[x3d-public] Viewpoint centerOfRotation global or local

[Don Brutzman]

Thanks for the in-depth testing Andreas, very interesting and important.  This is a very subtle issue.  Am happily surprised that you found so much consistency already.

Hopefully x3dom and view3dscene can be adjusted to perform similarly, by you and Michalis respectively.  Sounds do-able now that you have isolated the anomaly.

Additional thoughts... It was a big step when we made viewAll a behavior of the currently bound viewpoint starting at the user's current position.  It allows a momentary transition from local coordinates to global perspective, a recentering (if you will) of context before the user's next move.  To me this seems like a much more effective alternative than some the navigational mode approaches we tried before.

The ability to "think globally while acting locally" is powerful.  Author still has to think about alternatives that are being offered to user, viewAll is not a magic wand, but it does add a lot to our most powerful user-navigation tool: viewpoints.

X3D4 addition of an automatically bound NavigationInfo for each Viewpoint, plus simpler exposure of nearDistance/farDistance culling, also offer promise for more effective exploration and interaction simply by using carefully authored Viewpoint nodes.  I sometimes like to think of a list of descriptions for corresponding Viewpoints as a scene storyboard.

As mentioned earlier, getting navigation paradigms such as viewAll consistently aligned across browsers can help in development of recognizable/intuitive modes that users find effective.  This certainly helps with larger Web3DUX objectives also.

Once browsers also start providing support for ViewpointGroup (a selectable list of Viewpoint nodes whose visibility is scoped within a virtual box) then I think we will also have a good path for large-scale composition of many models within larger scenes.

Have fun looking at things with X3D4!  8)


On 12/6/2020 7:48 PM, Andreas Plesch wrote:
> 
> Hi Don,
> 
> as suggested, there is a need to test browsers in that regard.
> 
> https://gist.github.com/andreasplesch/b96f20ccf0b6dd7af0cfb69f4a66e0fc
> has a simple scene with a transformed Viewpoint, as provided by a
> member of the x3dom-user mailing list. The center of rotation (COR)
> should also be transformed, and be centered on the green sphere.
> 
> Here are the results for some browsers using
> https://gist.githubusercontent.com/andreasplesch/b96f20ccf0b6dd7af0cfb69f4a66e0fc/raw/f8d03849300390dc814db52a506732eb00b4baf6/ViewpointCOR.x3d
> 
> All browsers transform the position of the viewpoint correctly.
> 
> x3dom: https://gist.githack.com/andreasplesch/b96f20ccf0b6dd7af0cfb69f4a66e0fc/raw/f8d03849300390dc814db52a506732eb00b4baf6/Viewpoint.html
>   rotates around red sphere, eg. COR is not transformed
> 
> view3dscene:
>   puts the center of rotation somewhere in between the red and green
> sphere, probably at the center of the total scene.
> 
> freeWrl:
>   center of rotation at green sphere, correct
> 
> Octaga:
>   center of rotation at green sphere, correct
> 
> InstantPlayer:
>   center of rotation at green sphere, correct
> 
> So there is a majority which follows your analysis. I agree that this
> is the most expected and least surprising behaviour.
> 
> Best regards,
> 
> -Andreas
> 
> On Sun, Dec 6, 2020 at 10:59 AM Don Brutzman <brutzman at nps.edu> wrote:
>>
>> To maintain momentum towards ballot, have added the following (flagged as editorsNote for continuing review/improvement if needed).
>>
>> On 12/5/2020 9:16 PM, Don Brutzman wrote:
>>> Simplest specification revision is to add a sentence to 23.3.1 X3DViewpointNode stating,
>>>
>>>       "The /centerOfRotation/ field is in the same local coordinate system as the current viewpoint node's /position/ and /orientation/ fields, in accordance with _4.3.5 Transformation hierarchy_.
>>
>> * X3D4 Navigation component, 23.3.1 X3DViewpointNode
>>     https://www.web3d.org/documents/specifications/19775-1/V3.3/Part01/components/navigation.html#X3DViewpointNode
>>
>> Cosmetic: am expecting that Dick and I will also sort the field descriptions in this clause, alphabetically by name.
>>
>> Also checked Mantis and found you had reported this issue in 2015 and it never got resolved (ouch).  Have flagged it as X3D4, updated, ready for final review.
>>
>> * Mantis 811, 23.3.1 X3DViewpointNode - Clarification of centerOfRotation
>>     https://www.web3d.org/member-only/mantis/view.php?id=811
>>
>> Noted related issue which can also be reviewed for completion:
>>
>> * Mantis 1194, 23.4.4 NavigationInfo VIEWALL (or X3DViewpointNode viewAll) should be a required navigation behaviour
>>     https://www.web3d.org/member-only/mantis/view.php?id=1194
>>
>> Thanks, keep 'em coming!  8)

all the best, Don
-- 
Don Brutzman  Naval Postgraduate School, Code USW/Br       brutzman at nps.edu
Watkins 270,  MOVES Institute, Monterey CA 93943-5000 USA   +1.831.656.2149
X3D graphics, virtual worlds, navy robotics http://faculty.nps.edu/brutzman