Just look at these Stereophile measurements: https://www.stereophile.com/content/primaluna-dialogue-premium-power-amplifier-measurements it has a pretty bad load dependent frequency response.
For those who don't read the article, the response is on a simulated speaker load.
Transformers are designed to match an impedance. When they don't, frequency response curves like the those shown are the result. ALL transformer output amps will deliver a similar curve. If one were to log the current rather than the voltage on a solid state amp, a similar non-flat curve would result.
If one runs a room response graph with a tube amp and a solid state amp on the same speakers, they will overlay indicating the power to the speaker is the same.
I have a PL Prologue 5 driving 85db/W 3.6-12Ω speakers on the 8Ω tap. On some material it is scarily good. Bass is tight and detailed. It plays as loud as 100wpc solid state which is only 4db more power than the PL 36wpc. I can get 100db peaks on wide dynamic range orchestra material. My room is 16 * 33 and I sit 14' from the speakers. A tube amp cam be pushed farther than solid state, which get ugly FAST!
One visitor said "Your system is so precise." when I played him the PL.
Internally, the PL amps are gorgeous. No PCB / tube socket connection to fracture. It is my 5th tube amp dating back to Citation II, 2 x ARC and 1 custom KT-88 monster.
I've never heard a solid state amp ever come close to the sound stage presentation of a good tube amp. Sure they're colored, but so is every thing!
If you can afford it, dual mono are always the way to go.
Definitely get a refundable demo.
DISCLAIMER: I also appreciate a good solid state amp.