The Narbys <narby@centurytel.net> wrote:
> Seriously, the whole speed creep thing is entirely unverified, and until
I hear
> more numbers from radar guns, I am going to continue to dismiss it,
personally.
It just so happens that I have actual measured speed data from every single
autox run I have made since mid 1999. This is taken directly off the wheel
speed sensor, at 500 samples/second, and then averaged to 100
samples/second.
>From this, I can generate maximum, average, and median speed data, as well
as speed histograms.
I have not, as yet, compiled this data into anything one might consider
useful. Once and I while, I trot it out for Roger's amusement, but I
haven't tried condensing it into a report of any kind.
I could - but Lord n'butter, that'd be a LOT of work.
I wonder, though, if having access to such data would really _prove_
anything. Over the years, I have gotten more seat time, and the car is more
heavily modified. There had better damn well be evidence of "speed creep"
in there, or I'm wasting my time and money!
In order to prove course-design "speed creep", there would have to be some
way to factor out driver skill increases and car performance potential
increases from the data. Anybody who can think of a good way to filter this
out is better at math than me.
> Driver estimates of how fast they were going are notoriously unreliable
(and
> inflated).
Amen, Brother. I've lost track of how many times I've said "No way did I do
x" only to have the datalogger contradict me.
DG
|