Will changing gain affect frequency response?

In addition to how changing the gain of a preamp can change sound due to nonlinearity of the gain control, another thing to consider is where you achieve the gain.  For instance, I can change the gain on my phono preamp as well as on my main preamp.  I have tried it both ways and using the main preamp for increased gain sounds better than increasing the gain in my phono stage.

This discussion poses a question in regards to the Preamp-Amp gain/volume control. Should I run the amps [Mcintosh MC 60s] full out with the gain controls available on each amp and use a lesser level of volume on my preamp,[Zesto Leto tube pre]? Does trimming the gain on the amps give me less headroom when listening at higher volumes?

The output impedance of the Preamp is relatively low 150-250 Ohms to the input impedance of the amps which is 500K ohms. The preamp has two settings for gain, set internally, of 3 db and 12 db and came to me set at 3 db. I have very efficient speakers so the headroom of the amp and pre is sufficient either way. I rarely have the volume on the pre at 12, more at 10-11 max with the gain on the amps maxed out.

I ask this because I do fight brightness a bit and would like to optimally control the tone of the system towards a bit more mellow direction. 

The preamp has two settings for gain, set internally, of 3 db and 12 db and came to me set at 3 db.

Back to my first post, which everyone overlooked, how is this reduction achieved? Is it via the feedback network??
If so, this can influence the tonal quality, of where these gain setting positions are used, and they have nothing to do with the vc knob being high or low physical position for the same given volume.

Cheers George

I second what rlawry says above about the non-linearity of the particular gain control, versus that of the volume control itself in particular.

Almost as an aside, Steve Deckert of Decware advocates introducing an extra gain stage control inserted after the source, in addition to the one at your amp (ZBOX). This allows you to do some "gain riding" as he calls it, letting you either turn up the gain going out of your source while also turning the gain down going into your amp....or vice versa. I haven’t tried it that way (yet), but he says that increasing the source output gain while lowering the amp input gain, definitely changes the overall character of the sound...in the right adjustment zone, he says the sound in that instance is much more dynamic. Presumably you can take this idea too far and it may begin to deteriorate the sound in some other regards the further your adjustments go, but his point is that you have a choice of what sounds best to you. If this arrangement produces that result, then perhaps it is not hard at all to accept that you are hearing, ghosthouse, a slight difference in tonal response in your set up (and that Al would be correct by way of what he has already pointed out, as usual, but I defer to Al and others on most matters ee).

Here is the ZBOX link which has a description and a link to the owner’s manual: www.decware.com/newsite/zbox.htm Steve makes mention of both impedance and frequency response.

Claims are a bit out there.
How can a tube buffer with 1000ohms output impedance, be a better impedance match than a 100ohms or even less output impedance source with it's own VC, into say the low 10kohm load  of one of the Pass Labs poweramps? 
And this 1000ohms output impedance will be even higher at lf as it's bound to be capacitor coupled!

Cheers George