autox
[Top] [All Lists]

Re: Autocross Timing/Scoring Software

To: "Mark J. Andy" <marka@telerama.com>,
Subject: Re: Autocross Timing/Scoring Software
From: "Jay Mitchell" <jemitchell@compuserve.com>
Date: Thu, 30 Sep 1999 10:09:24 -0500
Mark Andy wrote:

<snip>

>I'd be interested to know what system architectures and designs
folks out
>there really end up using for timing systems.

The Solo II timing system sold by Manzanita Micro a number of
years ago used a Motorola 6809. Actually, it used the entire
motherboard from a Tandy Color Computer. If memory serves, the
first generation JACircuits systems also used 6809s. The
Manzanita Micro system had custom firmware, presumably written in
Assembler, and used a standard B/W CRT monitor (or a cheap TV run
from the Tandy's RF modulator output) for display. Motorola had
an operating system (OS9) for the 6809 processors, in addition to
a C compiler and an interesting version of Basic (Basic09) in
which the source code looked suspiciously like Pascal: no line
numbers, indented procedure structure, bracket delimiters for
conditionals, etc. I've still got manuals for the C compiler and
for OS9, although I don't know why I've held on to them. ;^)  I
believe this family of micros - or possibly a newer CMOS family
with similar architecture - is still in use in some embedded
microcontroller applications, but I don't really keep up with
that field any more.

There's no problem with timing resolution with systems based on
8-bit micros. Absolute accuracy would be a function of the system
clock stability. It's possible to use relatively exotic
techniques - like housing the crystal oscillator in a heated box
inside the computer - to stabilize the system clock. Whether or
not this is justified for Solo II timing systems is another
matter.

The bottom line for a computer-based timing system is whether the
time-keeping is done by the host CPU or by external (could be
installed on an I/O bus) hardware. In the latter case, the
timekeeping hardware can easily buffer enough data so that none
would ever be lost between interrupt services. In the case of
Windows computers, I wouldn't really want to rely on the resident
CPU to keep time, as any application that tried to implement that
approach would be at the mercy of Windows' interrupt priorities.

Jay


<Prev in Thread] Current Thread [Next in Thread>