They are single-sided up to 2V p-p, unless Rupert has changed his design philosophy. -Same as the 9098i, 5534s can be made to sound FABULOUS, and operate in single-sided mode for ALL small signals, reverting to class AB for larger peaks only. I must confess, I don't recall them saying it was discrete anywhere... If they do say "single sided" and they're using the 9098i trick, I bet it's a marketing decision, and one that I'm VERY uncomfortable with, so we probably agree on that, but discrete it very plainly isn't. -Perhaps the marketing guys insist on saying single-sided and leave everyone's minds to fill in the 'discrete' part?
Quad did a very similar same thing with their current-dumping topology. Class A where it matters (delicate, small microdynamics) and Class AB where there's a distinct advantage (blood-and-thunder headroom).
Yeah, the real reason that I 'jumped' so hard on that 'sum of cheap parts' bit was not to get on the poster's case, as to try and balance the issue.
Posters who read this are not all balanced, reasonable people, and -picking a gearslutz-mentality example- may find this thread having searched "portico circuitry" for example. -If they scan it and read stuff like "cheap parts", that's how "It's just an expensive name on an ordinary product" takes root.
Incidentally, I heard of an interesting test recently: The sales manager for Rupert Neve Designs set up two sets of output paths in Pro-Tools, one with 16 channels summed through a 'typical' console, and a second one with the smae sixteen channels summed through the porticos. Both paths were aligned using 1kHz level reference tones...
...and then the Portico paths were all REDUCED by 1dB.
The client was asked to say which sounded louder, with the meters hidden. He chose the Porticos. When they ran tone through them and showed him the meter readings, he bought sixteen.
Keith