Brickwall Limiting

Back in the day, broadcasters used limiting so they could get the RMS signal as loud as possible without going over their allotted signal strength. While the program material was oftentimes highly compressed, there was no mentality of making a record as loud as possible because, unlike radio, the loudness was limited by the media’s physical attributes. For example, if vinyl was mastered too loud, the grooves would have too wide an excursion and reduce the overall playing time. With cassettes, printing too “hot” caused distortion. It was a balancing act; the signal level compared to the inherent noise of the medium had to be balanced with the ill effects of pushing too hard on these formats.
Author:
Publish date:

Fast forward to the digital age: New digital processing options allow for true, realtime peak limiting, where the attack time can be zero. No more overages! This lets people close the gap between their peak signal and their RMS signal, meaning that while the unlimited version and the limited version will peak at 0dB, the limited version will give the impression of being much louder. With CDs as the prevalent medium, the singular concern was simply in having no peaks over 0dB, as jumping needles and tape compression weren’t an issue; limiters became ubiquitous in mix buses, and mastering studios adopted “brickwall” limiting as the new protocol.

BUT WHY?

One anonymous mastering engineer with well over 500 credits explains: “The labels would send a track to three different mastering engineers, and they wouldn’t pick the best-sounding one, but the loudest one.” This spawned what became known as the “Loudness Wars,” fueled by labels and artists noticing the relative loudness of certain CDs that were brickwall limited, and requesting the same treatment from mastering engineers.

WHAT’S WRONG WITH HAVING THINGS AS LOUD AS POSSIBLE?

Nothing . . . if you don’t mind sacrificing dynamics, which brickwall limiting does. For example, in one very popular song, the verse is sparsely populated with sound, while the chorus is the band in full throttle. But, when that chorus comes, the music suddenly sounds smaller, and the drums practically disappear. This is because the recording was brickwall limited. The verse was already as loud as could be, and when the chorus was supposed to get louder, more limiting kicked in. You have to wonder if that’s congruent with the original intent of the band — after all, they added extra instrumentation and velocity to the chorus probably to make it “bigger” than the verse, but the brickwall limiting sure neutered that idea.

Brickwall limiting also messes with the mix’s balance. Note the waveform peaks in Figure 1, which are kick and snare. Brickwall limiting makes those transients quieter, so that everything else can be louder; in Figure 2, these transient peaks no longer exist. So all that time spent mixing, getting the drums just right, can be for naught once brickwall limiting comes into play. And if you make the mistake of limiting your mix, there’s no room for dynamic tweaking in the mastering process. Compression and limiting are completely different processes; limiting the mix really impedes the ability to touch the master with compression — which can “glue” everything together nicely, and let the end product “breathe.” It just doesn’t work as well after the fact.

There’s also the distortion artifact factor. After comparing 1/4" masters to the finished product, I (and many others) have noticed that certain limiters add distortion when brickwalled.

Also consider what’s going on beyond the master. As the preamps in many consumer CD players cannot tolerate the “hotness” of brickwall-limited signal levels without running out of headroom, while you may not have any digital distortion from your actual product, you will encounter distortion to some extent from many “civilian” players. Furthermore, when the music is converted to MP3, WMA, or AAC format, any prior distortion becomes amplified, and distortion that didn’t exist before may miraculously appear. The more dynamics a track has, and the quieter it is overall, the easier it is to convert to these formats; less limiting and loudness means less information, which means more opportunity to conceal the negative effects of this conversion compression.

But to come full circle to the genesis of limiting, if the music ends up being played via radio stations, brickwall limiting is even more damaging, as the radio station has to limit again to pump a track out over the airwaves, resulting in some pretty bad-sounding songs emanating from your car. A program director of one popular L.A. radio station told me that tracks given the loudness treatment pre-broadcast have treble that is “very scratchy.” And another big station in L.A. says that they received complaints from listeners that “couldn’t understand the lyrics,” which the station attributed to a byproduct of pile-on limiting.

WHY DO PEOPLE INSIST ON BRICKWALL LIMITING?

I don’t know, and most mastering engineers admit privately that they want it to stop. I like to imagine a world where mixers and mastering engineers weren’t forced to acquiesce to the requests out of fear of losing their business; a world where artists realized that making their albums as loud as possible doesn’t equate with better sound; a world where listeners were made to tackle their level requests by reaching over to the knob on their playback system labeled “volume” — and just turning the damn thing up.