Second Life Oculus Rift viewer 4.1.0.317313: update and JIRA

The Second Life Oculus Rift project viewer has been updated to support the Oculus CV-1 - but not without issues

The Second Life Oculus Rift project viewer has been updated to support the Oculus CV-1 – but not without issues

Update: July 8th: Linden Lab has suspended viewer support for the Oculus Rift. This article has been updated accordingly, notably with strikethroughs on links which are no longer valid.t.

On July 2nd, I posted about the release of the latest Oculus Rift project viewer, version 4.1.0.317313. As I’m actually Riftless, I could do little more than take a surface poke at the viewer and leave it to others to have a more detailed look – and they have done so, and found things to be less than favourable.

Ai Austin / Austin Tate is perhaps best placed in terms of overall feedback having gone through using an Oculus Rift HMD with the new viewer in several scenarios, all of which he has documented in his own blog, some of his initial finding having been reported in the comments following my original article. The problems he’s encountered include:

  • In all cases and with all graphics settings tried, the HMD view is over bright and washed out pastel in colour. He also notes the Pixel Luminance Overdrive setting, which had been present in earlier versions of the Rift project viewer is now absent
  • A failure to show any objects, wither in-world or attached to an avatar, with full or partial transparency when in HMD Mode (so, avatar hair, for example will not render). AI found that disabling Atmospheric Shaders in the viewer resolved this – but is not entirely a desirable solution
  • The image resolution in the HMD is low and jagged, and altering the viewer’s graphics settings apparently has no impact
  • Numerous UI-related issues in HMD mode, including: UI elements in fixed positions which cannot be changed via viewer settings; object and avatar labels and interaction icons fail to show; mouse pointer fails to display
  • Additional visual and display issues.
AI Austin illustrates one of the visual issues with the new Oculus Rift projects viewer: one the left, a scene rendered in the viewer when not in HMD mode; on the right, when rendered in HMD - note the washed-out Linden water in particular

Ai Austin illustrates one of the visual issues with the new Oculus Rift projects viewer: one the left, a scene rendered in the viewer when not in HMD mode; on the right, when rendered in HMD – note the missing transparencies in thew avatar’s hair, giving the impression of hair loss and the “missing” hot tub water. Credit: Ai Austin / Austin Tate

His experience mirrors that of other Rift users, including TTech (who also commented on this blog), and a number of users who have tried this 4.1.0 release of the viewer and have commented on the Oculus Rift forum thread ( see feedback commencing with this message onwards).

Commenting on the viewer at the Simulator User Group meeting on Tuesday, July 5th, Oz Linden said:

At this point, I don’t have any real comment.  It’s a Project Viewer, and one explicitly labelled Experimental at that… the point is for people to try it and let us know what they find out.

To help let the Lab know what people find out in using the viewer, I’d like to point to a bug report – BUG-20130 – raised by Rai Fargis.

While I am flattered that staff at the Lab do read this blog, if you are experimenting / trying the new Oculus Rift project viewer and experience specific issues, please add them to bug report rather than documenting them in the comments following this article (general feedback here is welcome, obviously). Doing so, and including with information on your system set-up, relevant log files, etc., guarantees your feedback is seen and recorded by the Lab, encouraging them to investigate issues.

When reporting problems, one thing to keep in mind is that this version has leapt forward several iterations in terms of the Oculus SDK; therefore comparisons with earlier versions of the viewer may not be helpful (outside of possibly pointing to removed options which proved useful in dealing with specific issues in the past). Rather, specifics of issues encountered with this version will offer a better means for the Lab to start / continue investigations.

At the moment, viewers operating in HMD mode have no means to tell the simulator they are doing so. Therefore, the Lab doesn’t have a means of accurately determining the numbers of people using Rift HMDs – and metric which could be useful in the future; as such, it is something which may change with a future update to the viewer.

With thanks to Ai Austin, TTech, Rai Fargis, and Jeanette Doobie

10 thoughts on “Second Life Oculus Rift viewer 4.1.0.317313: update and JIRA

  1. Ai Austin

    Good summary thanks Inara… Its “Ai” with a small “i” by the way rather than the call capitalised “AI”… in any case people think I am an NPC all the time in world🙂

    Like

    Reply
    1. Inara Pey Post author

      I did actually know that… but I think my brain has been out to lunch the last few days. Either that, or it’s a purely subconscious thing with me!. Fixed & apologies!🙂 .

      Like

      Reply
  2. Ai Austin

    For anyone testing, here is my suggestion to make progress… in Settings -> Graphics… turn Atmospheric Shaders off. That allows MUCH more to display and returns the view to near normality.

    Turn off the UI with Ctrl+Shift+U so its not obscuring you view as its likely it may be in a fixed position across the middle of your screen. The Rift HMD Mode UI adjustment settings in Settings -> Move & View -> Head mounted Display do not appear to work at the moment.

    Use the 2D/non-HMD Mode view for all interaction with objects.

    But then you can at least view ar3aes in Second Life in VR on DK2 and CV1 though the colours will be very bright and washed out.

    Like

    Reply
  3. Ai Austin

    The two images that show the water issues I think are two separate problems.

    a) No show for any object with any level of transparency of textures with transparency in them. The water in the hot tub, and the balcony railings, etc are examples of this. This can be fixed by selecting only ONE of “Atmospheric Shaders” or “Advanced Lighting Model” but not BOTH at the same time.

    a) Drop out of some sea patches for entire regions, even ones directly adjacent to the current position. Several such patches can drop out and this seems to happen for some further away (multiple) water patches even when the fix above is used.

    Like

    Reply
    1. Inara Pey Post author

      Yup… I was going to add commentary on the transparencies (particularly noticeable in your hair), but opted to focus on the washed-out sea rather than have a too long an explanation of the transparency issue. I’ll add a brief pointer to the bullet point.

      Like

      Reply
        1. Inara Pey Post author

          OK. I was taking your lead from your comment preceding the images, vis: “In the HMD view, one or more sea patches show as light blue rather than the correct water surface. This improved with Atmospheric Shaders off, but some sea area blue patch drop outs still occur.”🙂 .

          Like

          Reply
          1. Ai Austin

            Ah, I see. The “blue sea patch drop out” I am referring to is what we see for the sea surface of a while region where it shows light blue and sometimes flashes. Its been a problem for along time when Object Occlusion is on and was fixed in recent LL viewers with an adjustment of the Z level of the water surface (I Think). Its a typical “Z fighting” problem. My guess is this is occurring I this viewer as the image resolution is so low, so the fix made still leave the sea surface patch fighting for the se level Z… Long explanation. I guess it is in a way a Linden labs water se level issue actually.

            But note that the hot tub in the image is not Linden Water, its 5m above the normal region sea water level and is a simple partially transparent animated texture object, so is not rendered at all if you have BOTH Atmospheric Shader and Advanced Lighting on.

            I hope LL can find the root cause of all these issues and issue an updated viewer soon.

            Like

            Reply
            1. Inara Pey Post author

              Yup. the Occlusion flickering had been a long-standing issue, recently fixed.

              Anyway, I’ve revised the caption to focus on the transparency issue. Apologies for the the confusion. As you’d already demonstrated the the transparency issue in an earlier image, I assumed this was more focused on the Linden water issue + the “washed out issue you’d previously referred to.

              Like

              Reply
              1. Ai Austin

                You are right, the order and use of my illustrations was not very good. I modified the text and order of the image a bit to try to keep the issues separate. The truth is that both sets of images I used illustrate multiple problems.

                Like

                Reply

Have any thoughts?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s