Tinyman you are awesome 
The ER4S will be more neutral than the hf5 with less treble and actually less bass. I too have a hard time believing that little bass is percieved as accurately neutral, but that's a different discussion. It's more important/efficient to compare iems like the AS2 and hf2 accurately than debate the definitions we use. Describing comparative perceived frequency responses is the easiest way to do it. "Bass at 'y' frequency is more emphasized in 'x' iem as evidenced by this instrument in this track."
Without comparisons these reviews can be very difficult imo. Drawing FR graphs is actually a really good idea.
The ER4S will be more neutral than the hf5 with less treble and actually less bass. I too have a hard time believing that little bass is percieved as accurately neutral, but that's a different discussion. It's more important/efficient to compare iems like the AS2 and hf2 accurately than debate the definitions we use. Describing comparative perceived frequency responses is the easiest way to do it. "Bass at 'y' frequency is more emphasized in 'x' iem as evidenced by this instrument in this track."
Without comparisons these reviews can be very difficult imo. Drawing FR graphs is actually a really good idea.
but do think about it. Each of our descriptive word does correspond to a certain pattern(s) in an FR graph. Texture generally being a bump in the lower frequencies (or conversely slower decay, or both). Clarity is generally regarded as a small bump in the upper midrange. Dynamics compares peaks (ratios) within an FR graph to each other in a way. When we talk about treble splash, we see a bump in the highs. Sure, we aren't talking specifically about x Hz, or y kHz, but the estimate is close enough 
