But what is not logical to me in your line of argumentation is that with the identical measurement method & position (single point in my case) I am getting a very good result with Audiolense, and a less good result with Trinnov (lets say, good result is even FR). So to me it looks like the DRC software is the reason, not the measurement method.
Yes, this has been my experience as well.
The important part is understanding how each DRC/DSP software/system works. Some DRC software are "designed" to use a single point measurement analysis at the MLP as they use an
algorithmic approach to average the low frequency correction. Multipoint measurements use the "brute force" approach of dropping the mic at various locations around the listening position and average the response. Here is the issue and
@Dimifoot has already said with respect to his measurements, "they look super, but they don’t sound as super as they are set now...which is less “super”
Having been there myself, I agree. And there are technical reasons for it.
Here is what is happening. The multipoint measurement approach uses the full inversion of each measured response for averaging. That is trying to account for too much variation, hence the over correction. While the bass measures smooth, it sounds (incoming subjective words) dry, lifeless, not dynamic, compressed, limited. It is especially prevalent in DRC systems that not only use the multi-point measurement/averaging approach but do not offer the user a control for "how much" (over - joking) correction is even applied. So more = better. And there is no way to change it. This has been my experience with (all of) these type of DRC systems that rely on this multi-point measurement/averaging approach. See section on
minimum phase invertibility to understand the context of my meaning.
In contrast, DRC software that is designed to use a single point measurement
analysis/algorithmic approach, with a correction filter designed and in circuit, not only measures just as well at multiple locations around the listening position, more importantly, the bass response sounds tight, dynamic and clear with no hint of over correction. If you know
how to read a step response chart, then you can see where I have used a single point measurement analysis to design and generate a correction. With the filter in the circuit, I took
14 measurements around a 6ft x 2ft grid area of both left and right speakers at every point. Referring to how to read a step response chart, one can see that the 14 measurements are near identical over a 6ft x 2ft grid area. The frequency responses of those steps are just as smooth and posted on ASR somewhere and in my DSP book. The step response is much more revealing of low frequency response anomalies not only the direct sound, but over time.
This is the major advantage of a single point measurement analysis approach, if the DSP software was designed to be used in this way. Plus the user is in complete control of all correction parameters, including by how much. This is why it is important to understand how each DRC package works and most importantly, based on the technology and approach used, how does it sound!
Further, low frequency excess phase correction with 65,536 or even 131,072 tap filters provides full bass control with excess phase correction windowing capabilities of 800 ms at 10 Hz. This is why in my step responses you see here on ASR follows the target response, not just for the first 5ms, but for 100's of milliseconds. The vast majority of DRC systems, whether software or h/w don't have this excess phase capability at all as they are using IIR filters at low frequencies. Even if the tech changed from IIR to FIR at the low frequencies, the current limitation on h/w DRC makes it technically impossible to hold 65,536 tap filters. Most are limited to 8072 taps or less. There is
not enough resolution to be effective under 100 Hz.
These are the technical reasons why these multi-point DRC measurements systems won't "sound as good" in the low frequencies as compared Acourate or Audiolense. PS. both Acourate and Audiolense can do multi-point measurements and analysis. But the tradeoffs are very similar to what has already been discussed because it is the approach that yields the issue. It sounds less over corrected than the multi-point systems because of the analysis algorithm, but myself and others prefer the sound of the single point measurement analysis/algo approach as the bass response sounds the best, and still measures great at multiple locations