Initially, in the mid 1920s, we saw many technologies brought to the table, but during the 1930s, the Academy of Motion Picture Arts and Sciences established a standard for sound-on-film.  A sliver of space to the side of the picture frame was home to an optical track through which light was driven and picked up by a photo sensor.  By varying the width of the opening, the amount of light delivered to the photo cell varies as the film passes, creating varying voltages in the sensor, and a soundtrack is created.  The basic principal of the Academy Optical Mono soundtrack is still employed in part today.

Technology advanced fast in the early half of the century, especially as it related to loudspeaker and amplifier technology.  Thus was born a need for the industry to achieve a mechanism by which a soundtrack could be crafted and presented with consistency, regardless of the varied and widespread caliber of hardware it would encounter in a run.

Ideally, a sound system should have as flat a frequency response as possible.  Reality in the 1930s was otherwise.  In particular, while "better" theaters and studio mixing facilities exhibited "good" response, there existed an overabundance of movie theaters with poor high end response and an almost non-existent low end.  In addition, the Academy Optical Mono track itself suffered from a curtailed high frequency response and an inherent hiss.  The mechanical barrier of a perforated movie screen further filtered high frequencies.

Ultimately, in 1938 the Academy mapped out the response of the "worst case" scenario of the day, and the Academy Curve was born.  Mixing stages and better theaters applied equalization to simulate the poor response of lesser exhibition facilities, ensuring that they were evaluating and crafting the soundtrack such as it would be heard by the public.   Thus was laid the foundation for standardization of soundtrack presentation.

The Academy Curve is also known as the Normal Curve and is defined as flat 100 Hz - 1.6 kHz, down 7 dB at 40 Hz, down 10 dB at 5 kHz, and down 18 dB at 8 kHz.

In layman's terms, this is a dramatic attenuation of the treble and a generous reduction of the bass.

Naturally, speakers and other audio hardware improved as the decades marched on, but the optical soundtrack did not.  Eliminating the Academy Curve would simply have exposed the inherent hiss which it masked.

When Dolby got involved with cinema sound in the 70s, the first thing they did was apply their noise reduction technology which was already well entrenched in the music industry.   The 1971 "A Clockwork Orange" demonstrated to the world an improved bandwidth and high frequency extension thanks to Dolby's A-Type noise reduction.  As a result, the world was finally able to give up the Academy Curve . . . but not for a flat response.

Research done a decade earlier by C.P. and C.R. Boner defined a need for a "house curve".  They based this on the fact that a flat electro-acoustic frequency response in a large room sounds too bright on well-balanced program material.  Or in simpler phrase, we perceive sound in a large room to have more treble.

Here stood an opportunity to put into place a curve system which could go beyond simply dictating a flat response, but also address the psycho-acoustic differences of sound in large, medium, and small theaters and mixing facilities.  The �X Curve� was born.

Also known as the wide-range or eXtended range curve, the X-Curve is defined in ISO Bulletin 2969 as pink noise, at the listening position in a dubbing situation or two-thirds of the way back in a theater, to be flat to 2 kHz, rolling off 3 dB/octave after that. 

The measurement of EQ is done using wideband pink noise excitation through the speaker. The larger the room, the longer and greater the reverberation buildup over time, resulting from this steady-state signal. If we measure EQ on a fixed bandwidth basis, like full octaves or third octaves, the reverb which tends to be stronger at low frequencies will tend to make the SPL read louder there than at the higher frequencies where it is better absorbed. By knowing how much the meter reading is influenced by this effect, a curve can be used to compensate. That's what the X curve does.   Unlike the Academy Curve before it, It has nothing to do with high frequency capability in the playback chain.

It turns out that while SPL meters are "stupid" and cannot distinguish the flat direct speaker sounds from the room sounds which arrive commingled, humans hear the true spectrum of the direct sound and interpret the room sound as reverb or spaciousness. It's a bit more complicated than that, but you get the idea.  Small, deader rooms need less and less of the full X-curve compensation, as they have less of the meter-confusing reverb present.

So while we have the basic ISO definition of the X-Curve, it is fundamental to realize that there is in fact a table that contains variations to the curve based on room size.  For example, there is a Small-Room X Curve, designed to be used in rooms with less than 150 cubic meters, or 5,300 cubic feet. This standard specifies flat response to 2 kHz, and then rolling off at a 1.5 dB/octave above 2 kHz. Some people use a modified small-room curve, starting the roll-off at 4 kHz, with a 3 dB/octave rate.

For almost 30 years, the X-Curve has provided the motion picture industry with a valuable standard that ensures plausible interchangeability of program material, from one studio to the next, from studio to theater, and from film to film, which takes into account the different perceived spectral response of different room sizes.

Its worth noting that in the same decade, Dolby also pioneered the concept of a calibrated playback level for mixing facilities and movie theaters.   A reference-level pink noise piped through the audio chain was calibrated to 85 dBc (the "c" means the curve is "C Weighted", vs. "A Weighted", referring to the way humans hear different frequencies).  With mixing facilities and theaters setting their master volume level by the same rule, the levels chosen by the film makers would be assured in the cinema.

With the Optical Stereo track still in use as a backup track, Dolby Digital blew the doors off dynamic range, extending it to 105 db (the difference between the softest and loudest sound is 105 dB).  Again, the calibrated level process remained the same, so ultimately Dolby Digital gave the film-maker more volume at the top end.  Unfortunately this headroom has been more abused than used.  You can read all about it in the excellent paper by Loan Allen of Dolby Labs entitled "Are Movies Too Loud?"

What does all this mean for us at home?

Although rare, there are old, classic films with their mono soundtrack intact on home video and could use the high frequency attenuation of the Academy Curve.  Several of what are thought to be the better surround sound processors offer such a filter, and I routinely encourage others to follow suite.  In the digital audio age, the cost is relatively minimal, and given that the Academy Curve brings with it no controversial stigma, it's a nice tool to have around.

Since the advent of Dolby Stereo though, the soundtrack of virtually every movie (be it Analog or Digital) is crafted for playback over a system equalized to the X-Curve and set at the reference playback level.   Unfortunately there is no entry in the X-Curve table for rooms as small as most home theaters, and even if there were, like theaters and studios, you'd need 1/3 octave equalization on each channel to conform.  Don't worry.  Read on.

If you could get your hand on the X-Curve table, one corollary would stick out for you:  As the rooms get smaller, less and less of a roll-off is defined, because as we said, smaller rooms have less of the reverb which the X-Curve addresses.

By the time we shrink a room down to typical home theater size, we can say that no X-Curve compensation is needed, much to the contrary of popular opinion. There is not an inherent overabundance of treble in motion picture soundtracks, at least not due to the X-Curve.

In 1988, the George Lucas company THX, which previously had established elevated standards for motion picture presentation, entered the consumer electronics market with its THX Home Theater Program.  Among the program's many facets were new and proprietary processes for THX surround sound processors, including one called Re-EQ.

Re-EQ is a filter applied to soundtracks after Pro Logic or AC-3 decoding.  Although developed using the X-curve as a reference, it is not an X-Curve per s�, but a rational choice of filters that address the common (and in my opinion true) perception that film soundtracks played at home are typically too �bright� (too much high frequency).  However, as we've said, the fact that the soundtracks were mixed for an X-Curved presentation has nothing to do with it.  While hunting for an X-Curve compensation, which technically should not be required, THX developed a filter that aptly addresses other issues in home theaters.

The factors which make home theaters sound subjectively "too bright" are varied and numerous, and no two home theaters are going to be the same in this regard.  Home theaters are usually more reverberant than studios, cinemas, and dubbing stages, sustaining high frequency energy much longer, despite their small size.   The speakers are often much brighter than properly set up theater arrays.  As we push the playback level up towards reference level, the problem gets worse.  If a system is genuinely capable of high output with low distortion, then it is possible to recreate the " movie theater experience" and enjoy it.  Unfortunately, there are far, far fewer systems capable of this than we are lead to believe.  Many so call high-end speakers and many receivers exhibit harmonic buildup at high output and tilt the spectrum in favor of the treble, further exacerbating the brightness problem.  Just because you paid a lot for it does not necessarily mean it is suitable for reference level movie playback.  Sorry.

We must also consider that the perception of film soundtracks being too bright in home theaters has perhaps less to do with the spectral balance of the recording, dubbing or home theater systems, and more to do with heavy handed and ill chosen mix techniques used during the mix process. With such mixes, Re-EQ can�t correct anything but can make it easier to listen to, one of many reason I'll continue to want it on my processor.

THX chose an X-Curve type filter since their listening tests show that it�s more likely to produce an accurate correction more of the time. Of course, playback systems and film transfers are sufficiently variable that a single curve is never going to be �correct� all of the time. It is not intended as an absolutely correct interpretation.  As with everything THX does, they look at the consumer's reality, the realities of the Consumer Electronics industry and the available technical tools, and then create a solution that makes rational sense for most consumers most of the time.

We've often been asked the question, "Why not just remix the soundtrack or apply a form of filter to it when mastering the home video?"

The problem we face right now is one of trial and error:  When we take that new release DVD rental home, will it be better served by having Re-Eq on or off?   Who knows? We as consumers have very little to go on in this regard other than our own ears.   THX (naturally) advocates the status quo with Re-Eq, in place while other industry professionals openly advocate remixing the soundtracks.

No one ever said being a home theater aficionado would be easy.

When the ATSC* chose Dolby Digital AC-3 as the audio format for HDTV, among other things they defined a 2-bit flag called "Room Type" which is said to define the type of mixing room the material was made in.  The flag can be set as follows:

00 not indicated
01 large room, X-curve monitor
10 small room, flat room monitor
11 reserved

Contrary to some opinions, this flag is not meant to be a "Re-Eq on/off" signal.  If the correct X-Curve was properly implemented in the studio, it should make no difference, at least in terms of EQ in a "good" home theater (one which does not sound too bright).

Conclusion

The lesson to glean from the movie industry's experience is that "the room" plays a huge role in what is realized from a playback system.

While the X-Curve puts all the commercial facilities on the same page, there's no real way to account for different reverb characteristics, since even rooms with identical Rt60** measurements can still sound different due to a different spectrum of absorption.   How do they finally decide the room is right or not after EQ adjustments have been performed if X-curve can still allow for some subjective variability? The answer is, listen to it with familiar reference material.  Does the system sound right, bright, or dull? If it is one of the latter two, further adjustment of EQ, acoustics, or other properties may be necessary.

I've always advocated Re-Eq and will continue to do so, but not because it is a fix for the X-Curve.  In my opinion, typical home theater spaces with typical home theater hardware playing typical contemporary material do sound too bright, and Re-Eq is a simple and effective fix most of the time.  Because that's a "most" and not an "all", I'm also big on processors that allow it to be engaged or disengaged independently of the other THX Post Processes.

Ultimately, I would like to see enthusiasts spend more time and energy on their room configuration, at least as much as they do on selecting their components.

Doing acoustical treatments in a home theater is the equivalent of applying an X-Curve in a commercial or professional facility.  While we don't have 1/3 octave EQ and a curve to shoot for, we can still fine tune our rooms to sound "right".  A little absorption here and some diffusion over there can make a monumental difference in a room.

After all, a modest system in a "good" room can easily outperform upscale, expensive equipment in a bad one.
 

� Copyright 2002 Secrets of Home Theater & High Fidelity
Return to Table of Contents for this Issue.