it depends how powerful the signal you want. many tubes working in-phase superpose the signal to one another thus increasing the output power.
3 responses Add your response
The best answer I could give a "newbie" - and one that applies to any kind of gear at all, not just tube amps - is that *everything* affects the sound. That means yes, the number and type of output devices will affect the sound of an amp, but not in isolation, or in any necessarily predictable way, because that is a variable which cannot be separated from the entire circuit design and implementation.
My short advice would be to not focus so much on how any given piece of gear is executed, as on how it sounds in a given application. Some broad generalizations are commonplace (if not nearly always accurate), and you could always say, definitionally speaking, that yes, having one tube in a circuit might make it sound somehow different than if the tube were not used - and because a tube was used, you could call that presumed difference "tube sound" and not be wrong (just like you could call a sound resulting from the use of a solid-state circuit "transistor sound").
But that's not really a productive way to look at the question, not only because you'll never be able to compare the "same" piece of gear without that tube (as I say, it could not be the same circuit if a solid-state device, for instance, or maybe two tubes, was used instead, so it's an apples and oranges question, even only in theory), but also since it is impossible to characterize any such monolithic quality as a "tube sound" to begin with. Tube amps sound different from one another, as do solid-state amps, and any one amp will sound different in different applications. In any case, it's more important to consider the sound of the music, than the sound of the gear.
You may get some answers which say, in effect, the lower the number of output devices, the "purer" the sound, or that all-solid-state amplifiers tend to sound this way, while amps with tubes tend to sound that way - and there can be some limited truth to such statements taken in general - but do not make the mistake of substituting such assumptions for actual listening. *All* design and implementation choices are trade-offs with pluses and minuses, and not only will they all work better in some situations than in others (and for certain listeners more than others), but it is probably possible to make a great amplifier or a lousy amplifier using any of the almost limitless number of different approaches available to be employed by amp designers.
To address the other implication of your question, a higher number of *output* tubes (AKA power tubes), all other things being roughly equal, are most commonly utilized in order to give higher output power capability.
The more tubes that you run, the greater your chances for circuit instability ( stray capacitance ) and signal decay. The same can be said for SS devices also, but to a lesser extent. This becomes even more true as both devices age i.e. SS devices will maintain their nominal operating characeristics for a longer period of time than tubes will.
SET amplifiers take this to an extreme but are typically limited to low power output. One can pick an out of the ordinary tube ( as far as audio goes ) and work around such things. Some folks are now going to high powered RF tubes in SET circuits and finding out that they can generate much higher levels of power while still maintaining the basic characteristics and sound of low powered SET circuits.
Whether or not you like the specific sound of a product or system has more to do with anything than how that sound is achieved. That is, so long as safety and reliability are both accounted for. Sean