View Single Post
Old 1st November 2019
Lives for gear
Jens Eklund's Avatar

Originally Posted by RyanC View Post
Thanks for joining in Jens-

Hmmm, I interpret all of Floyd Tooles work and what he shares with the public at large to be that "flat and smooth" is having the least amount of deviation from whatever curve it is...The "Harman Target Curve" is from Toole and company no?

In my personal direct experience, the main issue here is the bass lift, not the treble tapering. Lots of experimenting here with a bunch of engineers actively working and we always come up with +6 to +8 @20hz tapering down to flat around 100hz, each full range speaker measured separately.

In terms of tapering the treble a little bit, I can agree that's a little more dicey. But at the same time many commercial mixes are bright for my taste, and especially are bright to listen to for full work days. I find ~ -2dB to take the edge off, especially for long days in the studio, but I can concede there is more risk here of overbright mixes and agree that -4 ala B&K is too much. I should have been more clear to say that B&K is, to me, a good place to start the experimenting.

I think it's important to keep in mind that the real 'standard' has been set by hundreds of thousands of records that have been released and listened to by the masses and adopted as a sort of de facto benchmark after each person's lifetime of listening. It is certainly a vague target, but musicians, artists and engineers do have a fairly well established concept of it in their minds.

Regarding speaker design- (you know all this better than me, but for the sake of the OP to understand), designing a speaker in an anechoic environment to be flat on-axis obviously doesn't translate to trying to achieve a flat response in an non-anechoic environment. We should see some degree of boundary loading- which is to say every anechoic-ly flat speaker is completely expected to exhibit a rise in bass response in a real room (which as you know can be quite significant). Also it's much less likely that people are seated on the tweeter axis at home or in a car...

Add in a bunch of bass traps, and as you create a quasi anechoic environment it's reasonable to me for end that users to find a good balance there that suits their work.
I´ll try once more to explain what I´m trying to say. I know this is a tad technical and I might not be the best communicator …

First; I´m not against a non-flat target curve. The total opposite is true. When I calibrate our systems in studios, I normally make presets with a bass lift from +3 to +7 dB (sometimes even more if the client requests it). I never do a completely flat preset since no one ever asks for it. Most of my clients prefer the +5 or +7 dB preset.

What I´m trying to say is that in the upper range; since what we hear is dominated by the direct sound (yes, we can actually “hear trough the room” to some extent in the upper range, just like we can gate the impulse response to exclude the room contribution, but only for higher frequencies due to the bandwidth requirement); it make no sense to me to try to match the response to a target curve that is based on the steady state response (the entire or at least a large section of the impulse).

The B&K curve is the result of looking at the steady state response of a number of measurements made in domestic environments. If you where to look at the individual measurements and gate the impulse to include only direct sound; I wouldn’t be surprised at all if most of them would measure more or less flat in the upper range.

Look at the graphs in this post:

The Titanic, room curves and other GS style OT wanderings

Notice how the steady state response is similar to the B&K curve but the gated response (only direct sound) measures flat (again, in the upper range).

In a well treated control room, the difference between a gated response and the steady state response is minimal (in the upper range), so if you where to apply the B&K curve; you would effectively tilt your direct sound in the upper range without knowing what you are doing. Again, not saying that some might prefer this, but I also know that you can get used to pretty much any house curve if you spend some time with it …

So bottom line; use whatever target curve works for you but make sure you know what you´re doing or you might add a tilt to your direct sound unintentionally, and then getting used to it witch might end up being a problem later on when working in other studios with a flat direct sound in the upper range (since this is the standard).

More here:

Acoustics Issue

Acoustics Issue

The Titanic, room curves and other GS style OT wanderings