The topic has already been discussed quite extensively, but there are a few aspects to consider as opinions (not hard facts).
Obviously, the room and the placement within the room contribute to the overall sound. Think about the angles of reflections relative to each wall. Therefore, there are optimal distances for utilizing the walls, including the rear wall for central focus. Just a reminder to experiment.
Speaker toe-in changes room reflections. Since most of us use directional speakers, changing the speaker's orientation will affect the soundstage (including focus) because the proportion of reflections from different walls changes.
When discussing the shortcomings of stereo setups, Toole pointed out that the biggest issue is the shifting of the central focus as you move around the room. From this, you might draw a reverse conclusion: if you're seated in your listening position, a center speaker might not be necessary.
And now some subjective experiences:
The only system that maintains a natural sound image everywhere in the room is one with 360-degree speakers. With all other systems, you eventually notice when you've stepped out of the sweet spot. However, it is not as good in a precise listening spot as some other speakers.
The second-best option in terms of spatial performance is speakers with a wide directional profile for high frequencies.
Strangely, both of these speaker types sound best when placed relatively far from the back wall. The central focus is also better in this setup, although it might seem logical to assume the opposite—that proximity to the back wall would provide a stronger central focus. So, there has to be an optimum distance: not too far for a reasonable reflected sound level and not too close for a relatively spacious reflection result.
On the opposite end are strange issues with AVR upmixing. If your head is fixed in the position where the setup microphone was placed, everything sounds perfect. But just 20 cm to side from that spot, the sound becomes quite ordinary. While the sound fills the room evenly, it loses its sparkle and spatial information. This effect doesn’t occur with actual multi-channel audio, such as in movies. It seems that the human brain is still a better upmixer than today’s AVRs. Toole also rightly argues in his book that once audio is recorded, it can no longer be broken down into its original components for mixing differently without losing quality.
Perhaps it’s just a preconception, but using two speakers as the source always creates a straight line, and even when you move away from the sweet spot, the speakers remain aligned. The listener is simply off-axis to the soundstage. In a multi-channel system, the situation becomes more complex. The result is neither 2D nor 3D sound but a general sound mix coming from all directions. Room reflections and other factors calculated for multi-channel systems and directed to the respective speakers still rely on the listener’s position in the sweet spot, but the actual listening position has shifted, and the expected reflections, sound timing, etc., are no longer appropriate for your brain.
This entire text is an unproven hypothesis, but perhaps some points may seem interesting and help to achieve a better listening experience.
