okay, here it is with the same eq settings as previously used to set it as flat as possible, with the only difference being the gain settings. the chart below from bottom to top, is:
orange: gain=9:00
turquoise: gain= 10:30
violet: gain=12:00
yellow:gain=1:30
blue: gain=9:00
...as you can see, it doesn't change the frequency response curve much at all (starts rolling off @ 50hz), but it's mostly its amplitude. however, when you start clipping it into distortion with a bass, you tend to notice the mids and highs a little more since that's where you hear the 'grit' the most.
and no, it didn't make any noticeable difference with the gain button in or out other than to change the level.
A few additions to John's comments about tone vs Gain setting:
You have to be careful when testing, to make sure that you're making an apples-to-apples comparison. For example, I used a narrow, 30dB vertical scale on my graphs, in order to make any variations in frequency response more visible. John used a wider 70dB scale. This is more in line with what most amp manufacturers post, and gives you a much flatter line, for the same actual frequency response.
Mine:
Exact same response curve, but using the wider scale:
If you look at the legends on the two graphs, you can see that one line on the upper one, is worth 5dB. On the lower graph, each line is 10dB: twice as wide. Remember this, the next time you're looking at a manufacturer's brochure. This is why every kind of gear from mics, to speakers, to amps, tends to show a much better frequency response on paper, than you experience in person.
Another difference to consider, is that I use a 20-20K horizontal scale, whereas John's graphs go all the way down to 10Hz. That makes it look as though his amp has more roll-off at the low end.
Also, signal levels matter a lot. I have a fairly high-output bass, with EMG P5's in it. With my test tones level-matched to my bass, I got a flatter treble response from the the boost switch (ie; less treble rolloff), even with the Volume backed off to retain the same level. When I lowered my test tone levels, that effect wasn't visible. Similarly, I ran a set of curves with the signal too hot (as though I was using a boost pedal in front of the amp), and got a whoop-ass load of extra upper mids and treble. All extra harmonics from clipping, of course...
Last but not least, John only took the Gain up as far as 1:30. There's a distinct change in the graphs as you get up around 3:00 o'clock. If you keep your levels below the clipping point, you actually get a fairly hefty treble rolloff- I think this was illustrated in the first set of graphs that I posted.
