I'm having, is that when WebVR is inluded, my usual method of toggling the gyro doesn't work. (Just for some background, we use a JS/KrPano call to enable and disable the gyro plugin.)

    I've dug around the vtourskin.xml and the webvr.xml, but I can't see anything that would interfere, since my gyro toggle doesn't call WebVR.

    If someone has a clue how to separate the two, I'd greatly appreciate the tip.

    Thanks.

    We noticed today that the HTML5 tours don't seem to work that well in the latest version of Chrome (55), but only on Windows 7:

    I appreciate that Windows 7 is fairly old and no longer fully supported, but since a sizable number of people still use it would be good to know if there's a fix for this issue without having to resort to using Flash.

    Thanks,
    Dan

    I'll try to explain this as best I can.

    We have a client who wants us to include some of their old virtual tours in a new presentation we're making for them. We requested the original files, and they've gone to their original suppliers and come back to us with the "original images" used to create their presentation.

    The images in question seem to belong to a type of panorama I've never seen before consisting of eight portrait, rectangular images. Six around, one up, one down. The files seem to piece together to make a hexagonal prism - with the top and bottom faces overlapping.

    Their original tour was made in Pano2VR, but Pano2VR won't accept the files. As there's no overlap between the six horizontal images, we can't stitch them.

    Has anyone ever encountered this type of panorama before? How we can convert them to a format KRPano accepts?

    it looks fantastic, and just down the road from me!! just full screen option as mentioned above would be nice.

    was this done with just krpano? no panotour or other?

    Glad you enjoyed it!

    The lack of full screen was a tough choice, but there were a few elements of the project that lead to it being removed. Interestingly though, our analytics seem to show that the number of people who actually click the "full screen" buttons in our tours is relatively low - this could just be a quirk of our users, but it helped us make the decision.

    And yeah, it was all built in KRPano/JavaScript - we've never really used Panotour or its equivalents.