Ian's system

[krass]

What I'd really like to hear is SAM with
(a) the Quad ESL63, Peter Walker's original concentric rings design. Or even with the later versions in the 98* series or the 2*05 or 2*12 series. Those speakers were/are allegedly substantially phase correct and overdrive protected - suggesting that the SAM effect should be minimal other than at bass frequencies;

My understanding of SAM is that it's a processing-module that applies delay in the time-domain, to compensate for the speaker, crossover & different driver designs. At a simple level I imagine a simple time-lag for one driver compared to the other, but guess its more complex and that the amount of time-delay also changes with frequency (even if the signal ends up at the same driver). I postulate this because I think Devialet can only generate a sensible SAM profile if the speaker system under test has inherent time/phase discrepancies, which I guess they measure. With the concentric Quad design, I always thought they included time-delay within the electronics design to arrive at a different signal driving each of the rings. Since there are no individual drivers affected by time-errors, I don't think it would be meaningful or even possible for Devialet to generate a Quad concentric SAM profile.

That's probably what you already said !

I would really like to know more what they do; especially whether a SAM profile can be derived from the design (on paper) itself, or whether they have to use real speaker hardware to get measurement data.