Hi Andy,

In the 'dark ages' of DOS it was possible to 'reprogram' the DOS Hartbeat
interrupt to much shorter times (you had to take care of the timekeeping
yourself). This timer was very accurate and with a 40MHz 386 it was
possible to have an accurate timer at 20us intervals or so ...
Also direct acces of interrupts, serial and parallel ports etc... was
'peanuts' once you figured it out.

Unfortunately the Windows designers decided that they were next in line
after God and that the simple guys writing software had no longer the right
to access all these 'goodies' ...

(Break down the fences and we don't need Gates ...)

73, Rik

At 14:20 20/03/01 -0000, you wrote:

PowerBasic (the language I use) does incorporate a "microtimer"
instruction for measuring time to 1us so something has always been
available in the underlying hardware in the PC,   via the same timer

chip that generates sounds from the PC speaker.

However, it becomes inaccurate for intervals > 55ms as the timing
function get messed up by the DOS Heartbeat interrupt at 18.2Hz.  Also
(at least in PBasic) it cannot be used to set a delay, only for
measuring an interval.    Back in the Dark Ages, I used  a FOR / NEXT
loop with a fixed count, accurately timed this using the microtimer,
then generated a calibration factor for subsequently using in the loop
to generate software timings.   Once again this was upset every 55ms,
but did work satisfactorily with a Pico Products 8 bit A/D on the
printer port for first forays into DSP.  Then I abandonned software
timing by using an external interface (a PIC) on the serial port and
ended up with something very similar to the VE2IQ I/F.

Andy  G4JNT

-----Original Message-----
From: Alberto di Bene [mailto:[email protected]]
Sent: 2001-03-20 13:36
To: [email protected]
Subject: LF: Re: WOLF - am I missing something ?

> We all have to accept that doing any sort of accurate timing operation > on a PC is futile - whether Soundacard, serial port or, even worse, by

> software timing.   Therefore some sort of external interface really
> should be used to resynchronise Tx and Rx data samples        .

Hmm, perhaps not all is lost. The Pentium II class CPUs and the Athlons

do have the so-called Performance Timer, an hardware timer

that can be queried through Windoze APIs, with resolution around 1 usec.

This however would rule out old Pentiums, 486 and the AMD K6, which do
not have that counter.

73   Alberto   I2PHD

--
The Information contained in this E-Mail and any subsequent correspondence
is private and is intended solely for the intended recipient(s).

For those other than the recipient any disclosure, copying, distribution, or any action taken or omitted to be taken in reliance on such information is

prohibited and may be unlawful.