Other Free Encyclopedias » Online Encyclopedia » Encyclopedia - Featured Articles » Contributed Topics from A-E

Dead Reckoning

game schemes position moving

Definition: Dead reckoning is commonly used to compensate for packet losses and delays in online gaming.

With the aim of providing users with high levels of interactivity in the course of frenetic fast paced games, a large number of latency hiding techniques have been devised that aim at compensating transmission delays and packet losses. In essence, much of the focus on improving real-time interactions within online multi-player game sessions is on how to reduce response times experienced by players. For timely game state updates at player consoles, dead reckoning is commonly used to compensate for packet losses and delays. In this specific context, prominent works are those presented in .

In particular, dead reckoning schemes typically aim at predicting the actual positions of moving entities during the game evolution. These schemes are based on the idea of updating entities’ trajectories rather than static positions. Thus, for each entity, the position estimation is calculated based on previous moves of the entity.

Different schemes are possible for the prediction of a future event or position. As an example, the most common approach is that of considering entities moving at a constant speed. In this case, the position of a moving entity is predicted by resorting to a first order equation: x ( t i ) = x 0 + v ( t it 0 ), where t i and t 0 represent, respectively, the actual local time and the time of the last received event, while x ( t i ), x 0 and v are vectors in a 2D or 3D space. In particular, v represents the constant speed that identifies also the trajectory of the moving entity, and x 0 represents the known position of the entity at time t 0 . It is clear that such an approach may be fully exploited to predict the position of moving entities provided that events and correlated movements may be described based on a first order equation. Otherwise, alternative prediction schemes may be exploited such as, for example, second order equations. These assume that entities move with a constant acceleration i.e., x ( t i ) = x 0 + v ( t it 0 ) + ½ a ( t it 0 ) 2 . Finally, different prediction schemes could include methods that exploit constant speed circular motion.

It is quite obvious that these dead reckoning schemes work well for game entities that perform simple (and slow) trajectories such as, for example, tank movements in a (game) military simulation. However, the problem of dead reckoning is that moves may be incorrectly predicted, especially in case of (fast moving) complex trajectories. In this case, in fact, inconsistencies may occur. In such situation, dead reckoning schemes must be devised that adopt control techniques that allow identifying possible inconsistencies between the predicted entity’s position and the actual real position of the entity.

Finally, due to the fact that dead reckoning algorithms typically encode continuous move actions within game events, it results that synchronized clocks and game event timestamps must be exploited to process such game events at their exact points in time. In fact, network latencies could introduce inaccuracies in the entities trajectories if game events would be roughly processed as soon as they are received without any control on their specific time of generation.

Deadlock [next] [back] Dead Men Tell No Tales

User Comments

Your email address will be altered so spam harvesting bots can't read it easily.
Hide my email completely instead?

Cancel or