Question on making a DCP from Blu-Ray

[Martin Brooks]

One of the old ladies in the senior's apartment building where my mother lives now has told me that she doesn't watch movies any more because there's too much music(?) and she can't hear the dialog.

I wonder if it's something to do with hearing aids.

She must like movies, though, because she told me that she named her son Wade after Errol Flynn's character in Dodge City.

Not that everyone is going to rush out and buy new TV sets, especially the elderly and they won't necessarily be able to understand the UI, but ATSC 3.0 takes care of this - you will have the ability to change the level of dialog, effects and music; switch languages and in the case of sports, even switch announcers.

[Marcel Birgelen]

I don’t know how many times I had to explain to people what the word “bollocks” meant!

I've never considered myself especially intelligent, but sometimes I really doubt the general intellect of the masses... My vocabulary in the languages I pretend to speak is all but complete, yet, I can't hardly imagine any exchange between two or more human beings, where at least the general meaning of the word "bollocks" isn't immediately clear, simply by context.

I guess we do have the advantage of living in a part of the world where subtitles are generally accepted. Whenever someone speaks with a heavy accent, even in the "mother language" of where the film is being exhibited, it's pretty common to put subtitles in there. So, even a Flemish movie shown in the Netherlands will usually be presented with subtitles, even though many Dutch will be able to comprehend most if not all of the dialog without subtitles.

This is not to say that there aren't good 3-way (and 4-way) speakers that can do a great job...there are...they are just more rare and typically more expensive. They are still trying to match timbre and DI. I can tell you that the QSC SC-423 and SC-424 (my personal favorite cinema stage speaker, by far, in its price range) can do an incredible job despite its crossover points but they are the exception more than the rule.

I agree that, when you're budget-constrained, a good 2-way system beats any low-quality 3-way system, but if done right, in my experience, a good 3-way system actually improves dialog intelligibility.

A crossover will never be perfect, as you indicated yourself, you're changing between "cones" with different characteristics, but getting this "as right as possible" IS the bread and butter of good speaker design. Combined with modern room correction tools that can even reduce time domain related problems, you should be able to get your crossovers working in a way they don't really impair dialog intelligibility.

What I've found though, is that whenever a complex sound mix is playing that also uses the center channel for effects or music (this obviously includes mono mixes), a 3-way system will usually be able to render the output much cleaner than a 2-way system, which will also improve dialog intelligibility.

Getting back to hearing aids in general. What I've learned about them is that many of the tradiional "amplification style" hearing aids apparently perform a frequency shift for higher frequencies into the lower frequency domain, as many people with impaired hearing seem to have difficulties with the higher frequency domain. I don't know if this is individually adjusted, but I can imagine that such a frequency shift can also cause problems with dialog intelligibility, especially if there's more going on than just a single person speaking.

[Randy Stankey]

I can't hardly imagine any exchange between two or more human beings, where at least the general meaning of the word "bollocks" isn't immediately clear, simply by context.

The conversation would go like this...

First person: “What did he say?”

Second person: “Bollocks!”

First: “You don’t have to be rude! Just tell me what he said!”

Second: “Bollocks! He said, ‘Bollocks.”

[Steve Guttag]

Marcel...I stand behind, more often than not, a 3-way speaker hurts your cause. Key elements of dialog are in two sections of the frequency spectrum and 3-ways put their crossovers just about right where you don't want them...which is why 500Hz crossovers were/are still ideal...almost nothing dialogy is happening there. Sure that frequency is in the dialog range but nothing that is key to intelligibility. The closer you get to the 3KHz you get, the closer to danger you get and 3-ways almost always have their crossover in a 1.5KHz to 2.5KHz area. It is difficult to get a timbre and DI match between dissimilar drivers and crossovers can only hurt, not help that. Timbre is a function of the driver(s) and the physical box/horn that it works with. The best a crossover can do is get the most out of what is already there.

To put it in film terms...you HAVE to splice your film (for whatever reason). If you splice it at 500Hz, you are on the frame line. Few if anyone notices. As you move up towards 3KHz, you are splicing near the middle of the image and it is very difficult to hide that (we'll presume there is no loss of image so you are working on the same frame). Your technique (equipment/skill) may be able to keep it to a minimum but it is still there for all to see.

Time alignment in the crossover, again, is trying to reduce inherent damage done by the speaker design. Getting drivers in the same physical space is difficult, if not impossible. It is one reason coaxial drivers have an edge...the only difference is z-axis, which is a simple time offset. With other systems, you have vertical or even (gasp) horizontal displacement too. You'll never get that right across all seated areas...you can, at best, get it better and the severity is dependent on speaker design and how tight they can get the drivers to each other. Again, a large 2-way (e.g. JBL 4675, Altec A5, Altec A4) "hides" this problem down at 500Hz where you won't notice it in dialog. As you move to smaller horns, the 2-way crossover point moves up and once you get above 1KHz...you are really moving heavily into a dialog area and causing problems. A 2-way speaker crossing over at 500Hz is going to beat about 95% of the 3-ways speakers out there for dialog. The only real downsides to a 500Hz 2-way system is the horn is going to be large if it has pattern control and, depending on the throat/waveguide design, the turbulence happening in the throat.

[Marcel Birgelen]

Steve, I think we generally agree on the subject and your film analogy is a good one. That's also why I generally advocate two-way systems, unless you do have the budget to do it right. I agree on especially the spectrum around 3kHz being important for speech (2kHz to 4kHz is generally considered to be most important for speech). I've written some implementations of the Ogg Vorbis encoder and decoder about 15 or so years ago, I've read a lot of theory about what frequency domains are important for speech (and should therefore be less compressed). I have to admit, I've forgotten probably more than half of what I've learned back then, but still...

I also agree on coaxial drivers being a more ideal point-source and therefore reducing some of the timing issues, unfortunately, besides for car audio applications, they're nowadays increasingly rare and are very hard to get right, I guess that's why most speaker designers have since moved away from them.

Still, three-way systems can be done right. Good speaker design at least keeps the vertical alignment of drivers in-line, so the time domain issues are limited and good speaker design also involves matching drivers and making sure human speech gets reproduced as naturally as possible.

But my point was, a good 3-way system will have less trouble in rendering a complex soundtrack, one with many overlapping sounds in the same channel. Now, many mixes keep the front center somewhat free for dialog and don't very often mix it with music or other sound effects, but the exception would be a mono mix. And it's especially with difficult mono mixes that I think, a good 3-way system actually helps with dialog intelligibility. I think A Clockwork Orange is a nice reference of how well your system can handle a pretty difficult mono mix.

The conversation would go like this...

First person: “What did he say?”

Second person: “Bollocks!”

First: “You don’t have to be rude! Just tell me what he said!”

Second: “Bollocks! He said, ‘Bollocks.”

Sounds like your average British movie, really.

[Steve Guttag]

Marcel,

The reason I harp on the 3-way thing is not that their can't be a good 3-way...I even referenced that I use the QSC SC-423 and, by extension the SC-424 as a 4-way. It is that the overwhelming majority of 3-way speakers I see (and more importantly hear) used in the cinema industry are just not that good and notably inferior to the 2-ways that cost less and quite probably worse than speakers they replaced. There is this notion that just because you have three distinct drivers to cover their frequency ranges comfortably, there is an inherent superiority to them. They CAN be good but most of them are not. Then again, most 2-ways now don't use a large enough horn to get the crossover low enough too...they still, typically, crossover at or below 1000Hz except for the very small. Still anything higher than 500Hz is detrimental. The problem is, even crappy speakers make sound.

There is one 3-way people seem to like that I just don't get. Yeah, it's shallow but it sure hasn't impressed me yet. I think people are impressed by its size/price and that it can make sound.

[Marcel Birgelen]

"The problem is, even crappy speakers make sound."
I'm going to steal that quote.

But... You don't need to convince me. I'm not "impressed by 3-way" just because it's "3-way". It's not simply "more is better, because it's more". The most ideal speaker system is probably a one-way system, with a single driver that can handle the whole audible frequency domain and has a flat response across that same domain. The problem obviously is, nobody has yet been able to design this kind of driver, at least not for the output levels we're looking at, and it's unlikely somebody will do this in the near future. The more power a speaker needs to deliver, the bigger the problem will get, as the inertial mass of the driver needed for the lower frequencies will work against you at the higher frequencies.

I do think the SC-423 is a good speaker, at least in a tri-amped setup, but have no experience with the SC-424 yet. If I remember correctly, the crossover between mid and high for the SC-423 is somewhere around 1.6 kHz, which puts it at the underside of the "danger zone" for dialog.

By the way, this whole discussion reminded me about this paper about crossovers, which really is worth a read if you have nothing better to do.

[Bruce Cloutier]

You guys can argue off-line. The fact remains that it is a challenge to achieve the perfect audio reproduction and then, what is "perfect" is highly dependent on the individual. No one really knows what it would sound like to be running from a brigade of alien ships blasting the hell out of the city and your backside, let alone if you would be able to understand anything anyone was screaming during such a scene. The producer might even have intended the dialog to be difficult to hear. Sometimes even when someone physically stands before you in a scene comprised of your own back yard, you can't seem to comprehend the dialog.

Aside from the audio in such a situation your senses are really missing a lot more of the show. No I don't need my seat to be shaking and water spritzing in my face. But yet there would be vibrations beyond imagination. And, likely a huge odor. Thankfully, theaters have not tried to stimulate your sense of smell... at least not in any serious way. So... nothing is perfect.

I agree that a room can be terrible and difficult to work with. No two seats likely share identical audio characteristics ever and at all points in time. Even then would a room creating the perfect War of The Worlds audio experience really also get the light and sparse dialog from a romantic candlelight dinner set in the garden after dark with background breezes and perhaps a faint thunderstorm perfect? Besides, you might want to feel a little air movement to go along with that scene but not in that stagnant popcorn flavored air mass you are in. Again... not perfect.

I am not in any way saying that you shouldn't try to get it perfect. It is probably futile. But yet your gratification is important even if the patrons don't seem to all get it.

Your ear converts sounds into the frequency spectrum performing like a real time Fourier Transform and then converting that into an array of analog signals that then somewhere in your head you are continuously analyzing with complex pattern recognition. No two ears are wired the same, hearing impaired or otherwise. The note A (say 440Hz I think is one) played on the piano is not just a single frequency but a sequence of harmonics of varying amplitudes. That same A played on a violin or any number of wind instruments you would immediately recognize as being played by that instrument. Recording that perfectly is enough of a challenge. Trying to then reproduce the original sound perfectly is even more difficult and yet you are still hopefully able to recognize the sound.

I was amazed to learn that the low bass you hear from the organ in the cathedral (16.7Hz or something like that) is a note on the keyboard but a sound that the organ cannot physically reproduce. The pipe wouldn't fit in the building. Yet you hear it. I at first thought well you beat two frequencies differing by 16.7Hz against each other and then you have it. I am told that they recreate the balance of the spectrum for that note and your mind imagines the fundamental frequency which isn't there. But, that, might work no matter what seat you are in.

So I am not sure anyone's speakers are perfect at whatever price. But I would admit some seem much better than others. But does the amount of design effort you put in and the amount of money you expend somehow in your mind actually recreate the missing gratification for you in the sound? I wonder if patrons who think they are experiencing high technology audio might not actually experience a higher level of satisfaction in the soundtrack? I bet Dolby thinks that, in pushing the Atmos name in your face. Seems like an opportunity for a double-blind test. Although, one should not then be chastised for misrepresenting the theater's audio presentation.

Uh... We need to get theaters back open and people comfortable in coming. Seems to be a priority.

[Steve Guttag]

Bruce, most of what Marcel and I have been bickering about is dialog reproduction, which is a subset of sound and has its particular challenges aside from wide-spectrum audio, let alone some pipe organ. When I tested my subwoofers that can go down to 18Hz...I had to go outside of my house to hear (well...feel) it.

I don't think anyone here is talking about "perfect" audio...just trying to improve difficult situations and/or reproducing what the movie's creators put into the mix.

[Marcel Birgelen]

You guys can argue off-line.

So you're entitled to your opinion and post it here, but "we" should take it off-line? What's the point of an on-line discussion forum again? I think the only thing you can blame any of us is that the discussion went off-topic, but so did your post.

The fact remains that it is a challenge to achieve the perfect audio reproduction and then, what is "perfect" is highly dependent on the individual. No one really knows what it would sound like to be running from a brigade of alien ships blasting the hell out of the city and your backside, let alone if you would be able to understand anything anyone was screaming during such a scene. The producer might even have intended the dialog to be difficult to hear. Sometimes even when someone physically stands before you in a scene comprised of your own back yard, you can't seem to comprehend the dialog.

There is a difference between recording and playback. The whole idea of a good cinema sound system is to reproduce the sound the way the moviemaker intended it to be. So, an ideal reproduction sounds identical to when it was mastered. While we all know this isn't technically possible, the idea is to get as close as possible within what current technology and budget constraints lets us do.

Aside from the audio in such a situation your senses are really missing a lot more of the show. No I don't need my seat to be shaking and water spritzing in my face. But yet there would be vibrations beyond imagination. And, likely a huge odor. Thankfully, theaters have not tried to stimulate your sense of smell... at least not in any serious way. So... nothing is perfect.

There is barely anything theaters haven't tried before.

I agree that a room can be terrible and difficult to work with. No two seats likely share identical audio characteristics ever and at all points in time. Even then would a room creating the perfect War of The Worlds audio experience really also get the light and sparse dialog from a romantic candlelight dinner set in the garden after dark with background breezes and perhaps a faint thunderstorm perfect? Besides, you might want to feel a little air movement to go along with that scene but not in that stagnant popcorn flavored air mass you are in. Again... not perfect.

If you want wind and water in your face and your seat to shake, we've got 4DX for you right there. But again, a modern audio system should both be able to render that candle light dinner scene perfectly well, just as the follow-up scene, when the aliens attack and blow it all into atom-sized pieces. See, the whole idea of a good cinematic audio system is that it can do both: it can be subtle when it needs to be, but blast on all channels when the director decides all hell broke loose.

I am not in any way saying that you shouldn't try to get it perfect. It is probably futile. But yet your gratification is important even if the patrons don't seem to all get it.

Look, the excuse "the public doesn't get it anyway" is how we end with rendered-down, 13-in-a-dozen, mediocre products.

Your ear converts sounds into the frequency spectrum performing like a real time Fourier Transform and then converting that into an array of analog signals that then somewhere in your head you are continuously analyzing with complex pattern recognition. No two ears are wired the same, hearing impaired or otherwise. The note A (say 440Hz I think is one) played on the piano is not just a single frequency but a sequence of harmonics of varying amplitudes. That same A played on a violin or any number of wind instruments you would immediately recognize as being played by that instrument. Recording that perfectly is enough of a challenge. Trying to then reproduce the original sound perfectly is even more difficult and yet you are still hopefully able to recognize the sound.

That's the whole point of the discussion between e.g a 2-way system and a 3-way system. A driver needs to not only reproduce the primary frequency, but also a ton of harmonics layered on top of it. Designing a driver that can reproduce a simple sine wave is easy, designing a driver that can render complex frequency pattern for a large domain AND considerable output power is something else entirely. By splitting a speaker into multiple drivers that each are responsible for their own frequency domain, you decrease the amount of work an individual driver has to do. The result is, often, a clearer rendition of the original recording. The problem here though is that you need crossovers between those drivers. In many 3-way systems, one of the crossovers spans quite an important part of the most important frequency domain for speech. A not-so-good crossover can therefore have a big impact on how speech is rendered.

I was amazed to learn that the low bass you hear from the organ in the cathedral (16.7Hz or something like that) is a note on the keyboard but a sound that the organ cannot physically reproduce. The pipe wouldn't fit in the building. Yet you hear it. I at first thought well you beat two frequencies differing by 16.7Hz against each other and then you have it. I am told that they recreate the balance of the spectrum for that note and your mind imagines the fundamental frequency which isn't there. But, that, might work no matter what seat you are in.

Keep in mind that most humans won't be able to hear any note at 16.7Hz at all. A pipe reproducing such a frequency would need to be about 66ft tall. The trick may be to reproduce several harmonics, which would otherwise be there. Your mind can be tricked into thinking you're hearing that actual note.

So I am not sure anyone's speakers are perfect at whatever price. But I would admit some seem much better than others. But does the amount of design effort you put in and the amount of money you expend somehow in your mind actually recreate the missing gratification for you in the sound? I wonder if patrons who think they are experiencing high technology audio might not actually experience a higher level of satisfaction in the soundtrack? I bet Dolby thinks that, in pushing the Atmos name in your face. Seems like an opportunity for a double-blind test. Although, one should not then be chastised for misrepresenting the theater's audio presentation.

Nobodies speakers will ever be perfect, the whole idea is to work with the stuff you got to make it as good as possible.

Professional audio is expensive, but not as expensive as many think you need to make it. Most professionals in the field know about the subjectivity of "sound", yet there is practically nothing in sound that can't be measured. No speaker is perfect and no system is perfect, but with careful design and some modern "electronic voodoo" like room correction, we can get pretty close. Still, the excuse: The audience won't hear when we fuck up anyway is no excuse at all.

Mind you, I'm personally anything but an "audiophile", I don't trust my ears to be the final answer to anything, but I do trust my measurement instruments.

Uh... We need to get theaters back open and people comfortable in coming. Seems to be a priority.

Right now, almost all theaters that can realistically be open, are open. The faith of those theaters, unfortunately, isn't really in my hands and I doubt it's in anybodies hands around here. So, right now, there's probably not much we can do, besides come up, maybe, with creative new ideas. But creative new ideas don't come up in a vacuum, they're often the result of discussions... just like those right here...

[Steve Guttag]

Wow...talk about your bollocks!

[Marcel Birgelen]

The subtitles aren't working, what are you saying?

[Bruce Cloutier]

Haha. I guess I know now how to get quoted. We're just off topic I think and good (and impressive) points were made about speaker systems. It just seems that any ongoing 'discussion' over the use of 3-way vs. 2-way or any other combination of speakers deserves its own topic. And, frankly, nobody really knows how the moviemaker intended it to be. That's not to mention that there are a lot of people involved in making a movie and any single piece of it. I wonder if they can even tell if the way it is rendered in a particular house is exactly as they envisioned it? They might not even know who among themselves dictates the official opinion as to how it will be. And... my "opinion", is that it is too subjective to form any specific opinion about any part of the presentation. Personally, I look to the plot, the quality of the acting and the creativity in telling the story. Maybe that is just me. Most of the special effects to me seem to be added just to top the previous movie and might not add much to the story. You know, there needs to be a sex scene. Why, because it sells. But, what do I know? Just ignore me. I am used to it.

[Marcel Birgelen]

Yeah, we're pretty far off-topic, but the thing with splitting off a new topic is that those often end-up nowhere.

We can't look into the mind of a moviemaker, but we do have the movie as a representation of what hopefully* comes close to that vision. This movie was mixed in a mixing stage, according to certain specifications and guidelines. If we recreate those conditions of this mixing stage as well as we can inside the theater, the representation should come very close to what the director intended it to be.

The reality is, sure, every theater is different, even two identical twins in a multiplex are somewhat different. Your seating position has an impact on how you perceive not just the sound mix, but the entire movie. The amount of people eating popcorn makes an impact, the amount of people refusing to shut their mouth too...

How any individual perceives a certain sound mix is obviously extremely subjective. The best sound system in the world won't save a bad sound mix and it will also not fix bad writing or bad acting. I think you can generally describe a good sound system as a system you don't specifically "hear", as in, it doesn't get in your way into enjoying the movie. A bad sound system is one, where its flaws and shortcomings distract you from the movie. Now, what this entails is also pretty subjective, as you say, some patrons may not care, but quite a few will care and a lot will care, will not tell you anything but simply won't show up again.

* There may be quite a few movies out there that where actually saved by not matching entirely with their creator's vision, the original Star Wars trilogy seems to be such an example.

[Randy Stankey]

I agree. The perception of how well an audio system works to reproduce sound that is pleasing to the ear is almost entirely perspectival.

What sounds good to one person might not sound good to another for a whole host of reasons. It could be because of physiological differences in each person's hearing. It could be because of what one person prefers to hear versus another. It could even be a perceptual difference.

You know the old koan... If two people are looking at the very same American Flag, at the same time, in the same place, how do we really know that the colors one person sees as "red," "white" and "blue" appear the same to the other? The second person might actually see "green," "black," and "yellow" but nobody will ever really know the difference.

The same goes with sound. Even though two people seem to agree that they are hearing, for instance, the sound of a pipe organ, how do we know that both people perceive the sound the same way?