This surprised me as I wasn't expecting to spot the real amp as often as I did. As I went through the test I was actually identifying the real amp by a process of elimination, spotting obvious traits of modelled sounds. That's probably because I'm more familiar with modelled amps that real ones. I was definitely trying to spot the real amp rather than saying which one I preferred. I think that I could only honestly judge the 'best' sound if I'd been told in advance that the different sounds were either all real or all modelled. Similar reason why I like my JMP-1 and Vox ToneLab pre-amp / modellers - they have a real valves in them so they must be good, right?