Thanks, this is a little less vague. In general, from my personal perception, loudspeakers with narrow directivity patterns have been described by others to have a narrow sweet spot (extreme examples include planar dipolar speakers like Sanders and King Sound, also highly directive designs like JansZen), using descriptive terms like "vise-like" or "laser-thin" in terms of listening position. On the other hand, very widely directive speakers are often described as having a wide sweet spot, i.e. Roy Allison designs and innumerable others.
It increasingly seems to me that that this discussion, which has shifted from the original "directivity affects listener perception and other measurements beyond on-axis frequency response," may have broken down in multiple ways:
1. What should discussion of directivity mean when it comes to "controlled" or "constant," i.e. continuously changing or relatively flat DI, and if so, over what bandwidths? I had tried to introduce some discussion under point C here:
https://www.audiosciencereview.com/...-on-readings-of-lokki-bech-toole-et-al.27540/
2. For off-center "sweet spot" discussions, what are the parameters used? I believe that Allison was focussed primarily on timbre. In the first paper cited by
@markus , they seemed to have focussed on "wide-band signal" and image fusion. The second paper cited by
@markus appeared to conflate multiple factors: "Some of the parameters considered for comparison were: sound localization, apparent source width ~ASW!, coloration, and loudness." In my opinion, "sweet spot" discussions may increasingly vary between "realistic timbre" vs "realistic imaging" at increasingly off-center positions, perhaps related to directivity, but research may help define this further.
@markus seems to have declined to comment further on the papers that he himself cited.
@markus , please refrain from citing papers if you do not discuss them further.
3. For central "sweet spot" discussions, again, what are the parameters used, and how might directivity affect that? See again here:
https://www.tonmeister.ca/wordpress/2015/06/30/bo-tech-what-is-beam-width-control/. I believe that Geoff Martin's perceptions are compatible with those derived from Tapio Lokki's papers that I very crudely attempted to summarize and conflate above (#1).