The input impedance of the HX Stomp is the determining factor here, as well as its output impedance into the 1 Megohm input of the Rumble, which would be significantly lower than the output impedance of the P-bass pickups, which (approximately) doubles every octave; that is, the DC resistance of the pickups is equal to the impedance only at low frequencies, rising sharply with frequency. The HX Stomp output would be seen by the Rumble quite differently from the P-bass pickups. For an apples to apples comparison, a direct box with two equivalent outputs, one into each device, would level the playing field. Even though the direct box outputs wouldn't be the same as the output of the pickups themselves, it would at least be the same for each signal path.
I write this not to be a stickler, but because you clearly have a scientific interest in this and have gone to some length to seek an answer experimentally, with careful attention to detail.
@Bottom Octave,
Firstly, apologies for my terse response last night. Now that I've had a chance to sleep, I have to admit that your points raised a keen level of concern with me.
As you rightly indicate, the input impedance level could certainly have a significant impact on the modeler tones. And, further, I omitted one additional, and potentially highly significant, component from the signal chain I described. I pretty much always use a Line 6 Relay G10 wireless device both at home and live. So, not only was my bass not plugged directly into the Rumble Stage 800 when testing, it was never plugged directly into the Stomp either; but rather via the wireless.
And, though I was not trying to do a strict shoot-out between the two modelers, this was more simply to provide @Solude my subjective judgement informally comparing the quality of the common amp models the two share, I certainly do not want any glaring inconsistencies in the process to taint my results.
I am not that concerned about the effect the G10 had on the Stomp, because: 1) I have used it for a long time, both directly connected and via the wireless, and I have never had any tonal issues switching between the two, and 2) because my overall assessment was that its modeling edged out the Rumble Stage 800. However, especially in this forum, I want to make certain the Rumble was well-represented, and not impeded by the indirect instrument connection.
So, in order to test the effects the connectivity might have had on my assessment, rather than redo-ing my initial test, I ran this alternate test (
which tests both the initial review results, and illustrates differences among connectivity methods):
I setup another test project in my DAW, and created 3 instrument tracks to record each of my connection methods: 1) wireless via bypassed Stomp, 2) direct into bypassed Stomp, 3) direct into Rumble Stage 800. I used the standard "Basic Bassman Overdrive" preset which I find to be a nice balance of clean, grit, and thump, and recorded two passages using each of the connection methods. One passage was the bass riff in the intro of Rush's "
Xanadu" (
which covers 2 octaves), the other was the signature line of Stanley Clarke's "
Lopsy Lu" (
which covers 3 octaves). I did my best to record each with the same playing intensity.
Once recorded, I balanced the volumes among the tracks and compared the results. Not surprisingly, plugging the bass directly into the Rumble Stage 800 yielded the hottest recording level among the tracks. Surprisingly, the track using the G10 wireless was hotter than plugging directly into the Stomp.
However, once the track volume levels were balanced, there was not a significant effect to the modeling results. Certainly the guitar signal when directly connected was "cleaner", kinda like the treble or mids had been tweaked just a bit for more upper-end clarity; but not a significant change to the overall amp model tone.
Again, if one were strictly judging the accuracy of the models, or the responsiveness of the sag/bias/etc., then a strict adherence to connectivity methods would be a necessity. In the case of this general comparison, and given these testing results, I think my initial findings stand.
However, none of this is in any way intended to impugn the veracity and importance of your observations, and I appreciate that you raised them.
Hope this helps.