I appreciate all the different perspectives being offered here. And agree with some of what you’re saying above, Mike.
A flat frequency response is what I look for in most of my audio components, including the direct response of a loudspeaker. I think that’s about as close to neutral as you can probably get without knowing how a recording was actually mastered.
I have already mentioned Floyd Toole’s video here once. But maybe it deserves mentioning again, since he provides a pretty good overview of this subject, and challenges that are involved in actually acheiving a “neutral response” on loudspeakers.
Neutral for me in a headphone would be the response that best approximates the in-ear response of the intended listening space of a recording. If, for example, you have a recording that’s mastered on a Sennheiser HD600 with new pads, and that is also the producer’s intended listening space, then a headphone that matches the in-ear response of the HD600 with new pads on your head represents a neutral response. Pretty simple.
It would be nice if things were that easy. But in most cases they will not be. And we won’t be blessed with that kind of specificity. We won’t know how the recording was actually mastered. Or exactly how the producer may have intended you to hear it. Whether they wanted you to listen on headphones, or IEMs, or in your car, or in your home on a nice sound system. Or some combination of those spaces.
There are different approaches (and potential solutions) to this problem. An audiophile, for example, might tell you that neutral is when he (or she) feels like he’s there in the room or the concert hall with the performers where the recording was made. In many cases though (esp. with electronic music recordings), no such room or concert hall exists. Or it only exists virtually on a computer, or in the mind of the producer or sound editor.
You can also look at how different recordings are mastered, and try to form some more general assumptions from that. The methods of mastering content are now as varied as the devices used for listening though. And a recording certainly won’t sound the same on a computer or desktop system as it does on an IEM or a good set of speakers in a room. Good music producers will of course know this, and try to make a recording sound its best on the device (or devices) that their audience is most likely to use for listening. If they’re very good, they’ll try to make their recording sound at least passable on all of the above.
Another approach you can take is to set some standards for mastering, and also for listening, to try to bring the two a little closer together. And specify what type of devices to use for the production and reproduction of recordings. The “spinorama” system that Floyd Toole and others developed for loudspeakers is such a standard.
Consumers are the ones who will ultimately decide if this type of standard is successful and worthwhile. The trendlines in consumer speakers suggest that it is though. Because many of the newer speakers being designed by the major mfrs fit the spinorama model. This is also the type of listening space that many higher-end mastering facilities will try to replicate. And at least some headphone mfrs are also trying to replicate in their headphones.
If a recording was mastered to sound best on the tiny speakers on your computer though, then it may not sound that great on a pair of good speakers in semi-reflective room, or on a pair of headphones designed to approximate a similar response at your eardrum. Standards like the spinorama give us a place to go for solutions though, when other approaches to the question of “what is neutral” lead to more uncertainty or confusion.