refresh rate USB

David McFarlane mcfarla9 at msu.edu
Fri Jul 8 14:32:09 UTC 2011


Tobias,

I checked through my FAQ, and found the following foundational 
citation:  Segalowitz & Graves (1990), Beh Res Meth Instr Comp 
22:283-289.  It's rather outdated now, but it sets the stage, and if 
you search for later papers that cite this one then you will learn a lot.

At 7/8/2011 07:14 AM Friday, you wrote:
>In fact, we do have a SRBox and I so far refused to take it because
>the buttons are so hard to press.

When we use a SRBox for tapping time tasks, we do *not* use the 
buttons on the SRBox itself.  If you look at the documention for the 
SRBox, you will find that you can connect any simple button box you 
like through a connector inside the SRBox.  By that means, you can 
use buttons that suit your tastes, but still use the electronics of 
the SRBox to get the responses from your buttons into E-Prime, it is 
really quite trivial.  As I have mentioned in several other threads 
here and/or on the PST Forum, I just grab some parts from Radio 
Shack, or a local electronics supplier, or an online source (e.g., 
Newark, or Allied) -- we like to use high-quality easy-push "clicky" 
pushbuttons -- build a box and an adapter cable, plug it all in 
through the SRBox, and E-Prime never knows the difference.  I don't 
know why everybody doesn't do this.

>ANother thing is, I just learned that a parallel port would have best
>timing characteristics. Serial ports are still quite slow. Maybe we
>will just use single buttons sending a signal via individual pins to
>the parallel port. I have never used it but there seems to be a way of
>using parallel ports in E-Prime (if you look at "devices").

Indeed, the parallel port or an ordinary digital I/O expansion card 
could give you near-instant latencies (microsecond range?).  The down 
side, as you will find by reading the articles on the PST Knowledge 
Base, or the book "Parallel Port Complete", or threads on the Group 
or the Forum, using the parallel port does take some finesse.  In the 
old days before Windows XP we could add a digital I/O card to make 
the programming simpler, but now that requires specialized drivers 
and libraries simply to execute a single I/O command.

That said, if you think through the specifications of the SRBox, you 
will see that with proper configuration it achieves sub-millisecond 
latency, with minimal programming finesse.  That does not seem so bad to me.

Hmm, seems to me that we had much of this discussion in an earlier 
thread, but I do not have a reference handy.  But checking through my 
FAQ, I found references to a couple of pertinent links:  Empirisoft 
promises a keyboard with sub-millisecond latency, see 
http://www.empirisoft.com/Hardware.aspx?index=2 .  And Ergodex offers 
a way to build an arbitrary keyboard layout, although its timing 
characteristics may be no better than an ordinary keyboard, see 
www.ergodex.com .

Aha, using "Ergodex" as a search term on the Group, I found the 
earlier discussion, here is the 
link: 
http://groups.google.com/group/e-prime/browse_thread/thread/d42447cfc9a061af 
.  Among other things, this thread contains more citations to studies 
on timing performance of various input devices.  Have fun!

-- 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.



More information about the Eprime mailing list