[Elecraft] APF vs 10 Hz DSP, why they don't sound the same

[Eric Swartz -WA6HHQ, Elecraft]

The filter gain compensation is -post- crystal  filter. Its done inside 
the DSP.

If this is set well above what is really needed, the DSP may generate 
some distortion on really strong signals. We usually recommend +4 dB max.

73, Eric   WA6HHQ

William Carver wrote:
> The "DUAL PASSBAND" makes a pass at providing an APF type of function
> where you make the DSP passband as narrow as you choose, but there's a
> wider passband that's 20 dB down. Undoubtedly it isn't the same shape as
> APF amplitude v.s. frequency, but conceptually "a signal peak floating
> on a sea of weaker undesired signals" is similar.
>
> G3XJP implementated a DUAL PASSBAND function (that terminology hadn't
> been invented then) in his STAR transceiver, but there the amplitude of
> the wider passband can be dialed to be -20, -30, -40, .... -60 dB. It
> probably isn't too big a deal to make the "skirt attenuation" number a
> user-adjustable variable but I enjoyed being able to dial it up and down
> as QRM changed....which would ite up another programmable functio
> button.
>
> For those who suggested kicking the gain compensation up when narrower
> roofing filters are switched in, that's exactly what I did. I'm well
> aware that   IF   the roofing filter is setting the IMD capability of
> the receiver adding pre-filter gain reduces the intercept/increases IMD
> WHEN  that IMD happens to occur, and   IF  it happens to fall into the
> passband with the desired signal. I intentionally emphasized the string
> of IF..WHEN..IF.. because depending upon you penchant for living in
> pileups, or turning the rig off on big contest weeks, a 4 dB IMD
> increase  IF all of those things happened to occur may, or may not, be
> that big a deal. 
>
> A post-roofing filter gain increment would not affect intercept or IMD
> and could be a smoothly varying increase that could conceivably be
> matched to a users memory of APF or Heath QF-1 function.
>
> Bill
>   

_..._