My point here was that the important effect downstream of the controller is
that write current is reduced in some cases. Most post 1980 drives kept
track of the track on which the heads were sitting and didn't use the signal
on Pin 2. I'm glad that SOMEONE caught that!
Since floppy dives typically reduce the write current on the inner tracks in
order to make the write precompensation done by the controller effective,
the two effects are not separable. What's interesting about this is that
they do reduce the write current irrespective of the modulation being used
to record the data. Consequently, it's important that one KNOW whether or
not a given drive uses the signal pin 2 of the cable to reduce write current
or whether it's simply a spare. One can conclude from the presence of track
counters on disk drives with the sole purpose of modifying the write current
suggests it's important. Drive makers used to charge extra for every little
option. They didn't think this feature was an option.
It's important, therefore, that reduced write current be implemented on
drives requiring it, whether from the controller or from the drive itself.
How does this impact the current topic? Well it means that effort needs to
go into the write side of the task, i.e. the computation of the bitstream
going to the write head, precisely timed at as nearly the nominal rate as
possible, with proper precompensation. It's difficult to compute the
nominal location and duration of each pulse if your data rate is not a
precise harmonic of the nominal data rate. This does not apply to reading,
but it's critical for writing, unless you believe you can deal with a
certain amount of of misalignment of the physical transitions with respect
to the theoretical ones. That is pretty likely if your write clock is fast
enough, since, as I mentioned yesterday, the typical disk controller looks
for the levels at certain times rather than the edges. Consequently the
pulse train can have considerable variation from the nominal transition
times and still be "close enough." That will require study, however, since
between every write clock frequency that's not a harrmonic of the
theoretical bit rate, there can occur pattern sensitivities that will make
the process break down under certain circumstances. This could be
simulated in software easily enough, playing a theoretical sample rate
against all possible combinations of, say 4K bits. It would take a while,
but might be worth it for its educational value. Naturally it's more
practical simply to use a common harmonic frequency and be done with it.
Dick
----- Original Message -----
From: Tony Duell <ard_at_p850ug1.demon.co.uk>
To: <classiccmp_at_classiccmp.org>
Sent: Thursday, July 06, 2000 11:41 AM
Subject: Re: Tim's own version of the Catweasel/Compaticard/whatever
> > >You may be onto something, Tim, but I'd make one observation here. The
> > >signal on pin2 of the 8" drive cable, though often driven with the
> > >1793's TG43 signal, does not turn write precomp on and off, but,
rather,
> > >reduces the write current to the heads. This reduces the amplitude of
> > the
> > >signal
> >
> >
> > In many cases it's also used to alter write precomp. Most all have some
> > precomp (Esp DD controllers) and for the TG43 case they alter the
precomp
> > to further compensate for bit shift due to the close magnetic domains.
>
> To clarify (I hope) this somewhat. The TG43 output of the controller may
> well also be used to enable write precompensation _in the controller_.
> I've never seen a floppy drive that does write precompenstation internally
> -- it always seems to be a controller issue.
>
> So in a sense, the signal on pin 2 of an 8" drive does not enable write
> precompensation. It reduces the write current. The fact that the signal
> that produces the signal on pin 2 also happens to enable precompenstation
> in the controller is a separate issue.
>
> -tony
>
>
Received on Thu Jul 06 2000 - 17:01:13 BST