Evidence on e-prime timing accuracy compared to other similar software?

[David McFarlane]

At 4/21/2009 04:06 PM Tuesday, CB wrote:
>Here's one comparative review: http://www.ncbi.nlm.nih.gov/pubmed/16955731
>
>It's a feature comparison, not addressing timing at all, but might be
>of interest to some here.

Thanks, that's the one I referred to earlier, you 
saved me a lot of trouble!  This 2006 paper by 
Christoph Stahl compares DirectRT, E-Prime, 
Inquisit, and SuperLab (2004).  Here is the 
relevant excerpt from that paper -- note especially the final clause:


"All four packages rely on DirectX technology2 to 
support stimulus presentation and response 
registration with high temporal resolution, and 
to interact with the output (e.g., video and 
sound) and input interfaces (e.g., keyboard, 
mouse, and joystick) available for the Windows 
operating system. The reviewed packages claim 
millisecond accuracy in stimulus presentation and 
response registration; supporting evidence has 
been reported for Inquisit (De Clercq, Crombez, 
Buysse, & Roeyers, 2003). It is beyond the scope 
of this review to provide a test of that claim 
for the other packages. In general, some caution 
is necessary regarding timing accuracy on a 
Windows operating system: Because it supports 
multithreading, (i.e., multiple processes running 
at the same time, sharing one central processing 
unit), perfect timing accuracy cannot be 
warranted (see also Myors, 1999). If an 
experiment program—be it one of the above 
mentioned packages or self-programmed—is to 
present a stimulus at time t, it can do so 
accurately if no other processes are running. If, 
however, another program is running in addition 
to the experiment program, it might occupy the 
central processing unit at time t and thus delay 
stimulus presentation. Yet, within the DirectX 
framework, several measures can be taken to 
minimize this and other sources of timing error 
(e.g., Forster & Forster, 2003), and it can thus 
be assumed that timing accuracy is potentially 
high for the reviewed packages (MacInnes & 
Taylor, 2001; Plant, Hammond, & Whitehouse, 2002)."


Note that the E-Prime documentation also provides 
supporting evidence for PST's claim of millisecond accuracy.

-- David McFarlane, Professional Faultfinder


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "E-Prime" group.
To post to this group, send email to e-prime at googlegroups.com
To unsubscribe from this group, send email to e-prime+unsubscribe at googlegroups.com
For more options, visit this group at http://groups.google.com/group/e-prime?hl=en
-~----------~----~----~----~------~----~------~--~---