Sound engineer Brian McCarty, Harman's Director of Acoustic Research Sean Olive, and binaural-recording expert Bob Schulein discuss the recent AES conference on the future of audio-entertainment technology that Brian and Sean co-chaired and Bob participated in extensively. Topics include acoustics, the X-Curve, dialog intelligibility, bass management in cinemas and dubbing stages, the headphone target EQ curve, answers to chat-room questions, and more.
-
Our native mobile app has a new name: Fora Communities. Learn more.
The Future of Audio-Entertainment Technology
Attachments
Scott.. great podcast...
I have a couple of point/clarifications/comments and personal opinions on what was said.
1. Even if you look at the findings of the recent SMPTE report, the rooms in Hollywood aren't as radically off as Mr. McCarty suggests.. the industry, while we could certainly use more consistency and new measuring techniques, isn't in as sorry a state as he describes. I won't argue that consistency between dub stages and theaters isn't ideal. In the past 6 months I've mixed in 5 different rooms at two of the best facilities in the world. There are differences, to be sure. But it doesn't end up being an "oh my god, what happened to the mix." type of situation. The only place we find major issues is in the subs and low frequency response of the mains (under 35Hz) in my experience....
Look back at Brian Vessa's podcast for some good counter points..
No one is mixing wildly off the x-curve due to client or producer demands (at least where I have worked over the last 15 years...) That just doesn't happen as Mr. McCarty suggests.. as a matter of fact, Dolby won't certify a room if it doesn't meet the x-curve within a reasonable amount, and even at this point in time (with 35mm dying quickly off) Dolby is involved with almost every theatrical print master in town.. that just hasn't been my experience.
And with that being said, I think the proof is in the pudding... we, as a community, are making some fairly spectacular sounding films...
2. Many near field mixes are done with bass management.
3. While Mr. McCarty keeps banging the drum for woven screens, that is a utopian solution which isn't going to happen and time soon IMO. It's hard enough to get exhibitors to do anything regarding presentation that costs money.. asking them to replace screens without any increase in revenue isn't going to work. We need to find better solutions to the concerns and issues raised.
I have a couple of point/clarifications/comments and personal opinions on what was said.
1. Even if you look at the findings of the recent SMPTE report, the rooms in Hollywood aren't as radically off as Mr. McCarty suggests.. the industry, while we could certainly use more consistency and new measuring techniques, isn't in as sorry a state as he describes. I won't argue that consistency between dub stages and theaters isn't ideal. In the past 6 months I've mixed in 5 different rooms at two of the best facilities in the world. There are differences, to be sure. But it doesn't end up being an "oh my god, what happened to the mix." type of situation. The only place we find major issues is in the subs and low frequency response of the mains (under 35Hz) in my experience....
Look back at Brian Vessa's podcast for some good counter points..
No one is mixing wildly off the x-curve due to client or producer demands (at least where I have worked over the last 15 years...) That just doesn't happen as Mr. McCarty suggests.. as a matter of fact, Dolby won't certify a room if it doesn't meet the x-curve within a reasonable amount, and even at this point in time (with 35mm dying quickly off) Dolby is involved with almost every theatrical print master in town.. that just hasn't been my experience.
And with that being said, I think the proof is in the pudding... we, as a community, are making some fairly spectacular sounding films...
2. Many near field mixes are done with bass management.
3. While Mr. McCarty keeps banging the drum for woven screens, that is a utopian solution which isn't going to happen and time soon IMO. It's hard enough to get exhibitors to do anything regarding presentation that costs money.. asking them to replace screens without any increase in revenue isn't going to work. We need to find better solutions to the concerns and issues raised.
Joined
·
2 Posts
Again congratulations on a great show, it always amazes me that you consider yourself a mere geek when in fact you and your guests are REALLY on top of things. (HPA, SMPTE, AES. to name a few). I am writing today because a question was asked and never answered. Scott asked why do we have to boost the LFE by 10 dB? Hopefully my answer will clarify some misconceptions.
The reason is pretty easy … way back when they where experimenting, the LFE was 1 to 1 with the rest of the loudspeakers, but they noticed that to give the - what I like to call "emotional impact" - the "printed to mag signal was overloading. The Fletcher-Munson equal-loudnness contours paper (1933) tells us we need a LOT of BASS to get the "feeling" of even frequency response. So it was decided to boost the monitoring level of the LFE by 10 dB SPL, so in effect the printed signal was lowered by 10 dB hence well below distortion levels. It assumes of course that the whole chain follows.
On another note, the reason why it's not necessary to have bass management in the theatres is because the behind the screen loudspeakers (L, C, R) are full frequency. There is no need to compensate for the low frequency capability of the loudspeaker (typically with a single or double 15 inch woofer), contrary to smaller venues (small studio and homes) where the real estate can't justify refrigerator size loudspeakers. Hence the introduction of bass management, with smaller loudspeakers we are able to "re-direct" the very low bass content from the main loudspeakers to the sub-woofer. (noticed I didn't write LFE).
Hope it sheds a bit of light on the subject,
cheers,
R
The reason is pretty easy … way back when they where experimenting, the LFE was 1 to 1 with the rest of the loudspeakers, but they noticed that to give the - what I like to call "emotional impact" - the "printed to mag signal was overloading. The Fletcher-Munson equal-loudnness contours paper (1933) tells us we need a LOT of BASS to get the "feeling" of even frequency response. So it was decided to boost the monitoring level of the LFE by 10 dB SPL, so in effect the printed signal was lowered by 10 dB hence well below distortion levels. It assumes of course that the whole chain follows.
On another note, the reason why it's not necessary to have bass management in the theatres is because the behind the screen loudspeakers (L, C, R) are full frequency. There is no need to compensate for the low frequency capability of the loudspeaker (typically with a single or double 15 inch woofer), contrary to smaller venues (small studio and homes) where the real estate can't justify refrigerator size loudspeakers. Hence the introduction of bass management, with smaller loudspeakers we are able to "re-direct" the very low bass content from the main loudspeakers to the sub-woofer. (noticed I didn't write LFE).
Hope it sheds a bit of light on the subject,
cheers,
R
I don't want to argue with any of you on that, just saying after reading the report I got the impression that they more or less followed the specifications, contrary to the conclusions made by the authors.
Certainly.
Of the few newer films I have seen lately, I found them all to have entertaining, good sound, and I also appreciate that soundtracks actually are different.
To try to make a shift in the right direction, in a time where interest for sound and sound quality may seem on a decline, I think it is beneficial to point out what is good and tell and even better - demonstrate - to as many people as possible, what a great expreience movie sound is.
First, thanks to Scott Wilkinson for making this episode, with those very knowledgeable heavy-weights in the industry.
The episode covers so much, it is of course only possible to briefly enter each topic in a one hour discourse.
The one thing I want to comment, is the dialogue intelligibility issue.
I see this mentioned occasionally - dialogue is hard to understand, clarity is bad, often it is blamed on too low dialogue sound level.
Since I myself never experience those issues, at least not in a similar way, I suspect something must be wrong.
Do I have better hearing (golden ears..), do I listen louder than everyone else, is my set-up so much better?
We can rule out the golden ears, so then we are left with playback level and set-up, i.e. speakers and room and configuration.
As for level, I find dialogue to be equally intelligible also at lower playback levels, for normal soundtracks at least down to -30dB.
This is for typical movie soundtracks; tv documentaries, news and so, are usually still good at much lower level as well - they are mixed differently, as they should be, not meant for loud playback volume.
However, if the dialogue is too loud, that can cause problems - on some movies dialogue is too loud when played at 0dB, and sounds better at lower playback, turning it down to -10dB can actually improve intelligibility.
If it gets too loud, the ears get overloaded, it does not sound pleasant, like someone yelling right in to your ears, it simply is too loud.
But then just turn it down, right?
The problem with that is that you also turn everything else down, and loose all the power and impact and tactile feel from other sounds and especially sound effects.
Too many soundtracks have too loud dialogue, and newer films have tended to be worse, though some of the recent films I have watched have actually been quite good.
Then we are left with set-up - speakers, room acoustics, configuration adjustments.
To achieve decent dialogue, one can assume it is required to have a reasonable flat frequency response through the midrange, and early reflections and reverb must be controlled.
This should not be an impossible task to achieve.
Would be interesting to see if it is possible to see a common denominator for those situations where bad dialogue is experienced, then it may be possible to come up with some simple rules for system implementation that could prevent such issues.
Then the producers don't need to make the dialogue yelling loud anymore, and I can turn up my system to experience that addictive, powerful sound.
The episode covers so much, it is of course only possible to briefly enter each topic in a one hour discourse.
The one thing I want to comment, is the dialogue intelligibility issue.
I see this mentioned occasionally - dialogue is hard to understand, clarity is bad, often it is blamed on too low dialogue sound level.
Since I myself never experience those issues, at least not in a similar way, I suspect something must be wrong.
Do I have better hearing (golden ears..), do I listen louder than everyone else, is my set-up so much better?
We can rule out the golden ears, so then we are left with playback level and set-up, i.e. speakers and room and configuration.
As for level, I find dialogue to be equally intelligible also at lower playback levels, for normal soundtracks at least down to -30dB.
This is for typical movie soundtracks; tv documentaries, news and so, are usually still good at much lower level as well - they are mixed differently, as they should be, not meant for loud playback volume.
However, if the dialogue is too loud, that can cause problems - on some movies dialogue is too loud when played at 0dB, and sounds better at lower playback, turning it down to -10dB can actually improve intelligibility.
If it gets too loud, the ears get overloaded, it does not sound pleasant, like someone yelling right in to your ears, it simply is too loud.
But then just turn it down, right?
The problem with that is that you also turn everything else down, and loose all the power and impact and tactile feel from other sounds and especially sound effects.
Too many soundtracks have too loud dialogue, and newer films have tended to be worse, though some of the recent films I have watched have actually been quite good.
Then we are left with set-up - speakers, room acoustics, configuration adjustments.
To achieve decent dialogue, one can assume it is required to have a reasonable flat frequency response through the midrange, and early reflections and reverb must be controlled.
This should not be an impossible task to achieve.
Would be interesting to see if it is possible to see a common denominator for those situations where bad dialogue is experienced, then it may be possible to come up with some simple rules for system implementation that could prevent such issues.
Then the producers don't need to make the dialogue yelling loud anymore, and I can turn up my system to experience that addictive, powerful sound.
Except that even with 15" drivers, we aren't approaching
I am well out of my wheel house so please don't put me on blast 
. I enjoyed the video. It was pretty informative. But I do have some layman questions\observations if anybody doesn't mind. I understand that each dubbing studio may be different. But why does it seem that some movies are done perfectly for bass and SPL levels and some are not. It seems that some movies had very high filters in them that cut a lot, if not all of the bass out. The Man of Steel was one I believe. The 47 Ronin was another. Then you have movies where the levels are just so high that it creates distortion and clipping in the sound track. The Immortals, Tron, and I think Star Trek: Into the Darkness fall into that category. I guess my question is why?
How come people just don't mix movies with those things in mind(full range, reasonable levels etc....) I know I am layman and I don't want to rip on anybody who is in the industry who may be on this site. But when you have movies like Oblivion, War of The Worlds. Edge of Tomorrow, Olympus Has Fallen and whole bunch of other films that have seemed to get it "right" in the home environment, I guess I am wondering why there are the others that fail. Now, I know "right" is subjective to me and me alone. But I am just wondering why there are movies that have not translated well. I am probably over simplifying the process and if I am, I am sorry. It just that we spend money on our systems to try and get movies sounding great. I hope I did not come off like an ass with this post. I am just very curious about this aspect of movies that's all.
. I enjoyed the video. It was pretty informative. But I do have some layman questions\observations if anybody doesn't mind. I understand that each dubbing studio may be different. But why does it seem that some movies are done perfectly for bass and SPL levels and some are not. It seems that some movies had very high filters in them that cut a lot, if not all of the bass out. The Man of Steel was one I believe. The 47 Ronin was another. Then you have movies where the levels are just so high that it creates distortion and clipping in the sound track. The Immortals, Tron, and I think Star Trek: Into the Darkness fall into that category. I guess my question is why? How come people just don't mix movies with those things in mind(full range, reasonable levels etc....) I know I am layman and I don't want to rip on anybody who is in the industry who may be on this site. But when you have movies like Oblivion, War of The Worlds. Edge of Tomorrow, Olympus Has Fallen and whole bunch of other films that have seemed to get it "right" in the home environment, I guess I am wondering why there are the others that fail. Now, I know "right" is subjective to me and me alone. But I am just wondering why there are movies that have not translated well. I am probably over simplifying the process and if I am, I am sorry
. It just that we spend money on our systems to try and get movies sounding great. I hope I did not come off like an ass with this post. I am just very curious about this aspect of movies that's all.I would really appreciate thatMike.... I will give you a detailed response when I'm not mixing
.
Joined
·
936 Posts
Scott,
Utopianemo(Utopia-Nemo) here. I thoroughly enjoyed this episode; not least because it's the first one I've been able to watch from the chat room. So much covered!
I have to restate my (respectful) doubts that the real-world matches up with Harman's listening curve. a 230 some-odd person test is respectable but in no way could we consider something on that scale to be conclusive. My own experience as a teenager in the 90's, listening to music with other teenagers("It's all about that bass"), to discussing the original Beats headphone characteristics with non-audiophile youth(there's a reason they were tuned that way), to comparing the low and high end frequency spectrum of rap/r&b/pop directed at today's youth with music aimed at the same target group 25 years ago.....this experience tells me that American youth like a similar curve to what Sean Olive described, only MUCH more so.
Moreover, if one looks at the patterns of various genres of music in this country, as well as comparing music production in different parts of the world, it's obvious that there are general differences in how different types of music(generally correlating to age groups or types of people who prefer those types) and different cultures tend to like their music to sound. Generally, Japanese-produced pop tends to have much less bass and much more emphasis on crisp high end sonics. British pop tends to have a slightly more pronounced midrange than our pop tends to be(or maybe it's that the highs and lows aren't as pronounced). And Jazz and Classical tend to have a more balanced sound, Rap and R&B and Pop tend to have waay overemphasized low ends, and modern country is somewhere in between.
These are generalizations but they point to the fact that different groups of people have significantly varied tastes in the types of sonic signatures they most enjoy.
Utopianemo(Utopia-Nemo) here. I thoroughly enjoyed this episode; not least because it's the first one I've been able to watch from the chat room. So much covered!
I have to restate my (respectful) doubts that the real-world matches up with Harman's listening curve. a 230 some-odd person test is respectable but in no way could we consider something on that scale to be conclusive. My own experience as a teenager in the 90's, listening to music with other teenagers("It's all about that bass"), to discussing the original Beats headphone characteristics with non-audiophile youth(there's a reason they were tuned that way), to comparing the low and high end frequency spectrum of rap/r&b/pop directed at today's youth with music aimed at the same target group 25 years ago.....this experience tells me that American youth like a similar curve to what Sean Olive described, only MUCH more so.
Moreover, if one looks at the patterns of various genres of music in this country, as well as comparing music production in different parts of the world, it's obvious that there are general differences in how different types of music(generally correlating to age groups or types of people who prefer those types) and different cultures tend to like their music to sound. Generally, Japanese-produced pop tends to have much less bass and much more emphasis on crisp high end sonics. British pop tends to have a slightly more pronounced midrange than our pop tends to be(or maybe it's that the highs and lows aren't as pronounced). And Jazz and Classical tend to have a more balanced sound, Rap and R&B and Pop tend to have waay overemphasized low ends, and modern country is somewhere in between.
These are generalizations but they point to the fact that different groups of people have significantly varied tastes in the types of sonic signatures they most enjoy.
Appreciate Brian's insights -
https://www.editorsguild.com/FromTheGuild.cfm?FromTheGuildid=505
https://www.editorsguild.com/FromTheGuild.cfm?FromTheGuildid=505
Joined
·
936 Posts
Quoting the abstract: "Critical listening experience was a significant factor in the listeners’ performance and preferences."
My main idea, which I should have spelled out more clearly, is that listening preferences are not the same across the board and change given geographical, cultural, and temporal variations. Given this, it doesn't make sense to make the Harman curve a standard target--although it may seem to follow the mean preferences of the folks that Harman polled, it's still a subjective target. Why not follow the only objective target--flat response--and let individuals tailor their own systems as they see fit?
1 - 14 of 14 Posts
