Another dumb question - I've heard about the triode/pentode switch, but exactly what does it change? Tone, power/loudness, or both?
Look at the
6L6GC data sheet. The top graph on page 4 shows the normal pentode-mode curves for this tube. Now scroll up to the graph on page 3. These are the triode curves for the same tube.
The shape difference should say that the tube reacts very differently, depending on what mode is used.
Look at the curves for Ec1 = 0v on both sets of curves.
If you pick almost any plate voltage, the pentode mode has a higher corresponding current. So you might conclude that pentode mode has the potential for a bigger current output.
If you also look at the shape of the pentode 0v gridline, it rises pretty steeply at a fairly low plate voltage, before bending and running roughly horizontal. The bend, or knee, of the curve occurs at a relatively low plate voltage, and you might say allows the pentode to have a wider variation of plate voltage. This becomes noticeable if you actually draw a loadline on the curves; the pentode can swing its plate much closer to 0v than a triode can.
So pentode mode can possibly create bigger output current and voltage swings, which attacks the equation for power 2 ways. We can conclude that most of the time, pentode mode will output more power than the same tube in triode mode.
Look also at the slope of the gridlines on those graphs. If the lines were perfectly horizontal, that would indicate infinite resistance; voltage could change to any value, and current would stay unchanged. Conversely, a vertical line would indicate zero resistance. The slope of the gridline at the operating point corresponds to the internal resistance of the tube. Pentode mode lines are closer to horizontal, indicating a high internal resistance. Triode mode lines are steeper, indicating a much lower internal resistance. Under some circumstances, the lower output impedance might be a useful feature.
One thing that's harder to explain/justify without drawing some diagrams is that triode mode distortion will generally be lower. We shouldn't just say "triodes make even harmonic distortion, and pentodes make odd" because 6L6's and some other beam power tubes produce more even harmonic distortion than true pentodes. Either way, even harmonic distortion gets cancelled in a push-pull OT if everything is balanced.
Distortion might make the sound seem "brighter" or more present, even before it is obviously distortion. So the lower distortion, and that most will be cancelled in the OT (until the grid is driven positive), so triode mode will likely sound "darker" and cleaner.