Is a FLAT response the IDEAL?


Sounds in nature are not a flat response, quite often, there are natural attenuators, accelerators and amplifiers, including horns (caves), wind and water, let alone reflections, absorption and diffraction.

Similarly the holy grail (one of them) of recreating outdoor, concert or live music, and so on, abound with these shifts in the environment or context where the experience happens and the recording takes place. Are we depending on the mic positioning, and mic performance, along with mixing equipment, format and so on, to enable recreation of the environment when moving to playback. How does a flat response curve help?

Of course, we have DSP. For Club, Hall, Rock, Indoor, Outdoor and may other shifts to music recordings. And mastering adds reverb as another way to create a 3D version of context/venue. These are averaging processes that apply universal shifts to shape a standard curve across the music stream continuously.

So why is it that we pursue flat response curves? Or DSP generated fixed curves? How does flat recreate that live ’being there’ experience.

When designing equipment including components, such as DACs, and speakers, most seek to judge against a flat frequency response.

Mind you, how on earth can we allow other than flat. Turntables as most here know, use the RIAA curve to fix the problems of hearing that itself is not flat. But even that is aimed to deliver a flat hearing response.

I don’t understand. If we are trying to model or capture the original event, how does flattening everything help? And, what are the alternatives? How do we achieve close to the venue or location, given so many unique variables, that our approximations just don’t seem close to the original. It’s no wonder... Have we selected flat because it is the best average we’ve got?

Do immersive audio methods of sound reproduction do it better? Some prefer pure stereo, some like DSP, some multi-channel and multi-speaker methods including ambiophonics.

Where does the ’flat curve’ fit into the equation here, vs say cross-over design or powered speakers or upgrades as a priority? Should we care about it?

Well that’s enough to launch this inquiry...

128x128johnread57

I don't know if a flat response is the ideal sound for me or not. I just know all the albums I have I grade D, C, B, A, and A+ based on how good they sound to me. A flat response is one of the least important criteria I listen for.

Unfortunately, the word "flat" carries an unattractive connotation. The word implies dullness and lack of life. What it means in the hifi game, though, is truthfulness. It means that whatever music you put into the system, it’ll reach your ears with the qualities the creators intended to convey.

There is no ideal response. What you need is custom tuning. Every track you play requires its own curve. But nobody has yet invented a system which can store a different EQ preset for every track. The ideal curve is not fixed because it will depend on the room, your ears, speaker placement, recording and your preference. On some days you might want a little more highs, on other days you might want less. 

Flat is just used as a marketing tool by the speaker companies to sell their speakers. A flat response looks impressive. A ragged curve does not. 

The studio pro folks love flat tuning. They dont care about sound quality they just want flat so thats what the speaker companies give 'em. 

Hifi folks are generally more fussy so you have a variety of curves being offered on the hifi market. Some are flat as a pancake but some are not. 

If you want to change your response, you need an EQ. You can then have whatever curve you like. 

I love how even a topic that has only one possible answer still becomes a debate. Darn audiophiles!  This one was a gimme!

There are many factors involved in reproducing a recorded signal. Some are more important than others which is why when some factors are not done well the sound may still be better than when other factors are done well. But the goal is for every factor to be accurate. Flat response should always be a goal although I consider it a secondary factor. I find dynamic linearity, minimal compression of level changes of all magnitudes from micro to macro a primary factor in sounding real. But anyone not attempting to achieve flat response to be in error.