User`s guide

E-Prime User’s Guide
Chapter 3: Critical Timing
Page 101
millisecond) are unlikely to result in a missed detection of a refresh. In contrast, if duration is set
to the expected time of the refresh, in half of the cases the refresh would occur before starting to
look for it. Think of the 10ms rule as providing robustness to account for the typical variability
between cards of the same model (0.1ms), E-Prime variability (0.5ms), and operating system
timing variability (depends on processor speed and configuration, see missed tick estimation in
Appendix A). It is important to note that regardless what anomalies occur during an experiment,
E-Prime can record the stimulus onset times and delays necessary to compute and report the
actual display times that the subject witnessed.
The next table shows the actual display durations that should be specified to obtain 1, 3, or 6
refresh cycles on a 60.15Hz monitor. The Specified Duration is the value that would be entered
into the Duration property of the E-Prime stimulus presentation objects (e.g., TextDisplay,
ImageDisplay, Slide).
Refresh Rate 60.15Hz Target Duration
Number of Refreshes 1 3 6
Target/Actual times 16.63 49.89 99.78
Specified Duration 7 40 90
If Event timing mode is in use, the actual duration for which the subjects sees a stimulus is equal
to the Duration specified for the stimulus object plus the OnsetDelay of the next object that alters
the visual location of the first stimulus. For example, if the duration for the first stimulus was set
to 90ms, and the next stimulus had an onset delay of 10ms, the true duration of the first stimulus
would be 100ms.
3.4.2.2 E-Prime takes time to perform housekeeping
functions
If the experiment must account for every millisecond, the user must be aware of when E-Prime is
doing housekeeping operations and when data is being logged. Both actions take small amounts
of time and may trigger operating system events, such as memory reorganization. If PreRelease
times are set properly, no significant delays should result in the experiment. However, it is useful
to be aware of when extra functions are occurring if running experiments with very high-speed
demands (e.g., long continuous streams of 50ms displays).
In general, E-Prime consumes about 10ms on a 266MHz computer when selecting a stimulus,
setting up the procedure, and logging the data. The selection of the stimulus occurs before the
procedure is performed, and the logging of the data occurs at the end of the procedure (see
Figure 13). Typically, when a PreRelease time of at least 100ms is used, the E-Prime setup time
will be absorbed and not influence what the subject sees (see section 3.3.1.1 Technique 1: Using
Prerelease, for special considerations related to potential interactions between the use of
PreRelease and end of trial data logging). In general, most logging events take minimal time (2-
5ms). However, when disk-writes occur, this could increase to as much as 20ms.