Why do subs sound bloated or slow?


The use of subs in 2 channel audio is controversial around A’gon. Detractors argue that subs usually make a system sound bloated or slow.

IME, the two biggest challenges for integrating a sub into a 2 channel system are optimizing frequency response and optimizing transient response. When frequency response isn’t flat, the bass sounds bloated. When transient response isn’t time aligned, the bass sounds slow.

Here is my pet theory about why systems that use subs often sound bloated or slow: Under many circumstances, optimizing frequency response and optimizing transient response is a zero sum game. In other words, getting one right usually means you get the other wrong.

Thoughts?

Bryon
bryoncunningham

Showing 2 responses by audiokinesis

"Under many circumstances, optimizing frequency response and optimizing transient response is a zero sum game. In other words, getting one right usually means you get the other wrong."

Strongly disagree. When you get one right, you get the other one right at the same time.

Both the bloat and the slowness you describe are primarily frequency response problems; at low frequencies the ear has very poor resolution in the time domain so it doesn't really pick up the transient response in and of itself. The ear cannot even detect the presence of bass energy from less than a singe cycle, and it takes several cycles to determine the pitch. There is a correlation between transient response and sound quality at low frequencies, but that is because of the kind of frequency resposne curve that inherently accompanies good transient response. More on this later.

Now the ear is quite good at hearing the sort of frequency response anomalies rooms typically impose on subwoofer systems.

Given the wavelengths of bass energy and the time it takes for the ear to hear bass, by the time you hear it, the room has already very significantly altered the native frequency response of the sub. You have two primary mechanisms going on (really just different manifestations of the same mechanism): Severe peaks and dips imposed by room interaction, and a general trend of boosting the bass as we go down in frequency (room gain).

One effective solution is to use multiple subs. Each sub will interact with the room differently, and the sum of the two dissimilar peak-and-dip patterns will be much smoother than one sub all by itself. If you pay attention to internet posts where someone has gone from using one sub to using two subs, they invariably report more natural-sounding bass. So I believe that one of the reasons a lot of people prefer the bass of a good pair of speakers over satellite+sub bass is that the latter doesn't benefit from the smoothing we get from having multiple bass sources... unless of course you use more than one sub.

The observation that low-Q sealed subs sound tighter and more natural than vented subs goes back to room gain. Typical room gain is +3 dB per octave below 100 Hz. A low-Q sealed box rolls off at -6 dB per octave below system resonance, so combined with room gain that gives good extension without over-emphasizing the low bass. On the other hand a vented box is probably "flat" down to a much lower frequency before room gain is factored in. Suppose our vented box is "flat" to 25 hz; after room gain, it's now +6 dB at 25 Hz! That will definitely sound heavy and sluggish. As proof that it's the frequency response and not the transient response that matters, take both subs outdoors where there is no room gain, and the vented sub will sound tight and solid but the low-Q sealed sub will sound thin and anemic.

So to recap, audible bass anomalies arise primarily from frequency response problems, which in turn are primarily caused by room interaction issues.

Duke
dealer/manufacturer
"Many subs have group delays of up to or more than 1 cycle. Of course this is audible."

I don't remember the exact group delay figures used in the study I read, but they were consistent with what one would expect from a well-designed vented system. Using digital signal processing, group delay was isolated from frequency response (don't ask me how they did it). On test signals group delay was slightly audible, but on music it was statistically inaudible. The audible effects normally attributed to group delay were replicated by a frequency response curve mimicing the high group-delay system, even if the group delay is eliminated by DSP. So while large group delay correlates very well with perceived "slowness" in the bass region, it is not the actual cause.