Application of feedback Quad amps

I am wanting to know the rule of thumb for frequency
shaping a feedback path.

I have noticed different approaches to this where
some manufacturers do their best to roll off
frequency fairly quickly where other models from
the same manufacturer more gently approach this
even providing emphasis for what can only be
frequency restoration before being applied back
to the input.Was there argument in the
board room, or was the maths wrong, or does one
design require different approach- if so why?

For example look at Quad 405 vs 303 , here the 405
rolls off pretty quickly,- in later models its 4k7 and
100uf followed by 22k. Early models 10k 100uf 10k
A fairly sharp decline in frequency at the 100uf cap point.

Whereas in the 303 its 82k ( more resistance which
might explain) then a junction with resistive elements
of 2.2k variable of 4.7K and 8.2K to ground and 12uf
between the variable resistor start and ground
( fairly gentle treatment ) and 22k with emphasis
restoration of 100pf before reaching the input.

Two totally different approaches to what is at the
end of the day is feedback application. Could this
partly explain the different sounds these amplifiers
reproduce ( apologies to P Walker )
Cheers / Chris
:smash: