"Imaging" has become a huge buzzwords in audiophile circles. It has been elevated to heights that it doesn't belong to. Go to a live concert and tell me where the imaging is. All you hear is a diffused field. No way should imaging have a large space in our vocabulary.
Yuck. I have done live stages for decades in my amateur music activities and setting up mics and figuring out the mix between the speakers for a live show is a huge part of the effort because it matters. Especially tough are the ones where people can hear the direct sound as well as mic'ed sounds. Because you don't want the imaging from the speakers they are hearing be different from the imaging they get from the direct sound. Large halls require a different analysis but it is still a thing.
If a concert has a diffused field then the audio is set up badly or you have lousy seats.
Imaging is a thing whether you believe it or not. I cannot believe someone would be believing otherwise in this day and age.
Do I loose sleep then that I did not test for "imaging" in the way you imagine, no. I have confidence that extremely well measuring and sounding speakers in mono also do extremely well in stereo. This is what I quoted from Dr. Toole and Olive research.
I agree with you on the tonal quality of what comes out of the speakers and believe the study supports you on it. But that is only half the story. To pretend it is the whole story as you are doing is very odd to me, more religion than science.
And what exactly do you want me to do anyway when it comes to imaging? How the heck would I describe and compare to two sets of speakers in this regard? How would I answer if anyone challenged to my assessment?
These are valid questions and you may be confusing what I have said with a lot of what some others have said.
Just because you are unable to measure something doesn't mean it doesn't exist or it can be dismissed as unimportant. I can understand it from an engineering perspective of saying this is the best bridge we can build given the tools we have but even then the limitations are usually documented. But ignoring it and calling it science is a misnomer. It is like flat-earthers saying at some point that the world is round doesn't matter because we have no way of measuring how round it is.
The situation here is not that different from what existed before the Klippel system (model and measurement) came about to compute a predicted room response based on measurements around a speaker. One could have simply said before then that anechoic is all that matters and room interaction is not a thing. It would be equally misguided as the model has showed. The fact that someone invented a way to model and measure doesn't immediately make it a thing while it wasn't before.
All I am saying is that we don't have a similar system yet to predict a stereo response from the measurements of a single speakers nor a good handle on what parameters of a single speaker measurement contribute to that prediction with a model. I am sure someone will invent it at some point, then we will have a complete picture of relating what we measure to what people hear.
Now, if my current speaker evaluations were a failure, maybe we would look for more ways to do things. But that is not the case. Science and research has led us correctly to identify many excellent speakers, and rule out some not so excellent ones.
This is what I don't agree with. This is a very dogmatic approach not a scientific one - that what we cannot measure (or have tools for) does not exist.
You see it as a binary - success or failure of the current system. I don't.
It is perfectly fine in science (or engineering) to set forth the limitations and caveats of a study, measurement or system (and in many cases necessary to do so for validity).
I think the system here is just fine for measuring the tonal qualities of speakers and how they compare with each other. But it doesn't necessarily fully model what one enjoys about the equipment to pretend it is the end of story. So, its ability to make recommendations is limited. For example, it may be quite sufficient for comparison across very similar speakers (in size, type of technology, radiation patterns, etc) but not necessarily across very different types of speakers if that has a significant impact on the imaging, for instance. I think this issue also arises with dipole speakers as an example.
I don't see a problem in saying that this is a great measuring system for looking at tonal quality and accuracy of reproduction as a transducer but its behavior in a stereo system isn't predictable yet because of the limitations of the current models and measurement equipment. So, it cannot be the final word in speaker selection or evaluation.