It wasn't a guess My guess is that I'm not making my point very well, so that's my fault. I completely understand and agree that the "interaction" is captured in the DI profile. That isn't the issue. The issue is trying to replicate that with a SS power amp. With the with the kemper, and any SS power amp, you are essentially reamping a profile of a tube amp rig through a SS power amp and guitars speakers. Since SS and Tube amps do not react in the same manner to speaker loads and impedance changes, the feel(and/or "compression" as stated by the OP) will not be the same. It is impossible to replicate that in this scenario. Simply amplifying that interaction is not the same.
I personally don't mind this sense of extra compression in the profiles. I find that it is present to varying degrees in all digital amps/modelers. Actually the Kemper does a great job of capturing a lot of nuances that some modelers miss. Assuming the profile is well done, the KPA is great.
Mr Kemper is right. You are wrong. Resonances in a load are a function of frequency, not voltage. This means you can profile this resonance (interaction) provided you’re using the same speaker and the same gain and it theoretically should be able to sound exactly the same. You can profile these resonant points as said, and then the class D power amp can just replicate this. The maths works out.
When you make a direct profile properly you connect both the speaker and the amp to the DI box, and the kemper monitors the change in voltage as you play and records these resonant points. Now I’m not arguing how well it does this, just that mathematically it can be done.