Game networking

The flashcards below were created by user Zidanny5 on FreezingBlue Flashcards.

  1. DiffServ and IntServ differences
    DiffServ provides QOS by differentiating traffic For example, DiffServ will reduce the latency in traffic containing voice or streaming video, while providing best effort service to traffic containing file transfers. whereas IntServ does so by building a virtual circuit internet using the bandwidth reservation technique i.e. Resource Reservation Protocol (RSVP).

    DiffServ does not require nodes of the network to remember any state information of flow whereas IntServ does by remembering state information in routers.

    • Furthermore, reserving paths and remembering state information in a busy
    • network such as the Internet would be a tedious task. Therefore, implementing IntServ would be practically difficult in the Internet. Because of that, IntServ would be suitable for smaller private networks whereas DiffServ is much suitable for the Internet.
  2. How would you use Latency to Cheat?
    Look-ahead cheating is a method of cheating in a peer-to-peer multiplayer gaming architecture where the cheating client gains an unfair advantage by delaying his actions to see what other players do before announcing its own action.

    • A client can cheat using this method by acting as if it is suffering from high latency; the outgoing packet is forged by attaching a time-stamp that is prior to the actual moment the packet is sent, thereby fooling other clients into thinking that the action was sent at the correct time, but was delayed in
    • arrival. A partial solution is the Lockstep protocol.
  3. QOS: DiffServ & IntServ explain QOS and these two services.
    • QOS (Quality of Service) refers to the overall performance of a computer network particularly the performance seen by the
    • users of the network. Network congestion is defined as packet rate>output rate. A service that resolves congestion issues known as DiffServ prioritises traffic flow. It does this by providing low latency for critical traffic and best effort (DiffServ should be better then best effort I think)for non-critical traffic. 
    • IntServ uses a parameterised approach by using Resource Reservation Protocol (RSVP) to request and reserve resources through a network.
  4. What are the advantages of using compression in network game communications?
    In network game communications, game data which is transmitted over the network may be compressed. The main advantage of using compression techniques is to reduce the overall network load.

    • For this purpose, compression techniques such as delta compression may be used. This means that at each update, the server does not need to send a snapshot of the entire game world state, and instead sends only
    • the changes, called a delta snapshot, that have taken place since the last update.

    • Since full-world snapshots are only sent when a game starts or when a client suffers temporary loss of connectivity, the overall required data transmission is greatly reduced, which can result in lower latency as
    • clients have less data to process at each update.
  5. What is meant by Interest Management and describe how IM may be used in P2P Massively
    Multiplayer Online Gaming?
    • The aim of interest management is to identify when objects inhabiting a game world can interact, while preventing those that cannot
    • interact from exchanging messages.

    • Interest management is especially important in Massively-Multiplayer Online Gaming, where a huge number of players exist in a
    • shared virtual space. For this reason, interest management is used to limit each player’s view of the game world to a relatively static Area of Interest (AOI).

    One of the most common methods of AOI management is the Region-based model, which partitions the game world into predefined sections. When used in a P2P MMOG, this means that a single player’s AOI would be limited to those players inhabiting its current region. This would result in a lower networking overhead in comparison to if interest management was not used, thereby increasing the viability of a P2P MMOG.
  6. Describe the concept of Potentially Visible Sets (PVS) and discuss how and why different flavours of PVS might be used in network games.
    A Potentially Visible Set (PVS) is a collection of objects in a game world which may be visible from a given location in that world.

    • The primary purpose of PVS is to compute the polygons which are visible from a player or object’s point of view. However, to maximise processing efficiency, it is often better to calculate PVS on a per-object as
    • opposed to per-polygon basis

    PVS may be used in networked games as a component of Area of Interest (AOI) management. As a result of the type of PVS used, latency may occur.

    Many games use a conservative PVS approximation, which is the fastest method of PVS, however, it may result in excess rendering and lower frame-rates which can cause latency.

    • In slow-paced networked games such as turn-based strategy games, this may not be a huge issue. However in the case of fast-paced, graphically-intensive games such as first-person shooters, a more exact PVS
    • algorithm may be used in order to lower rendering times and maximise frame-rates, and in turn, decreasing latency.
  7. Compare and contrast two different transport layer protocols that can be used in network game communications.
    • Two transport layer protocols that can be used in gaming are UDP and TCP. UDP is a connectionless, unidirectional protocol that is designed for lightweight application messaging. When an error is found a UDP datagram is simply discarded. TCP is connection orientated, reliable, bidirectional and guarantees delivery of data, sequencing, no duplicated data and resending lost data. UDP is the better of the two protocols (for gaming) as it is faster and also the fact it is transaction orientated so there is no need
    • for a three-way handshake to begin communications. The lack of retransmission
    • delays also make it suitable for online gaming, unlike TCP where lost data has
    • to be resent which in turn slows down the gameplay as users wait for the lost packets to be resent.
  8. Discuss how network latency can impact on network game communications and describe
    methods to overcome this problem.
    • In networked gaming, latency refers to the amount of time taken for game states to be transmitted. It is arguably the most important aspect of a game network as it is the difference in the time between when a
    • user acts and when they see the result.

    A high level of latency can make a game almost unplayable due to a lack of correlation between input and response.

    • There are four sources of latency in a networked game:
    • - frame-rate latency
    • - network protocol latency
    • - transmission latency
    • - processing latency

    • It is not possible to completely eradicate these sources,
    • however they can be reduced.
    • Frame-rate latency can be reduced with improvements to frame-rates.
    • Network protocol latency can be reduced by switching to a lower latency protocol e.g.
    • TCP vs UDP.
    • Processing latency can be reduced by upgrading server hardware.
    • Transmission latency can be reduced by upgrading the physical network e.g. dial-up to DSL.
Card Set
Game networking
Networking and games
Show Answers