[hpsdr] Frequency Tuning Speed

[Philip Covington]

On 6/6/07, n3wt at comcast.net <n3wt at comcast.net> wrote:
> ***** High Performance Software Defined Radio Discussion List *****
>
>
>  -------------- Original message ----------------------
> From: "Philip Covington" <p.covington at gmail.com>
> > ***** High Performance Software Defined Radio Discussion List *****
> >
> > On 6/6/07, john_eckert at agilent.com <john_eckert at agilent.com> wrote:
> > > ***** High Performance Software Defined Radio Discussion List *****
> > >
> > >
> > > I've been using AJO for a few days now and I'm quite satisfied
> > > with it, except for the VFO tuning speed.  It is way too slow. I
> > > doubt that the radio can be used in a contest because of this.
> > >
> > > Does anyone know if this has already been optimized or is there
> > > hope that it can be remedied.
> > >
> > > TNX,
> > > John
> >
> > I am sure it has not been optimized.  It just will take someone with
> > the time to spare to look at it.
> >
> > 73 Phil N8VB
> > _______________________________________________
>
> I noticed the same slow tuning speed and slow turn around times .  For CW if using break-in,  the first dit is missed on CW.   Unitl the system is optimized as Phil pointed out,  I  went back to  the standard parallel port for control,  but continue to use Janus/Ozy  as the I/Q source.  Works fine this way.
>
> 73  John, N3WT

I haven't looked yet at how tuning has been implemented currently, but
I suspect that the tuning word is being sent in-band with the I/Q
data.  If it is, this limits the tuning speed since the data can only
be sent at the I/Q sampling rate and will have more latency for 48kHz
than it would have for 192kHz (for example).

In QS1R I send all control data on EP0 out of band of the I/Q data.
If this isn't how it is currently done with PowerSDR, maybe someone
can implement it.  Tuning should then be virtually instantaneous.

73 Phil N8VB