Visibility CS 446: Real-Time Rendering

1
Visibility CS 446 Real-Time Rendering Game
Technology

• David Luebke
• University of Virginia

2
Happy Valentines Day!
3
Bittersweets Dejected sayings

• I MISS MY EX
• PEAKED AT 17
• MAIL ORDER
• TABLE FOR 1
• I CRY ON Q
• U C MY BLOG?
• REJECT PILE
• PILLOW HUGGIN

• ASYLUM BOUND
• DIGNITY FREE
• PROG FAN
• STATIC CLING
• WE HAD PLANS
• XANADU 2NITE
• SETTLE 4LESS
• NOT AGAIN

4
Bittersweets Dysfunctional sayings

• RUMORS TRUE
• PRENUP OKAY?
• HE CAN LISTEN
• GAME ON TV
• CALL A 900
• P.S. I LUV ME
• DO MY DISHES
• UWATCH CMT

• PAROLE IS UP!
• BE MY YOKO
• UMEGRIEF
• I WANT HALF
• RETURN 2 PIT
• NOT MY MOMMY
• BE MY PRISON
• C THAT DOOR?

5
Demo Time Ken Arthur
6
Homework

• The pitch, round 2
• Many ideas way too hard
• Art assets
• (Much) character animation
• Elaborate AI/gameplay
• Some of my favorites so far
• Gallery
• Mariokart
• Pirates (but just the ships)
• Combat billiards
• Orb (a la marble madness)
• Waverace

• Too hard
• Total War
• (Full-blown) RPG
• Ninjas
• UVA v. VaTech
• Yes to
• Terrain LOD, mocap data, NPR, Shadows
• No to
• Combat animations, AI, big worlds
• Out-of-the-box ideas?

7
Project preview

• Next 48 hours
• Pitch a few more ideas on forum
• Submit signup sheet by e-mail
• List 3 project ideas youd be happy to work on
• I assign groups on Thursday based roughly on
  interest overlap
• Note I assign groups, not project ideas!
• Assignment 3
• Groups build a basic engine
• Groups propose their final idea
• Later may reorganize groups, allow 3rd-party
  engines

8
5 Steps to Efficient Rendering

• Step 1 reduce vertices/triangle
• Step 2 reduce overhead of each vertex
• Step 3 recognize the existence of the vertex
  cache
• Step 4 display lists (deprecated re
  performance)
• Step 5 learn about the hardware
• Meta-step 0 profile and analyze code

9
Other things you should know

• Most interesting stuff in OpenGL isnt in red
  book
• Extension registry
• http//oss.sgi.com/projects/ogl-sample/registry/
• Extensions you should know
• ARB_multitexture
• EXT_framebuffer_object
• ARB_occlusion_query
• ARB_point_sprite
• ARB_texture_rectangle
• ARB_draw_buffers
• ARB_texture_float (_framebuffer_float,
  _half_float_pixel, etc)
• ARB_depth_texture
• ARB_vertex_buffer_object (_pixel_buffer_object)

10
Other things you should know

• Most interesting stuff in OpenGL isnt in red
  book
• Extension registry
• http//oss.sgi.com/projects/ogl-sample/registry/
• Developer sites
• http//developer.nvidia.com
• http//www.ati.com/developer
• Find demos, tools, tutorials, presentations,
  white papers

11
Visibility Calculations

• Motivation
• Avoid rendering redundant geometry
• Complements LOD, potentially even more powerful
• Basic idea dont render what cant be seen
• Off-screen view-frustum culling
• Occluded by other objects occlusion culling

12
Motivation

• The obvious question why bother?
• Off-screen geometry solved by clipping
• Occluded geometry solved by Z-buffer
• The (obvious) answer efficiency
• Clipping and Z-buffering take time linear to the
  number of primitives

13
The Goal

• Our goal quickly eliminate large portions of the
  scene which will not be visible in the final
  image
• Not the exact visibility solution, but a
  quick-and-dirty conservative estimate of which
  primitives may be visible
• Z-buffer clip this for the exact solution
• This conservative estimate is called the
  potentially visible set or PVS

14
View-Frustum Culling

• An old idea (Clark 76)
• Organize primitives into clumps
• Before rendering the primitives in a clump, test
  a bounding volume against the view frustum
• If the clump is entirely outside the view
  frustum, dont render any of the primitives
• If the clump intersects the view frustum, add to
  PVS and render normally

15
Efficient View-Frustum Culling

• How big should the clumps be?
• Choose minimum size socost testing bounding
  volume ltlt cost clipping primitives
• Choose minimum size so primitives can be rendered
  efficiently
• At least 500 triangles/clump on todays hardware
• Organize clumps into a hierarchy of bounding
  volumes for more efficient testing
• If a clump is entirely outside or entirely inside
  view frustum, no need to test its children

16
Efficient View-Frustum Culling

• What shape should bounding volumes be?
• Spheres and axis-aligned bounding boxes simple
  to calculate, cheap to test
• Oriented bounding boxes converge asymptotically
  faster in theory
• Lots of other volumes have been proposed
• Capsules, ellipsoids, k-DOPs
• but most use spheres or AABBs.

17
Cells Portals

• Goal walk through architectural models
  (buildings, cities, catacombs)
• These divide naturally into cells
• Rooms, alcoves, corridors
• Transparent portals connect cells
• Doorways, entrances, windows
• Notice cells only see other cells through portals

18
Cells Portals

• An example

19
Cells Portals

• Idea
• Cells form the basic unit of PVS
• Create an adjacency graph of cells
• Starting with cell containing eyepoint, traverse
  graph, rendering visible cells
• A cell is only visible if it can be seen through
  a sequence of portals
• So cell visibility reduces to testing portal
  sequences for a line of sight

20
Cells Portals
A
D
E
F
C
B
G
H
E
A
B
C
D
F
G
H
21
Cells Portals
A
D
E
F
C
B
G
H
E
A
B
C
D
F
G
H
22
Cells Portals
A
D
E
F
C
B
G
H
E
A
B
C
D
F
G
H
23
Cells Portals
A
D
E
F
C
B
G
H
E
A
B
C
D
F
G
H
24
Cells Portals
A
D
E
F
C
B
G
H
E
A
B
C
D
F
G
H
25
Cells Portals
A
D
E
?
F
C
B
G
H
E
A
B
C
D
F
G
?
H
26
Cells Portals
27
Cells Portals

• View-independent solution find all cells a
  particular cell could possibly see
• C can only see A, D, E, and H

A
D
E
H
28
Cells Portals

• View-independent solution find all cells a
  particular cell could possibly see
• H will never see F

A
D
E
C
B
G
29
Cells and Portals

• Questions
• How can we detect whether a given cell is visible
  from a given viewpoint?
• How can we detect view-independent visibility
  between cells?
• The key insight
• These problems reduce to eye-portal and
  portal-portal visibility

30
Cells and Portals History

• Airey (1990) view-independent only
• Portal-portal visibility determined by
  ray-casting
• Non-conservative portal-portal test resulted in
  occasional errors in PVS
• Slow preprocess
• Order-of-magnitude speedups

31
Cells and Portals History

• Teller (1993) view-independent view-dependent
• Portal-portal visibility calculated by line
  stabbing using linear program
• Cell-cell visibility stored in stab trees
• View-dependent eye-portal visibility stage
  further refines PVS at run time
• Slow preprocess
• Elegant, exact scheme

32
Cells and Portals History

• Luebke Georges (1995) view-dependent only
• Eye-portal visibility determined by intersecting
  portal cull boxes
• No preprocess (integrate w/ modeling)
• Quick, simple hack
• Now-archaic public-domain library pfPortals

33
pfPortals Algorithm

• Depth-first adjacency graph traversal
• Render cell containing viewer
• Treat portals as special polygons
• If portal is visible, render adjacent cell
• But clip to boundaries of portal!
• Recursively check portals in that cell against
  new clip boundaries (and render)
• Each visible portal sequence amounts to a series
  of nested portal boundaries
• Kept implicitly on recursion stack

34
pfPortals Algorithm

• Recursively rendering cells while clipping to
  portal boundaries is not new
• Visible-surface algorithm (Jones 1971) general
  polygon-polygon clipping
• Elegant, expensive, complicated
• Conservative overestimate (pfPortals) use
  portals cull box
• Cull box x-y screenspace bounding box
• Cheap to compute, very cheap to intersect