There's a lot of pontificating about statistical analysis with 'sparse data' when you do DBLTs. Which is why we did ABC tests rather than ABX tests. 'Statistical Significance' and other buzz words a LOT faster.
But a DBLT is a MEASUREMENT. Your instrument is your DBLT panel. Like any good instrument, it has an accuracy that you need to check regularly.
A good sounding cone doesn't have UGLY cone breakup.
A metal cone resonance is UGLY .. as is the straight sided cone behaviour shown in Arie Kaizer's early papers. You see the ugliness, not only on a SCanned Laser Plot, but also the frequency response and the 'waterfall'
But in a good sounding cone, the 'breakup' is controlled and doesn't show up in any of these 'measurements'. What you see is a transition from 'piston' behaviour to 'shrinking' behaviour.
I think it could be arranged to send you copies of books and papers.
As I see it today, a problem we have is that not all previous published DBT research was done very well. There is much we don't know about what they did or didn't do. We probably don't know enough to exactly replicate their research. To many unpublished little details that may have affect the results. We don't know if they tossed outliers, don't know if they changed the research question or deviated from planned procedures during or after the point of conducting the experiment, etc. IOW, they may have done things that today would be considered quite inappropriate.
However, there is no research money today to replicate what was done already, and no research money to study audio reproduction-related subject matter not undertaken before.
They had FTs and analog spectrum analyzers, etc., so they could measure and study more or less FR, steady state distortion and noise floor, and so on, but that leaves out a lot of things that are harder to measure and or that became more common problems as digital audio continued to develop (no matter how undesirable such effects may be).
Last edited: