The ATSC 3.0 Audio Standard: Tim Caroll, VP Linear Acoustic Has A View

Tim Carroll, a member of the ATSC committee deciding an audio standard for 3.0, offers an insider’s view of the selection process.

Since 2002, Tim Carroll has been the chief technical officer at The Telos Alliance and founder of Linear Acoustic. He also sits on the audio committee deciding the ATSC 3.0 audio standard, which is now in a tight race between Dolby Labs (AC-4) and Fraunhofer, Technicolor and Qualcomm (MPEG-H) to become the next generation audio system. The ATSC says a decision is due early next year.

In his earlier days, Carroll worked as the product manager for the Pro Audio Division of Dolby Laboratories in San Francisco. While at Dolby, he contributed to the development of Dolby Digital (AC-3) and the Dolby E encoding systems for DVD and high-definition television (HDTV) applications.

Today, Carroll’s company is a licensee of both technologies and he said he respects the technical quality of both audio codecs.

Though sworn to secrecy as to the deliberations of the ATSC 3.0 committee working to choose an audio standard, Carroll has been concerned about the tone of recent articles about the delays in the committee choosing an audio standard. He spoke to Frank Beacham at The Broadcast Bridge in an effort to shed a little light on the selection process.

Broadcast Bridge: What concerns you about the public’s perception of the deliberations over the ATSC 3.0 audio standard?

Carroll: I’m concerned about statements like this is another example of the customer not getting the better of the two technologies because of money-hungry trolls, meaning licensing costs. That’s not fair and is totally inaccurate. It doesn’t really come down to that. What the public sees is different from what a licensee sees.

Outside of the ATSC committee, my company is a licensee of both companies. I have gotten very specific answers and seen very specific breakdowns. I have also had to sign several NDAs (non-disclosure agreements). Telos will be implementing both standards. As a company that supplies products worldwide, it makes sense for us to support both codecs.

But whether the ATSC process has been riddled with delays that I feel could have been avoided, well it is something where potentially billions of dollars are at stake. So the process has to follow its course. It’s not just frustrating for the proponents who have both done a good job with their codecs. I will say that what I have learned as more details have become available is that one side has really innovated from the get-go and they actually have a commercially viable system. It’s the system now available on Blu-rays, Hulu and Netflix. I mean it’s out there.

When I worked at Dolby, I started on the cinema side. Some of it was dealing with classical music. Nothing prepared me for the beatings I would get on the film stages when producers were preparing the samples to make sure the film sounded exactly the way they were mixing it. I totally get it. It’s a tough crowd and Dolby had to develop their technology in that environment.

From there though, the discussions inside the committee’s deliberations became less technical. With a name like Technicolor, they certainly have a history in cinema. But the last time I checked, it was on the visual side of cinema. When you say we’ve been in cinema for over a hundred years, that disingenuous. That’s not exactly true. That’s where it got political. From my company, we will implement both technologies. But (on the committee) side they took it into the mud and the other side stayed on the high ground.

The audio portion of ATSC 3.0 is likely to become a candidate standard in early 2016.

The audio portion of ATSC 3.0 is likely to become a candidate standard in early 2016.

Broadcast Bridge: So how will the winner be decided? Is there an actual up-down vote?

Carroll: No, there is no vote initially. Without getting into what has already been done, I will say that we have made very good, very positive progress that is beneficial to both proponents. You can read into that whatever you want. If Mark Richer wants to get pissed off, let him come and get me. But I think the industry as a whole will be benefited by the unique way that the ATSC is doing this. So while it was stressful, I think the end result will be really better than any of us expected.

The way this works is it begins with a technical group. This is all available in the bylaws. In this case, it’s Technical Group 3. TG3 sits above a bunch of specialist groups. One of them is S34, for the presentation layer. Below S34 are ad hoc groups. Those groups focus on things like video, audio, conceptual model/framework and interactivity. The level of interest really gets focused as you get closer to the ad hoc group, where the experts in that area do the heavy lifting.

The group above will generally say 'OK, you come to a consensus.’ The ad hoc group cannot vote. It’s just a consensus. But I will tell you it has to be a very strong consensus. The ad hoc group will provide to the specialist group their consensus opinion. The specialist group will then provide that to the technical group, which in this case is S34. The actual voting takes place when something is sent out to become candidate standard. That comes from TG3.

Tim Carroll, Vice President/Executive Director at Linear Acoustic.

Tim Carroll, Vice President/Executive Director at Linear Acoustic.

Broadcast Bridge: Has that ad hoc group made a decision yet?

Carroll: I can’t tell you that. But I can say that the decision is two-thirds there.

Broadcast Bridge: Can both of the final choices become standards or does it have to be one or the other?

Carroll: It has to be what the members decide it has to be. The proposal did not say it shall be one and only one. But, remember part of this membership is consumer electronics. Who in consumer electronics would want to implement two different decoders?

Broadcast Bridge: OK, which one costs the most?

Carroll: We have gotten an actual answer from Dolby. From MPEG-H, the information is coming from the Blackstone Group, and it’s only an estimate. So I don’t know. I will tell you that my company, Telos, has gotten pricing from Dolby. We pay a fee for implementing Dolby Digital or Dolby Digital Plus encoding or decoding. We pay Dolby a fee for AC-3. AC-4 will be no additional charge.

Broadcast Bridge: Wait a minute. Are you saying Dolby will not charge any more for the new technology if they win? But you don’t know about MPEG-H? Isn’t that a major determining factor?

Carroll: At the ATSC, we can’t talk about licensing fees until a candidate standard is published. We discuss the technology only. That candidate standard gives the industry the ability to start asking these questions. But because my company is a licensee, we asked the questions and they provided that information to the licensee, not to the committee. But if Dolby wins, there will no additional charge over what people are paying today for AC-3.

Broadcast Bridge: When might this audio selection be made?

Carroll: That I don’t know. But I would say it is imminent. Because of the meeting cycle, I think it will happen early next year. And by ‘done,’ I mean it goes to candidate standard. Because of the way the voting works, that is actually the first vote that takes place.

I can say I’m ending this year tired, but pretty satisfied at this point. We don’t take sides on this. But we certainly don’t want to re-do everything in the U.S. that we spent 20 years helping broadcasters install. Dolby-centric isn’t the right word. Part of my job there years ago was to help people do things in a standardized way — almost a SMPTE kind of situation.

Broadcast Bridge: It certainly sounds like you are giving Dolby the advantage — not only cost-wise but with the comfort zone of broadcasters using their technology.

Carroll: Yes, but the economics are nothing new. They had this same policy when Dolby Digital Plus was released. We got an upgraded licensing agreement that said here’s the new code, but there is no change in the royalty you pay.

And the comfort level is also true. I think that started when we started to referring to these things as systems, rather than codecs. MPEG-H is a solid codec. It’s really good technical work. The guys at Fraunhofer should be proud of themselves. We wouldn’t implement it if its wasn’t good.

But I would say that when you look at it as a system, our experience has been that when we’ve added encoding to our products we’ve never experienced a decoder that had issues with Dolby. They have tested every implementation and that gives us a technical perspective that Dolby stands alone.

I want to be able to call somebody and I want to know I can get somebody on the phone in San Francisco during the Academy Awards. As a licensee from both Fraunhofer and Dolby, I would say there is a big difference between a system and codec. Both companies provide codecs.

Our plans are to implement Fraunhofer’s code and it’s going to be used in other parts of the world. Who knows exactly where it is going to be used? Nothing is off the table at this point, at least until the slightly overweight lady hits the high note sometime next quarter.

You might also like...

Digital Audio: Part 18 - Synchronization

Digital audio interfaces were developed as a way of avoiding generation loss between devices.

Digital Audio: Part 17 - Filters And Stability

The recursive filter has the advantage of using less hardware, but is more complex to understand.

Microphones: Part 7 - Microphones For Stereophony

Once the basic requirements for reproducing sound were in place, the most significant next step was to reproduce to some extent the spatial attributes of sound. Stereophony, using two channels, was the first successful system.

Microphones: Part 6 - Omnidirectional Response In Practice

Having looked at how microphones are supposed to work, here we see that what happens in practice isn’t quite the same because the ideal and the actual are somewhat different.

Digital Audio: Part 16 - Filters, Direct Implementation

There are two approaches to digital filtering. One is to implement the impulse response directly. The other is to use recursion. Here we look at the direct implementation.