Once I have one speaker dialled in, at least in terms of its linearity and room interactions, I then work on the other speaker, starting by placing it in a point that’s symmetrical to the listening seat. That may or may not be symmetrical to the room – and few rooms are genuinely symmetrical – but it will create an isosceles triangle with the other speaker and the listening seat. Equal arrival time at your ears is absolutely everything, if you want the two speakers to act together to create a single, coherent, musical whole. Room measurements simply aren’t accurate enough for that. In the best case, they’re a guide, but the actual listening distance can only be measured from the listening seat.
I’m currently using a multi-axis laser device (the Huepar Pro ZK04CG). It’s not particular expensive but it lays a perpendicular line in all three axes – front-to-back, side-to-side and vertically. I also really like that the line it throws is really visible and narrow, which makes working with it a whole lot easier – which is as we said earlier, a real contribution to successful results. If I lay a laser line along the front of the speakers, using for instance the front, inner spikes, if I place the laser intersection at the exact mid-point between the speakers, then I get a perpendicular line from that point which should bisect the listening seat. Suddenly, the way that a small movement in speaker position impacts the listening distance becomes really apparent. The easier it is to envisage theses things, the easier it is to work with and optimize them. The second speaker will always need to be moved from its initial ‘symmetrical’ placement, while the attitude of both speakers will almost certainly need to alter too. Each shift in angle or position impacts the listening distance, often by a tiny, tiny amount. It’s compensating for those shifts, one speaker to the other that really takes the time.
RG. I think that one thing that is important to understand is the role that actual numerical measurement plays in the results you achieve. A lot of people have seen the measurements and especially the angular measurements that you have arrived at in different systems and have dismissed them as impossible to replicate or maintain with any consistency. They suggest that adjusting the position or angle of a speaker to three decimal places is impossible. But adjusting a speaker to three decimal places and measuring it to three decimal places are very different things.
ST. Exactly. The numbers are a reference. They tell me where I am relative to where I was. The numbers themselves are not so important – at least not to that degree off accuracy. Instead, they tend to reflect just how small the movements I’m making really are – especially towards the end of the process. The numbers are used IN-PROCESS, not to replicate a final position. Sure, they’ll get you back into the ballpark if a speaker gets moved, but no closer than that. For example, if I have the Digipass placed on top of a speaker, even a speaker with a flat, machined top-plate, like a Stenheim, moving the location of the level will alter the reading. One of the things that I love about the Digipass is that, once it’s placed, I don’t have to touch it, so the consistency of its read out is much better than a level that I keep switching or moving. In angular terms, it acts as a reference for where I’m going, how far I’ve gone and, where I might need to get back to. Believe me, getting back to the same point – exactly – is hard.
