|
|
|
Forum Newbie
Group: Forum Members
Last Login: 6/14/2008 7:42:58 AM
Posts: 2,
Visits: 7
|
|
| My experiment is a change blindness study using gradual change paradigm. Subjects have to press bar bottom when they see gradual change occour. I've used this one for many months without any problem, but now it stops without any E-prime message! I have to get "task Manager procedure" (ctrl+alt+canc) to esc from E-prime 2.0 (i've successfully installed it yet). When it doesn't stop, "WARNING: Clock.SystemTimeDrift exceeds threshold value. Data from this run may not be usable. Contact PST tech support to determine diagnosis of this clock drift issue" sometimes appears at the end of experiment. can you help me about this?
|
|
|
|
|
Forum Guru
Group: Forum Members
Last Login: 2 days ago @ 5:23:48 PM
Posts: 137,
Visits: 380
|
|
Lisa,
This has been discussed before in other threads going back to Dec 2006, see
support.pstnet.com/forum/Topic111-12-1.aspx
support.pstnet.com/forum/Topic559-7-1.aspx
support.pstnet.com/forum/Topic735-12-1.aspx
support.pstnet.com/forum/Topic819-12-1.aspx
The documentation you want is in a PST Knowledge Base article at www.pstnet.com/e-prime/support/kb.asp?TopicID=2989 (you will have to login to view this).
In particular, it says there, "If the RefreshClockTest experiment or error indication after running an E-Prime experiment indicate possible drift values, the machine can be checked to see if drift is coming from the clock on the motherboard or if the clock that is keeping track of the system date and time. It is possible that the SystemTimeDrift checks put in place by E-Prime and RefreshClockTest will report a number of false positives." [emphasis added]
That is, Clock.SystemTimeDrift only compares the high-resolution E-Prime clock to the computer's slower time-of-day clock, and E-Prime has no way to know which of those clocks is more accurate than the other. So if the the time-of-day clock is poor (i.e., almost always), then you will still get a Clock.SystemTimeDrift error even though the high-resolution clock is perfectly good, i.e., you get a false positive. So the error message is of dubious utility, unless you have a good engineer on site who knows how to follow it up (you do have an engineer for your lab, don't you?).
-- David McFarlane, Professional Faultfinder
|
|
|
|
