Back when I had two VH100R's sat next to each other, one with EL34's and one with 6L6's, I didn't really expect them to sound that different. But the difference was huge. It wasn't subtle. The 6L6 one sounded more metal out of the box, and the EL34 one sounded more rock out of the box. The mids characteristics both in terms of tonal balance but also distortion character, where completely different.
It was quite an eye opener. And thus far, I'm not hearing anything in the SLO that couldn't be explained by such a physical hardware difference.
Because I know at least a little bit about the DSP behind this, I can say some things for certain. If you oversmooth the nonlinear solver, you get high frequency rolloff. If you get a filter coefficient wrong, you can filter frequencies you didn't mean to. If you use a median filter to average out data, you can blur transient response.
So when I hear a particular characteristic, like a subdued high frequency region, I cannot help but wonder if one of these types of things are at play. That's just how I think about it.
The end result for me - taking all analysis out of it - isn't much more complicated than turn up gain, turn up volume, cut mids, cut bass, boost treble and presence.... does it do what I expect it to do based on what my real amps do? If it doesn't, something is broken.
Those Laney VH100R's ... if you boost the presence on them full whack, the amp will oscillate like a bastard!!