Depends on how far the parameter changes go. It sounds like the equivalent of selecting specific values for components to me, which is not really deviating from the 'real world' design. Most designers are going to select passive components like resistors and capacitors that can easily vary by as much as 20% in their measured value from the schematic 'design' value.
When you go to buy an amp, in your specific real world amp, all those fuzzy 5,10,20 percent windows all resolve to some specific value for the one you have and it will NOT be the same as the value set printed on the schematic or in the prototypes the manufacturer built. The result is an incredible amount of variation some audible, some not. This is why specific tube amps of the same design all have the same character since schematically they are the same within tolerance but there are the occasional magical mojo specimens where all the stars aligned on a specific set of measured values that sound extra good.
While I am sure FAS tries to procure prime specimens with intrinsic mojo, I have zero problems with them going through and hand selecting specific values which give optimal performance which may vary slightly from their on site specimen and IMHO, this in no way diminishes the claim 'this is a model of amp X'. Now if they were to do something like change the feedback topology of the power amp... Yes, this is a modified amp, it should be named to reflect that so people's expectations are aligned but that doesn't sound like what Cliff is saying.