Hi @Fatcat , thanks for posting. I can see that the axis are not clearly labelled.
X axis is frequency in Hertz.
Y axis is magnitude of transfer function of applied filter gain in dB.
Indeed your observations are correct. What I was trying to show was that in the computational modelling that the Linn software undertakes to correct for low frequency room resonances:-
small adjustments are made to the runtime filters to correct for the changes to speed of sound waves as influenced by temperature and humidity.
The variations come from changing temperature and humidity values from default 20 deg C and 50% relative humidity that the software has coded within it.
If you want to read up in detail about the various problems with undertaking computational fluid dynamics evaluation for real spaces there is quite a bit of material out on the web.
Linn have chosen to use one of the more ‘simple’ approaches, perhaps because it is easier to code if writing from scratch. This is the Finite Difference Time Domain method, which literally calculates the velocity and pressure equations for the air volume in the room step by step in a time evaluation, based on a pre allocated grid density of computation points.
There are several other algorithms that one can use. What a good modeller is supposed to do is cross validate between results obtained using one method (e.g. FDTD) with another method (eg Finite Element Moddeling or FEM).
I did have quite a discussion with the Linn Tech Director on this topic during my ‘Fitness for Purpose’ discussion with Linn.
If you look to my thread, there is an appendix added to my improved method for tuning one’s room with Linn SO.
The link to the appendix is here….
FYI, there are well established commercial tools that one can buy for which there are CFD add-ons. COMSOL multiphysics is such a tool.
Linn used the COMSOL tool (video evidence is available for this) to design their new 360 loudspeaker range, so should have had the possibility of at least cross validating anything they wrote internally.
So it is explicit that the calculations of Linn software are entirely due to the modelled temperature and humidity adjustments, the real question is do those calculations bear any resemblance to reality. Which can be confirmed by acoustic measurements at the appropriate associated temperature and relative humidity’s.