[x3d-public] X3D meeting minutes 18 FEB 2022: SpotLight angles, Volume Rendering component

Sat Feb 19 10:27:41 PST 2022

Attendees: Anita Havele, Michalis Kamburelis, Vince Marchetti, Nicholas
Polys, Dick Puk, Don Brutzman

0.	Preliminaries

a.	Vince discussed new spreadsheet showing X3D 4 player coverage,
updated to show view3dscene coverage.  Several additional refinements
discussed and applied.  This discussion continues on the x3d-public mailing
list.

*
http://web3d.org/pipermail/x3d-public_web3d.org/2022-February/016788.html
*
http://web3d.org/pipermail/x3d-public_web3d.org/2022-February/016789.html

b.	Anita discussed planning for upcoming Webinars, including X3D4 Web
Audio and Advanced Rendering.  She is reaching out to candidate presenters.

c.	Discussions in progress for Web3D 2022 Conference, please contact
Anita if you are interested in being an organizer. 

*	https://web3d.siggraph.org 

This week's minutes are integrated with the following original agenda in
order to provide details about really excellent discussions.

From: Brutzman, Donald (Don) (CIV) <brutzman at nps.edu> 
Sent: Tuesday, February 15, 2022 8:53 PM
To: Nicholas Polys <npolys at vt.edu>; Michalis Kamburelis
(michalis.kambi at gmail.com) <michalis.kambi at gmail.com>
Cc: X3D Public Mailing List (x3d-public at web3d.org) <x3d-public at web3d.org>;
Brutzman, Donald (Don) (CIV) <brutzman at nps.edu>
Subject: X3D meeting agenda 18 FEB 2022: SpotLight angles, Volume Rendering
component

1.	Regular Friday meeting 08-0900 Friday, call information below.

Videoconference Connectivity:

*
https://us02web.zoom.us/j/81634670698?pwd=a1VPeU5tN01rc21Oa3hScUlHK0Rxdz09 
*	https://zoom.us/j/148206572  Password 483805 
*	https://www.web3d.org/member/teleconference-information 

= = = = = = = = = = = = 

2.	Spiraling questions about SpotLight field relationships and default
values, beamwidth and cutoffAngle, has our team looking closely at potential
refinements to improve X3D4 modeling.  We spent an hour together focused on
this topic.

*	X3D4 Architecture, Lighting component, clause 17.4.4 SpotLight
*
https://www.web3d.org/specifications/X3Dv4Draft/ISO-IEC19775-1v4-CD1/Part01/
components/lighting.html#SpotLight

excerpted from X3D4 committee-draft specification:

SpotLight : X3DLightNode {

  SFFloat [in,out] beamWidth        π/4      (0,π/2]

  SFFloat [in,out] cutOffAngle      π/2      (0,π/2]

}

*	X3D4 Architecture, Lighting component, Figure 17.1 - SpotLight node
*
https://www.web3d.org/specifications/X3Dv4Draft/ISO-IEC19775-1v4-CD1/Part01/
components/lighting.html#f-SpotLightnode

*	[x3d-public] Defaults of SpotLight.cutOffAngle, beamWidth changed
(maybe by mistake) in X3D 3.3
*
https://web3d.org/pipermail/x3d-public_web3d.org/2022-February/016744.html

*	Mantis 441: 17.4.3 SpotLight -- Default Values (resolved 2009)
*	https://www.web3d.org/member-only/mantis/view.php?id=441

Michalis agreed to walk us through this.  Goal outcomes for confirmation,
plus conclusions from discussion:

a.	is Spotlight explanation diagram OK?  YES (although not very pretty)
b.	Is relationship between fields correct, typically beamWidth <=
cutoffAngle ?  YES
c.	Should we modify default values?  YES
d.	Should we improve prose for clarity?  NOT NEEDED (except for
changing < to <= wherever appropriate)

We noted that the default values were switched in X3D3.2 so that current
default values for beamWidth and cutOffAngle (highlighted in yellow above)
were similar to the geometric relationships shown in diagram.

Michalis walked us through the issues with tremendous detail, drawing live
in real time and demonstrating corresponding authoring-tool settings plus
live rendering updates.  Wow, super helpful, much appreciated!

Of especial interest is that glTF defaults (outerConeAngle=cutoffAngle pi/4,
innerConeAngle=beamWidth 0) are not matched by Blender or Unity, which have
even-narrower cutoffAngle values.  Here are results from investigation by
Michalis:

======================================

glTF:

cutOffAngle = pi / 4 (in X3D <= 3.2)

beamWidth 0

blender

cutOffAngle 45 degrees / 2 = 22.5 = pi / 8

beamWidth = don't know

unity

cutOffAngle 30 degrees / 2 = 15 deg = pi / 12

beamWidth = don't know

x3d 3.3 and 4

cutOffAngle pi / 2 (too large)

beamWidth = pi / 4

x3d 3.2

cutOffAngle = pi / 4 (consistent with glTF)

beamWidth = pi / 2 (beamWidth gets overridden and so is actually misleading
since it is > cutOffAngle)

instantreality docs: x3d <= 3.2

castle game engine / view3dscene: x3d <= 3.2

======================================

Vince proposal:

cutOffAngle = pi * 1/3 = 60 deg

beamWidth = pi * 2/8 =

Dick proposal:

cutOffAngle = pi / 2 / 4 = pi / 8

beamWidth = pi / 4 / 4 = pi / 16

Michalis preference(s): improve the defaults. 

Nicholas preference: improve the defaults.

Don preference: improve the defaults...  X3D3.2 defaults are dysfunctional
and not desirable. glTF defaults are most appealing to me because glTF-X3D
converters will work and then we don't have to talk about it anymore).
Nevertheless can happily "live with" and deploy any other improvement to
defaults.

Points of shared agreement:

a.	Revising default values will be useful.
b.	Avoid default values that are contrary to diagram.
c.	beamWidth=0 does not provide functional effect of full-intensity
lighting  but matches glTF defaults.
d.	Note that many legacy browsers did not change from X3D3.2 defaults -
which, in effect, results in full intensity out to cutoffAngle and then no
dropoff, which is in fact counterintuitive to the beamWidth and cutOffAngle
field definitions.

Point of history:  use of Spotlight capabilities in X3D models is quite
rare.  Several of us thought that the reason was likely because the legacy
default values make no sense.  So, there is not a strong sense that prior
(specifically two versions of) X3D defaults need to be preserved, since it
is unlikely that any legacy X3D model content is explicitly dependent on
default values.  Of course, default values of critical lighting values are
often intentionally/easily ignored by authors who provide desired values.

So we have gotten very close to complete consensus, and expect to
deliberately revise default values for Spotlight beamWidth and cutOffAngle.

Plan: think about it, discuss on x3d-public mailing list, and decide on new
default values no later than next week's X3D meeting.

Five hours following the meeting, Michalis posted further-detailed rationale
with excellent rendered-X3D images showing effects of different Spotlight
default values.  I plan to tweet high-resolution versions of these images
when posting the summary of this week's meeting on Twitter.

*
http://web3d.org/pipermail/x3d-public_web3d.org/2022-February/016794.html

"Implement + Evaluate" is one of our X3D design cornerstones. Everyone is
warmly invited to follow his advice as we "zero in" on final X3D4
resolution.  Deadline is next Friday's meeting.

Here are screenshots. I encourage everyone who participated today to

not just think about these numbers, but actually try them out in

existing X3D browsers on some existing scenes :) My screenshots may

well be biased by a particular scene I used now for testing. Go and

pick your own, and show us screenshots :)

= = = = = = = = = = = = 

3.	Working on Mantis issues for Volume Rendering component

*	X3D4 Architecture (aka abstract spec)
*
https://www.web3d.org/specifications/X3Dv4Draft/ISO-IEC19775-1v4-CD1/Part01/
Architecture.html

*	X3D4 Architecture, Volume rendering component
*
https://www.web3d.org/specifications/X3Dv4Draft/ISO-IEC19775-1v4-CD1/Part01/
components/volume.html

Nicholas and group discussed following issues.

a.	ShadedVolumeStyle shadows

*	Mantis 917: Table 41.7 Volume rendering component support levels -
ShadedVolumeStyle shadows support unclear
*	https://www.web3d.org/member-only/mantis/view.php?id=917

We agreed that addition of shadow fields to lighting effects for X3D4 Shapes
had no relevance to X3D Volume rendering.  The ShadedVolumeStyle shadows
refer to voxels in a volume that might "cast shadows" within the volume as
an advanced visualization technique.

Context: if anyone reading is thinking "my goodness how weird is this
technogeek issue," then please consider future medical professionals
considering a volumetric scan going into the health record of someone you
know, trying to apply coloring/shadowing techniques to determine morphology
for diagnosis and treatment of a shadowy mass within an organ.  Our work is
important, folks, medical use cases show that these questions really do
matter.

We will think about X3D4 Architecture specification prose a little more and
expect to resolve final wording during the next week.  All opinions welcome.

= = =

b.	Can we remove 'ordered' field from table?  Review response: YES.
Mantis issue is resolved.

*	Mantis 709: 41.4.4 ComposedVolumeStyle - Remove reference to ordered
field
*	https://www.web3d.org/member-only/mantis/view.php?id=709

= = =

c.	Field definitions missing

*	41.4.2 BoundaryEnhancementVolumeStyle descriptions for opacityFactor
and retainedOpacity fields
*	https://www.web3d.org/member-only/mantis/view.php?id=545

We discussed the nature of these fields and will continue with review during
next week's meeting.

= = =

d.	Contradictory naming of enumeration values in X3D specification

*	Mantis 867: 41.4.1 BlendedVolumeStyle - Inconsistent enumerations,
ALPHA0 ALPHA1 or ALPHA1 ALPHA2
*	https://www.web3d.org/member-only/mantis/view.php?id=867
*
http://web3d.org/pipermail/x3d-public_web3d.org/2022-February/016777.html

Useful discussion.  We settled on ALPHA1 ALPHA2 for greatest clarity when
used. These changes will appear in next revision of X3D XML Schema, X3DUOM,
tooltips and autogenerated tools.

= = = =

e.	Addition of fields to further constrain ProjectionVolumeStyle
intensity for improved insight and repeatability of visualization
diagnostics.

*	Mantis 499: 41.4.8 ProjectionVolumeStyle - add Fields
intensityThresholdFloor, intensityThresholdCeiling ?
*	https://www.web3d.org/member-only/mantis/view.php?id=499
*
http://web3d.org/pipermail/x3d-public_web3d.org/2022-February/016778.html

Helpful discussion.  Will revisit (and hopefully resolve) next week.

We agreed to accept a dozen other recently resolved editorial issues.  The
Mantis issues list tells the story.

*	Web3D Consortium Mantis Issue Tracker
*	https://www.web3d.org/member-only/mantis/view_all_bug_page.php
*	Use filter "X3D4 Resolution"

4.	Other topics of group interest. are always welcome.

Membership has value! Please consider supporting Web graphics by joining
Web3D Consortium.  

*	https://www.web3D.org/join 

Thanks for all preparations and inputs.  Have fun with X3D!   8)

all the best, Don

-- 

Don Brutzman  Naval Postgraduate School, Code USW/Br
<mailto:brutzman at nps.edu> brutzman at nps.edu

Watkins 270,  MOVES Institute, Monterey CA 93943-5000 USA    +1.831.656.2149

X3D graphics, virtual worlds, Navy robotics https://
faculty.nps.edu/brutzman

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://web3d.org/pipermail/x3d-public_web3d.org/attachments/20220219/66f43e86/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 5353 bytes
Desc: not available
URL: <http://web3d.org/pipermail/x3d-public_web3d.org/attachments/20220219/66f43e86/attachment-0001.p7s>