An attempt to clarify a few things:
- SOv1 did not calculate the cubic air volume (affected by temperature and humidity I might add) of the room and how sound at all frequencies is energizing the space in the room. It only calculated length, width, and height modes and offered suggested cuts for frequencies below 200Hz based on the position of the given speaker in the given room dimensions - which were explicitly something that still needed to be further adjusted and verified by ear. Not every SOv1 calculation could or would be very accurate, as admitted by Linn back then.
- In addition to calculating the cubic air volume of the space based on the provided input, and in general performing a much more advanced calculation of the given room dimensions, materials coefficients, etc. (which is why it moved to the cloud - SOV1 could be done on standard computing hardware we all use whereas the maths happening with SOV2 are being done using GPU-based processing in the cloud on much more powerful machines so that you don't have to sit there and wait for hours - just that should tell you something about the sophistication of V2 vs V1) SOV2 uses a more advanced modeling system with respect to the speaker being used in order to determine the shape/dimension of the speaker, how it is interacting with the room surroundings, and what affect that has on both the speakers low frequency output and it's time-of-flight (AKA group delay) for all frequencies. This is audible with most speakers (even passive) once a proper TuneDem has been completed, by just ensuring the proper speaker has been selected in Konfig, without applying SO. The profile(s) that Linn have for the speakers in their database specifically take into account where the center point of the midrange driver is on the loudspeaker, the cabinet dimensions, and the resonant frequency of the port/radiator system used (if any). So @akamatsu is correct about the fact that SOv2, is indeed based on the characteristics of the speaker being used and it's assumed/presumed response and interaction within the given environment.
- Linn are (or at least have been in the past) very adamant that NONE of the above is a "magic bullet" and that SOV2, just like V1, would only give proper results when the speakers are TuneDem'd properly into the room. The idea is, you need to get the speaker as free of the room influences as possible before you do anything further in software. Otherwise you're just chasing your tail, and likely making things sound worse than they should. It has been my experience that this is the "Achille's heel" for getting results with SO, and as I've mentioned elsewhere, I've not personally met very many folks who are competent with this process. Again, speaking from a NA/US perspective (I think things are a bit different in the UK/EU, etc.) there are plenty of audio "gurus" and speaker setup "masters" who would not have the first clue as to how to properly TuneDem a pair of speakers, let alone know what to do with something like SOv2. I know because many of them have flat out told me to my face that the speaker setup process that Linn uses is wrong, and is an incorrect way to go about tuning a speaker into a room. So my assumption is, whenever I read or hear something to the effect of "we tried it and the customer didn't like it" or "I like the sound with it off better than with it on", that the individuals involved in the setup don't really know what they're doing.