It depends on the P-P amp, but most of them have increasing distortion as power is decreased below a certain point, usually about 5% or less of full power.
So to compare against a SET amp, you want one that is maybe about 1/2 the power of the SET. The reason is so that you will operate the amp above that distortion zone (this results in better detail) and assuming that the SET in question is being operated correctly, at no time will you be demanding more than about 20-25% of its full power. This means that the P-P amp will have about a 3 db advantage over the SET. It is a fact that P-P amps make less distortion at higher power levels, so they have a greater percentage of usable power compared to an SET.
Now if you are willing to ignore the 20-25% rule I stated earlier, then the amplifier power of the P-P amp can be increased. If this is the case, you will find out rapidly that SETs don't do so well at higher power. They make a lot of distortion. Initially the distortion only exists on transients, which means that an SET will likely sound more **dynamic** than the P-P amp. This is because the human ear uses higher ordered harmonics to calculate how loud a sound is (loudness cue) and those harmonics initially will be in the transients where more power is required.
This, BTW, is why SETs are often thought to be very 'dynamic' for their low power. But it can now be seen that the word 'dynamic' can be replaced by the word 'distorted' without changing the meaning of any conversation around SET dynamics. This is not true at lower power levels!! That is why I recommend the 20% rule.
So far I've not mentioned the issue of bandwidth, but in a nutshell the higher the power of the SET, the faster the bandwidth gets pinched. Practically speaking, 7-8 watts is the limit before full 20-20KHz bandwidth becomes very difficult, to say the least. This is one of the reasons the lower power triodes like the 2A3 and the type 45 tend to sound better- its not due to the tube so much as the output transformer's size.
The bottom line here is that if you need power, the P-P amp is going to be more neutral. This is why traditionally SETs are coupled with high efficiency speakers.
Sorry if this was a bit more than you might have expected.
So to compare against a SET amp, you want one that is maybe about 1/2 the power of the SET. The reason is so that you will operate the amp above that distortion zone (this results in better detail) and assuming that the SET in question is being operated correctly, at no time will you be demanding more than about 20-25% of its full power. This means that the P-P amp will have about a 3 db advantage over the SET. It is a fact that P-P amps make less distortion at higher power levels, so they have a greater percentage of usable power compared to an SET.
Now if you are willing to ignore the 20-25% rule I stated earlier, then the amplifier power of the P-P amp can be increased. If this is the case, you will find out rapidly that SETs don't do so well at higher power. They make a lot of distortion. Initially the distortion only exists on transients, which means that an SET will likely sound more **dynamic** than the P-P amp. This is because the human ear uses higher ordered harmonics to calculate how loud a sound is (loudness cue) and those harmonics initially will be in the transients where more power is required.
This, BTW, is why SETs are often thought to be very 'dynamic' for their low power. But it can now be seen that the word 'dynamic' can be replaced by the word 'distorted' without changing the meaning of any conversation around SET dynamics. This is not true at lower power levels!! That is why I recommend the 20% rule.
So far I've not mentioned the issue of bandwidth, but in a nutshell the higher the power of the SET, the faster the bandwidth gets pinched. Practically speaking, 7-8 watts is the limit before full 20-20KHz bandwidth becomes very difficult, to say the least. This is one of the reasons the lower power triodes like the 2A3 and the type 45 tend to sound better- its not due to the tube so much as the output transformer's size.
The bottom line here is that if you need power, the P-P amp is going to be more neutral. This is why traditionally SETs are coupled with high efficiency speakers.
Sorry if this was a bit more than you might have expected.