The inclusion of the slope variable actually adds some increased sensitivity to lower-Q deviations, just as we'd want psychoacoustically, so it's not as 'dumb' of a model as some naysayers make it out to be.We need to better model how different deviations form the target are perceived for a smarter updated score model.
First you need to present valid evidence it needs improving in the form of controlled blind listening tests.First the In-ear target needs to be improved.
I'm not aware of any blind test evidence of listeners finding the Harman in-ear target too 'mid-elevated' (in quiet conditions, which is how the vast majority of audiophiles and reviewers will judge sound quality).Find the cause for the mid-elevated result in it's current state, perhaps by using mobile subjects like USound.
Last edited: