posted 09-12-2012 12:37 PM                    

quote: Chuck McGregor

---

In fact, I would bet the best PA technology (loudspeakers) I know of against the best cinema technology.

---

posted 09-12-2012 05:06 PM                       

quote: Tony Bandiera Jr

---

Now a concert sound system (Sound Reinforcement System) IS designed for a wider frequency response and high dynamic range, BUT unlike a cinema sound system it is designed for extreme sound pressure levels (with a passing nod to accuracy) and, something you do NOT want in a cinema, broad, even coverage.

---

posted 12-22-2012 07:27 AM                    

Sorry, been away from the forum for a while hence this delayed reply...

Tony Bandiera said:

>> P.A. (Public Address)systems are designed on the basis of projecting VOICES over a large area.

No argument if using "PA system" in a strict and rather archaic definition of that term. "Concert systems" are routinely referred to as PA systems, so we differ in definition.
Would you call systems for Broadway musicals or a houses of worships or a football stadiums "concert systems"? They are used for the same voices and music found in movies. There are many other types of audio systems other than "concert" or "cinema." In the broader definition any audio system that addresses the public is a "PA system." I used the term more in this sense rather than a voice-only system. So I'll use "concert system" to mean any non-cinema system used to reproduce full-range audio to an audience that substantially(!!) exceeds more than one person. (I do like "sound reinforcement" better but longer/harder to type )

Semantics aside....

quote: Tony Bandiera Jr

---

Now a concert sound system (Sound Reinforcement System) IS designed for a wider frequency response and high dynamic range, BUT unlike a cinema sound system it is designed for extreme sound pressure levels (with a passing nod to accuracy) and, something you do NOT want in a cinema, broad, even coverage. (I'll explain why in the next paragraph.)

An ideal cinema sound system must have accurate frequency response within the standard curve, must be capable of accurate reproduction from a whisper to a bomb blast (Concert sound rarely deals with a whisper) and, most importantly, has to have the RIGHT degree of coverage to reach the extreme sides of the auditorium, but MUST maintain the necessary channel separation illusion regardless of seating area. (Unlike concert sound where they don't always want or care about "stereo separation" but rather want the entire band to be audible clearly from ANY seat.)

I worked live sound for a few years during my career and can state that live sound and cinema sound are literally worlds apart in their requirements.

---

Actually the best concert system loudspeakers are most certainly designed with a intense focus (not a nod) to accuracy. I have intimate knowledge of this having spent the better part of 20 years working at four different pro loudspeaker manufacturers (variously from early 70s until retiring in 2007, though I still consult for one). Among other things I was in charge of the acoustic and electronic measurements at three of them.

For example, I have two smaller PA loudspeakers. That's what everyone calls them though they are hardly voice-only. They are also extremely accurate as it was ensured that they sounded exactly like these before they were released as a product:

Gelec 1038B

...only mine get about 6 dB louder (and didn't cost $8000+ each!)

Yes SOME concert systems are designed for extreme sound pressure levels. However, for comparable audience sizes the loudspeakers used usually don't substantially exceed the SPLs that a properly designed cinema system would provide for that same space. Those that do are more the exception, not the norm, such as larger scale touring systems. The reason? There's a dollar figure attached to SPL capability and normally people paying for non-cinema systems aren't made of money. Budgets are budgets.

Operating a concert system in "stereo" is generally frowned upon by the best practitioners. However, when multi-channels are required, properly designed multi-channel concert systems provide coverage to every seat from each loudspeaker, as should front channel cinemas loudspeakers. The goals are exactly the same: maintaining channel separation for optimum localization while providing the same performance from each loudspeaker to each seat. For a well-designed, multi-channel system you automatically get the former with the latter. That's the "right" coverage. "Accusing" concert systems of broad, even coverage doesn't square with SMPTE standards which require the same thing.

ANY pro audio system worth its salt these days, including concert systems, is capable of a dynamic range that can easily handle whispers and to bomb blasts. If it can't, it's either faulty, crappy, or improperly adjusted equipment. In this regard cinema systems have NO advantage.

As to the "standard curve" (X-Curve) -- This concept for system tuning was known and documented by the non-cinema audio folks long before it was "discovered" by those working with cinema. In fact, the same concept is routinely taken into account and used in concert system tuning. However, the "standard" used is derived from the specific acoustics of the venue, not a graph. (Ironically, actual room acoustics were the reason for the X-Curve in the first place.) There seems to have been some lip service in cinema-world being paid in this direction with the idea of layering room size onto the X-Curve.

The last thing that should be pointed out is that cinema loudspeakers normally use drivers, horns, and enclosures that are fundamentally the same as those same manufacturers use for their non-cinema loudspeakers. As well there is no magic in cinema electronics regarding dynamic range (whisper to bomb blast). Thus basic loudspeaker performance and electronic dynamic range have to be closely similar.

Tony, your experience with "sound reinforcement" is not the same as mine and seems mired in stereotyping systems for this purpose. I would largely agree with your comments as applied to many mediocre practitioners and system designs. However, my experience has been with state-of-the-art audio systems for all kinds of purposes including larger (and way loud) systems for stadium concerts. As such, I don't care about mediocre systems for any purpose and don't generalize my judgments on audio system capabilities or design on what such systems do or don't do.

Suffice it to say that in designing concert systems the basic requirements are the same as those of cinema audio systems. Among other things are channel separation (when needed), uniformity of coverage, large dynamic range, headroom, reliability, and, the most important one, reproduction accuracy. For the latter there is only one ideal goal and that can be largely achieved with the right measurement tools and the right signal processing (involving a combination of IIR and FIR filtering, not 1/3 octave EQ but that would be another long post to explain).

In short your assessments of "concert systems" seem based on run-of-the-mill, mediocre system designs and/or installation and/or operation rather than the state-of-the-art systems I've been involved with and to which I was referring.

I should also point out that I am actually giving a lot of benefit of the doubt in accepting (sound unheard and measurements unseen) to what is said in this forum about the stellar performance of cinema systems.

The bottom line is that I've know of no difference in the "ideal" technical requirements or goals that differ between full-range non-cinema systems and cinema systems in terms of the B-chain. The ones you cited are not differences.

The introduction to Chapter 5 of this book, lays out fundamental principles for designing audio systems. By correctly and fully applying these principles to what cinema presentation requires, using available "PA" technology, and not using THX, SMPTE, et al standards, the result would be an audio system meeting or exceeding all the requirements you cited as well as those standards. (Disclaimer: I authored the introduction.)

Architectural Acoustics Principles and Practice (2nd Edition)

Cheers

posted 12-27-2012 12:36 AM                    

quote: Chuck McGregor

---

I should also point out that I am actually giving a lot of benefit of the doubt in accepting (sound unheard and measurements unseen) to what is said in this forum about the stellar performance of cinema systems.

---

posted 12-27-2012 07:49 AM                    

quote: Manny Knowles

---

Does this mean you've never heard a good-sounding cinema system?

---

Like I said, sound unheard and measurements unseen. I have only heard two cinema systems recently. In truth I have been only a very occasional movie theater goer over the years, my milieu being live sound. Note that this involved only some rock 'n roll. Think mostly orchestra, theater, opera, ballet, and the like.

Some years back I remember a large NYC cinema in the West Side 50s which was impressive. Contrarily I walked out of another NYC cinema around that same time with a refund. Intelligibility was...well...it wasn't.

Of the two cinema systems I've recently heard, one literally caused pain in my ears. I've listened to and evaluated countless audio systems of all types and quality levels over the years. I can't remember one causing actual real-time pain unless it was from being too loud. This wasn't the case.

The other system is the one I worked on. I know the loudspeakers are appropriate for the room in terms of pattern control, output capability, and they measure close to being anechoically transparent. The in-audience responses conform tightly to SMPTE specifications as to X-Curve and SPL uniformity. (I could post the measured data). The subjective opinions of others is that it sounds great. So I'll stick my neck out and say I think I've heard one good-sounding cinema system.

So, at best I'm 2 for 4 on good/bad cinemas that I recall, with any others being distinctly unmemorable in terms of the audio.

To be clear, I view a cinema audio system as just another variety of audio system. The details of how and why a cinema audio system is set up doesn't make it unique compared to other audio systems. In fact, I've found it was far more straight-forward to do the acoustical tuning of "my" cinema system than many of the systems with which I've dealt. Of particular note, I found the electronic gain structure is a piece of cake for cinema audio compared to non-cinema systems.

Only a couple of other cinemas that might be good to hear are an hour or more drive from here. There are also a couple of cineplexes around, but not much closer. I AM planning on going to one or more other cinemas at some point and promise to report. Aside from guessing, any known "good-sounding" cinemas in Maine?

For better or worse, that's my story...

"If bad sound were fatal, audio would be the leading cause of death." -- Don Davis

posted 12-27-2012 01:08 PM                    

quote: Chuck McGregor

---

The details of how and why a cinema audio system is set up doesn't make it unique compared to other audio systems.

---

posted 12-30-2012 09:22 PM                    

quote:

---

quote: Chuck McGregor
The details of how and why a cinema audio system is set up doesn't make it unique compared to other audio systems.
Tony Said:
With all due respect Chuck that statement right there causes you to lose all credibility in this discussion. Cinema systems ARE unique compared to other systems, just as concert sound systems ARE unique compared to other systems, as are home screening room systems.

---

quote:

---

Cinema sound is mixed differently from sound for DVD's and home video, from sound for television shows, from sound for concerts, etc. To think otherwise shows a clear lack of understanding of the differing disciplines of audio.

[I hope you are not implying that I have a clear lack of such understanding or even a modicum of such lack.]

I have been in this industry for over 30 years now and one thing I will say is that there is always something new to learn.

[About 45 for me but 100% agree!]

There is NO one "correct" EQ curve or equipment lineup for ALL cinemas.

[Awww, sure there is for the latter. Just ask any manufacturer ]

The same held true for my time working concerts and live sound shows, every venue presented unique challenges (and if that wasn't bad enough try mixing stage monitors sometime...talk about a constantly changing environment..oy vey.)

[Been there, done that, don't like it, but do know some people who do serious and a serious amount of monitor work. However, I've been involved in the technology one way or another from the days of "What's a stage monitor?" to the "More ME in the monitors" of today ]

---

posted 12-31-2012 12:42 PM                    

Tony,

Thanks for the kind reply. As to you being harsh (blunt)? I used to live on 43rd and 10th, NYC. You can do the math

quote: Tony Bandiera Jr

---

I think I was channeling another thread here where we had a sound mixer who was rather um, wrong about a great many things.

---

I can't believe there's a sound mixer who was wrong about a great many things.... (insert "isn't" in place of "was")

In truth there are few who are wrong, if at all, about very little...Robert Scoville comes to mind.

Having a knack for monitor work is something very few can claim. Like they say (reportedly) down under, good on ya.

Getting back to audio systems and "PA"/cinema audio:
Here's a link to a file about the theater I'm working on to give you an idea about how I approach things.

"My" Theater

The most important thing I start with for any "full fidelity" system is getting pristine anechoic performance from the loudspeakers. That means removing colorations. The problem is measuring them, identifying the sources, and correcting them.

The spectrogram used tells the tales. With the appropriate scaling and resolutions of the output spectrograph, visible deviations from the "perfect" loudspeaker correlate very will with audible coloration and vice versa. Calling that what it is, coloration = distortion. These distortions occur in both time and frequency domains and have little to do with traditional "distortion" data (harmonic, IM, et al). Measuring and identifying the sources of the distortions, understanding the mechanisms that cause them, and applying either physical or electronic correction is not a trivial task. Also identifying those that cannot be corrected is just as non-trivial.

The fact a loudspeaker has flat response gives one at best a tiny clue as to the "sound" of a loudspeaker, i.e. how much or little coloration it has. When you can subject one to the various analyises shown in the document I linked, a very different and less flattering picture about a loudspeaker's performance will usually appear.

I do not have hardware (as yet) to remove what colorations still remain, but I know the primary sources: reflections in the LF diaphragm, some horn honk, and reflections bouncing around between the phase plug and diaphragm in the HF driver that smear the upper HF transients. All else being equal, these particular colorations are both a major source of why loudspeakers "sound" different and, unfortunately, they are inherent in the design of these devices. You make one you get these anomalies. Only the magnitude varies with from design to design.

Fixing such things can require a combination of physical and (mostly) electronic solutions. The latter involves developing a single filter algorithm for the loudspeaker using a combination of IIR and FIR filters. The basis for this kind of algorithm was patented by my last boss, Dave Gunness. The patent goes into the more detail:

Patent

To me it should be self-evident that the direct sound performance of a loudspeaker (its anehchoic performance) is more important than any other factor (like the room's effects) in determining one's subjective notion of sound quality. The goal is transparency = lack of coloration.

Most all loudspeakers have colorations which is why we have discussions over which sounds "better" and why some loudspeakers sound seem to sound "better" on some source material than others. If all loudspeakers behaved as the proverbial "straight wire" there would be no such discussions. You want the input signal to arrive at the listen intact in time and frequency. Both are measurable and if not "intact" there will be colorations. As a result most all system tuning is a actually a process of doing the best as one can to make these distortions as "pleasing" as possible. Especially with only 1/3 octave equalizers, there is little else you can do. The colorations do not usually have 1/3 octave Qs, and 1/3 octave does nothing effective in the time domain, the main provenance of these colorations.

For "my" system, standard IIR DSP was used creatively to reduce or mask what colorations I could (I know a few tricks), but they are there and they can still be heard, albeit they are fairly subtle. Another thing I did was to time-align the HF and LF components to the temporal center of the seating so those in the front and rear row experience the least amount of offset.

As for how the system "sounds," one critique after the system was tuned as shown, was for a "Live at the Met" HD performance. A professor of music, life-long opera lover/performer/conductor/composer/teacher came up to me and said the sound was "stunning."

So there's a look at how I have applied some of my "PA" knowledge to a cinema system.

In case you're wondering, the L/C/R loudspeakers are the widely vilified (on this forum) EV TS940s. They were designed by Dave Gunness while he was an engineer at EV where he was also a primary design engineer for the HP series horns. As such, I have viewed ALL negative comments about this system as being the result of bad system tuning or faulty hardware. Tuned properly even with standard DSP (this cannot be done with 1/3 octave EQ or cinema processor crossovers) they provide excellent performance under stringent objective and subjective analysis.

I should note that the 1/3 octave responses in the linked document on the theater were taken at 1/48 octave resolution with SMAART. These spectrums were power averaged in SMAART and the averages smoothed to 1/3 octave with a proprietary audio measurement/analysis program called Rayleigh, of which the spectrogram is a part. Note that there was no EQ applied to the anechoic responses to get the X-Curves, but that's whole other discussion!

That's my story

posted 01-01-2013 06:28 AM                    

quote:

---

For "my" system, standard IIR DSP was used creatively to reduce or mask what colorations I could (I know a few tricks),...

Another thing I did was to time-align the HF and LF components to the temporal center of the seating so those in the front and rear row experience the least amount of offset.

...
Tuned properly even with standard DSP (this cannot be done with 1/3 octave EQ or cinema processor crossovers) they provide excellent performance under stringent objective and subjective analysis.

...
Note that there was no EQ applied to the anechoic responses to get the X-Curves, but that's whole other discussion!

---