I'm not sure if the Hypex I heard at Munich was the NC500; it was enough of an embarrassment that I preferred not to give it a mention. But how can you tell what goes wrong at a show- speaker placement, room problems, who's to know?
The 'Best' way to master an album is to avoid any processing whatsoever in the mastering process itself. Warm up the lathe by running it for about 20 minutes prior to operations. Do a few test cuts to verify the stylus temperature and angle (if correct, the groove will be so quiet its noise floor will be significantly quieter than the quietest electronics). Do a test recording of level on the outside of the lacquer beyond 12". I prefer to use a good tube amplifier- they interact better with the cutterhead due to the electrical characteristics. When done, regardless of the recording source (tape or digital master file) the LP will not have needed any EQ or compression to make it happen. This will give you the most neutral recording with the lowest distortion and widest bandwidth of any format available (playing it back is a different matter). It should not go without noting that any recording from any recording process will fit within the dynamc range of an LP without compression.
This will get you the best album, which will have the longest storage life of any media on the planet. And BTW much of this is exactly how LPs are done today. As far as I know, we are the only ones using a transformerless vacuum-tube mastering system. A strength of the amp we are using is that its distortion decreases linearly to unmeasurable at lower power levels, and it happens that on any LP mastering system the cutterhead uses no more than a small fraction of the total power of the mastering amp. What that means in this case is you can't measure the distortion of the amp at the levels being asked of it to drive the cutter. FWIW, this phenomena did not exist even 2 years ago- at that time, all the mastering amps in use had a distortion character that increased below a certain minimum power (which, I might add, they have in common with all class D amps...). IOW the amps we are using can do something that class D can't- essentially no distortion at lower power levels.
The problem you are up against as I pointed out earlier is while you can certainly model the target amplifier pretty well, you can't do that and then reproduce it without significant artifact. Class D and DSP just aren't there yet. I pointed out another issue with low level distortion in my response to Blizz above. For example if you want to model an SET, you won't be able to do it until you can sort out how to make the class D amp have no distortion at all at low power- that is something that all SETs have in common, and is part of their appeal. Class D can't do that; ergo it can't be used to model certain amps.
This BTW is why even a Hypex NC500 can't be used to model an Atma-Sphere MA-3. It can't make the low/non-existent distortion at lower power levels that the MA-3 can do with ease.