Evaluation of Viewport Size and Curvature of Large, High-Resolution Displays

1
Evaluation of Viewport Size and Curvature of
Large, High-Resolution Displays
Lauren Shupp, Robert Ball, John Booker, Beth
Yost, Chris North
Virginia Polytechnic and State University Center
for Human Computer Interaction, Department of
Computer Science http//infovis.cs.vt.edu
2
Outline

• Motivation Related Work
• Experimental Design
• Results Observations
• Conclusions Impact
• Questions

3
Related Work

• Physical constructionMark Hereld, Ivan R. Judson
  and Rick L. Stevens. Tutorial Introduction to
  building project-based tiled display systems
• Software for distributing the graphicsGreg
  Humphreys, et. al. Chromium A stream-processing
  framework for interactive rendering on clusters
• Most usability research is for collaborationScott
  Elrod, et. al. Liveboard A large interactive
  display supporting group meetings, presentations,
  and remote collaboration.

4
Related Work

• Partitioned spaces vs. one large spaceJonathan
  Grudin. Partitioning digital worlds Focal and
  peripheral awareness in multiple monitor use.

• Horizontal monitors as separate rooms and
  vertical as single spacesRobert Ball and Chris
  North. An analysis of user behavior on
  high-resolution tiled displays.

• Larger displays narrow gender gap on spatial
  performanceMary Czerwinski, Desney Tan and
  George Robertson. Women take a wider view.
• FocusContext screensPatrick Baudisch,
  Nathaniel Good, Victoria Bellotti and Pamela
  Schraedley. Keeping things in context A
  comparative evaluation of focus plus context
  screens, overviews, and zooming.

5
Related Work

• Large (3x3) high-resolution (38403072) displays
  can result in better performance than panning and
  zooming on smaller displaysRobert Ball and Chris
  North. Effects of tiled high-resolution display
  on basic visualization and navigation tasks.

6
Motivation Viewport

• Is bigger (viewport) better? Is there a display
  that is too big?

• More detailed data more context at once(3072
  x 10,240)
• Physical navigation using eye, head, and body
  movement

7
Motivation Curvature

• What happens if we curve the display?

• Reduce time physically navigating
• Change physical navigation to less strenuous
  turning rather than walking

8
Motivation Curvature

• All pixels are resolvable using only head and eye
  movements (2.75 times more resolvable pixels on
  our 24 monitor display)

9
Experimental Design

• Independent variable display condition
• Dependent variables time, accuracy, and
  mental-workload (mental demand, physical demand,
  effort, and frustration)

Flat Curved
1 monitor ?
12 monitors ? ?
24 monitors ? ?
8 users in each (40 total)
10
Experimental Design
Factor Level
viewport 1 monitor 12 monitors 24 monitors between subject
curvature flat curved between subject
task type search route tracing image comparison within subject
task difficulty easy hard within subject
11
Tasks

• Search

12
Route Tracing
portion of Expressway 402 East of Atlanta, GA
(labeled Highway 8)
13
Tasks

• Route Tracing

14
Tasks

• Image comparison

15
Results
Suggests both larger viewport sizes and curvature
improve user performance times
16
Results Task Specific
Both the 12 and 24 monitor conditions improve
performance over one monitor (approximately 2-6
times faster)
For route tasks, the 24 monitor condition
improves performance over the 12 monitor
condition (approximately 50 faster)
17
Results Task Specific
Curved displays improve performance over flat
displays independently of viewport size
For route tasks, the 24 monitor condition
improves performance over the 12 monitor
condition (approximately 50 faster)
18
Results
19
Observations

• Viewport Size
• One monitor used significantly more virtual
  navigation (pan zoom)
• 12 and 24 monitor used more physical navigation
  (standing, walking, leaning)
• Strategies changed by second Image comparison
  task (switch from serial pattern to overview ?
  target)

20
Observations

• Curvature
• Change in physical navigation
• Flat standing (5/8), walking
• Curved body and head turns
• Keyboard and mouse may have hindered users on the
  flat display from walkingmore
• Users changed their area of focus more frequently
  on the curved 24 monitor condition

21
Conclusions

• Larger viewport sizes improve performance
• Search 12 better than 1 mon
• Route 24 better than 12 mon
• Larger viewport sizes reduce virtual navigation
  and increase physical navigation
• Larger viewport sizes yield less frustration
• Curved displays improve performance time
• Physical navigation changes from standing and
  walking to turning when the display is curved

22
Impact

• 24 monitor display is not too big!
• Curve large displays for single users

23
Questions

• ?

Virginia Polytechnic and State University Center
for Human Computer Interaction, Department of
Computer Science http//infovis.cs.vt.edu