Mark Claypool and David Finkel claypooldfinkel cswpiedu Computer Science and Interactive Media amp Game Development Worcester Polytechnic Institute 1 Cloudbased Games Connectivity and capacity of networks growing ID: 328721
Download Presentation The PPT/PDF document "The Effects of Latency on Player Perform..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
The Effects of Latency on Player Performance in Cloud-based Games
Mark Claypool and David Finkel{claypool,dfinkel}@cs.wpi.eduComputer Science andInteractive Media & Game Development
Worcester Polytechnic Institute
1Slide2
Cloud-based Games
Connectivity and capacity of networks growing
Opportunity for cloud-based games
Game processing on servers in cloud
Stream game video down to client
Client displays video, sends player input up to server
2
Server
Server
Server
Thin Client
Cloud Servers
Player input
Game framesSlide3
Motivation and Challenges
Cloud-based gamesServer controls content, helps prevent piracyEasy to distribute games to many playersClient “thin”, so inexpensive ($100 for OnLive console vs. $400 for Playstation 4 console)Potentially less frequent client hardware upgradesChallengesRequires more downstream capacity than traditional network games (5000 Kb/s vs. 50 Kb/s [1])Latency since player input requires round-trip to server before player sees effects
3Slide4
Latency and Interactive Applications
Numerous studies on latency and interactive applications (e.g., VoIP)But interactions for games differentNumerous studies on latency and network games (e.g., car racing [2], role playing [3], first person shooter [4])But cloud-based has only thin client so results may differSome studies on latency cloud-based games (e.g., [6], [7])But do not measure impact on users
Few studies on latency cloud-based on users (e.g.,
[8])Still need more dataBut does not compare with traditional network games
Need to compare to traditional network games inform developers of game and cloud systems
4Slide5
Our Research
Measure impact of latency on players in cloud-based gamesTwo separate user studies – two games, two systemsUsers play games with controlled amounts of latencyMeasure objective (performance) and subjective (quality)Compare performance to traditional network games5Slide6
Teasers
Player performance degrades directly with increase in latencyEvery 100 ms latency means 25% decrease in performanceDegradation similar to traditional first person network gamesDespite difference in genres!E.g., third person game latency tolerance:Traditional network game 500 msCloud-based game only 100 ms
6Slide7
Outline
Introduction (done)User Studies (next)ResultsConclusion7Slide8
User Study 1
OnLive consoleConnects to laptop configured as routerDummynet on routerControl latency: 0-150msRouter connects to Internet then OnLive serversUsers play Crazy Taxi3rd person viewDeliver customers for pointsAbout 12 minutes totalUsers volunteers from campus8Slide9
User Study 2
GamingAnywhere LANClient connects to PC configured as routerDummynet on routerControl latency: 0-200msRouter connects to server on LANUsers play Neverball3rd person view“Roll” marble to goal as fast as possibleAbout 10 minutes totalUsers volunteers from campus9Slide10
User Study Summary
10Slide11
Outline
Introduction (done)User Studies (done)Results (next) DemographicsSubjectiveObjectiveTraditional network gamesConclusion11Slide12
Demographics
OnLive Crazy Taxi: 49 users95% 18-22 years old70% male75% “average+” game playing experienceGamingAnywhere Neverball: 34 users100% 18-22 years old90% male100%
“average+” game playing experience
12Slide13
QoE for OnLive
Crazy Taxi13Subjective opinions combined [12]Mean with standard errorLinear regression R2 0.92
QoE
drop of 30% over 150 msSlide14
QoE for GamingAnywhere
Neverball14QoE ranked for each testMean with standard errorFriedman test for correlation (p=0.002)Linear regression R2 0.86
QoE
drop of 40% over about 200
msSlide15
Points for OnLive Crazy Taxi
15Points for delivering customersMean with standard errorLinear regression R2 0.87
Score drop of 35% over
150 msSlide16
Times for GamingAnywhere Neverball
16Time to get marble to goalMean with standard errorLinear regression R2 0.70
Time increase of
35% over about 200 msSlide17
User Performance in Game Genres
17Traditional games Impact of latency depends upon perspective [5]First Person
most sensitive
Third Person less sensitiveOmnipresent least sensitiveSlide18
User Performance in Game Genres
18Convert objective measurements to performance degradationCloud-based games most closely follow first person avatar… Despite being
third person!Slide19
Conclusion
Cloud-based games increasingly relevantEffects of latency? Versus traditional network games?Two user studies on cloud-based games and latencyMeasure object (score) and subjective (QoE)Comparison with traditional gamesCloud-based games sensitive to modest latencies25% degradation for each 100 msMost similar to first person gamesEven if genre more tolerant to latency in traditional gamesFuture workAdditional user studies, other genres
Latency compensation for cloud-based gamesEffects of variation in latency (i.e., delay jitter)
19Slide20
Acknowledgements
GamingAnywhere NeverballJames AnounaZachary EstepMichael French OnLive Crazy TaxiRobert DabrowskiChrisitan ManuelRobert Smieja
20Slide21
The Effects of Latency on Player Performance in Cloud-based Games
Mark Claypool and David Finkel{claypool,dfinkel}@cs.wpi.eduComputer Science andInteractive Media & Game Development
Worcester Polytechnic Institute
21