Paul Foster wrote:
> Clay Horste wrote:
> >
> > Wouldn't environmental variables such as elevation, surface, humidity etc.
> > give more accuracy error from event to event than the hardware or OS. I
> > would think that as long as precise within the same event. I just
> > wonder if we are nitpicking the wrong things here.
>
> There have been numerous times where millisecond resolution was
> necessary to determine the outcome of an autocross. Speaking of
> nitpicking there is a difference between precision and accuracy. What is
> needed is actually precision. The accuracy is of lesser consequence.
> Take the '89 Nationals as an example. Oh, never mind. That's already
> been talked to death...
'89 Nationals - I don't think I've heard that story. (groan from the
audience, I'm sure)
Yes, precision is more important than accuracy for this application, but
really both are important. If you had to switch to backup timing
equipment in mid-event, you'd prefer that both systems were accurate.
Thermal drift and humidity effects can be significant, but good design
can compensate for real-world extremes. Performance specs would
tell you what the accuracy and precision is within a temp and humidity
range, but I don't remember anyone publishing this for their stuff.
Now instead of speculating on hypothetical design shortcomings, has anyone
actually compared two or more different systems to see if there are
differences?
As an engineer who does testing for a living, I think it would be a great
challenge to come up with an objective test scheme to measure how the
various systems perform. In fact, I think I'll be asking prospective
timing vendors to explain their procedures they used to determine that
their system actually works, and what the results were. That could be very
"illuminating"!
--
Jeffrey D. Blankenship Senior Technical Consultant
jblanken@itds.com ITDS - TRIS
neon enthusiast #478 Champaign, IL, USA
|