[JIRA] Updated: (VWR-16680) Add a 3rd viewer screen mode: 3D root window / background

Morgaine Dinova (JIRA) no-reply at lindenlab.cascadeo.com
Thu Jan 21 20:42:34 PST 2010


     [ http://jira.secondlife.com/browse/VWR-16680?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Morgaine Dinova updated VWR-16680:
----------------------------------

    Description: 
The Linden viewer and its direct derivatives have two alternative screen modes, windowed and fullscreen.  This exploratory Jira examines some of the issues of adding a third screen mode which renders to the "root window" or "background" of the desktop.

The reasons why this new screen mode could be of significant interest are several:

- Unlike in fullscreen mode, any applications which the user places on the chosen workspace of the desktop effectively become integrated into the viewer space without any work being required.

- The viewer would no longer need integrated 2D panels, but only edge decoration buttons and sliders to communicate with standalone applications.  Monolithic viewer bloat in 3D+GUI code would therefore plummet to 3D only, with everything else being refactored into application processes.

- The refactored code would be naturally disjoint concurrent system processes and therefore would not require any threading nor be subject to the many potential pitfalls of threading, such as race conditions and locking problems.

- The UI processes would naturally harness multiple cores automatically without any development work being needed.

- No code would be required for raising nor dismissing windows, because that is already handled by desktop launching and minimization.  Likewise resizing, docking, dragging across multiple screens, and so on --- it's all done by the operating system already.  Less code would mean a better, more maintainable and more stable viewer.

- The UI windows would be normal platform-compliant 2D applications using the native widgets, which improves visual consistency and makes their performance orders of magnitude faster than rendered in GL.  Likewise, input would be instantaneous instead of at the mercy of the always-laggy event handling in the GL rendering loop.

- Making the 3D virtual world your root or background window would be incredibly immersive, and would bring closer the vision of virtual worlds as the next generation of Internet interfaces.

Clearly to make best use of such a 3rd screen mode would require large scale refactoring and therefore is not a candidate for work at the present time.  However, refactoring the client has on numerous occasions been described as an important goal for Linden Lab, so there is merit in examining what kind of refactoring might be done and what potential benefits could arise from a good refactoring being chosen.

So many benefits would result from a refactoring into multiple applications plus root window rendering that it is worth considering this new screen mode as a potential goal, and reap the benefits of the viewer refactoring as a side effect.

Morgaine.

  was:
The Linden viewer and its direct derivatives have two alternative screen modes, windowed and fullscreen.  This exploratory Jira examines some of the issues of adding a third screen mode, "root window" or "background".

The reasons why this new screen mode could be of significant interest are several:

- Unlike in fullscreen mode, any applications which the user places on the chosen workspace of the desktop effectively become integrated into the viewer space without any work being required.

- The viewer would no longer need integrated 2D panels, but only edge decoration buttons and sliders to communicate with standalone applications.  Monolithic viewer bloat in 3D+GUI code would therefore plummet to 3D only, with everything else being refactored into application processes.

- The refactored code would be naturally disjoint concurrent system processes and therefore would not require any threading nor be subject to the many potential pitfalls of threading, such as race conditions and locking problems.

- The UI processes would naturally harness multiple cores automatically without any development work being needed.

- No code would be required for raising nor dismissing windows, because that is already handled by desktop launching and minimization.  Likewise resizing, docking, dragging across multiple screens, and so on --- it's all done by the operating system already.  Less code would mean a better, more maintainable and more stable viewer.

- The UI windows would be normal platform-compliant 2D applications using the native widgets, which improves visual consistency and makes their performance orders of magnitude faster than rendered in GL.  Likewise, input would be instantaneous instead of at the mercy of the always-laggy event handling in the GL rendering loop.

- Making the 3D virtual world your root or background window would be incredibly immersive, and would bring closer the vision of virtual worlds as the next generation of Internet interfaces.

Clearly to make best use of such a 3rd screen mode would require large scale refactoring and therefore is not a candidate for work at the present time.  However, refactoring the client has on numerous occasions been described as an important goal for Linden Lab, so there is merit in examining what kind of refactoring might be done and what potential benefits could arise from a good refactoring being chosen.

So many benefits would result from a refactoring into multiple applications plus root window rendering that it is worth considering this new screen mode as a potential goal, and reap the benefits of the viewer refactoring as a side effect.

Morgaine.


> Add a 3rd viewer screen mode:  3D root window / background
> ----------------------------------------------------------
>
>                 Key: VWR-16680
>                 URL: http://jira.secondlife.com/browse/VWR-16680
>             Project: 1. Second Life Viewer - VWR
>          Issue Type: New Feature
>          Components: Graphics
>         Environment: What compatibility is there between "root window / background" among platforms?
>            Reporter: Morgaine Dinova
>            Priority: Low
>
> The Linden viewer and its direct derivatives have two alternative screen modes, windowed and fullscreen.  This exploratory Jira examines some of the issues of adding a third screen mode which renders to the "root window" or "background" of the desktop.
> The reasons why this new screen mode could be of significant interest are several:
> - Unlike in fullscreen mode, any applications which the user places on the chosen workspace of the desktop effectively become integrated into the viewer space without any work being required.
> - The viewer would no longer need integrated 2D panels, but only edge decoration buttons and sliders to communicate with standalone applications.  Monolithic viewer bloat in 3D+GUI code would therefore plummet to 3D only, with everything else being refactored into application processes.
> - The refactored code would be naturally disjoint concurrent system processes and therefore would not require any threading nor be subject to the many potential pitfalls of threading, such as race conditions and locking problems.
> - The UI processes would naturally harness multiple cores automatically without any development work being needed.
> - No code would be required for raising nor dismissing windows, because that is already handled by desktop launching and minimization.  Likewise resizing, docking, dragging across multiple screens, and so on --- it's all done by the operating system already.  Less code would mean a better, more maintainable and more stable viewer.
> - The UI windows would be normal platform-compliant 2D applications using the native widgets, which improves visual consistency and makes their performance orders of magnitude faster than rendered in GL.  Likewise, input would be instantaneous instead of at the mercy of the always-laggy event handling in the GL rendering loop.
> - Making the 3D virtual world your root or background window would be incredibly immersive, and would bring closer the vision of virtual worlds as the next generation of Internet interfaces.
> Clearly to make best use of such a 3rd screen mode would require large scale refactoring and therefore is not a candidate for work at the present time.  However, refactoring the client has on numerous occasions been described as an important goal for Linden Lab, so there is merit in examining what kind of refactoring might be done and what potential benefits could arise from a good refactoring being chosen.
> So many benefits would result from a refactoring into multiple applications plus root window rendering that it is worth considering this new screen mode as a potential goal, and reap the benefits of the viewer refactoring as a side effect.
> Morgaine.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.secondlife.com/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        


More information about the Jira-notify mailing list