Tom Kenny, editorial director of Broadcast Engineering sister publications Mix, EM and Remix recently talked with Tom Sahara, senior director of remote operations and IT at Turner Networks. The two discussed how the Turner Network handles multichannel audio and the importance of good education for operators and engineers.
As senior director of remote operations and IT at Turner Networks, you deal with a lot of 5.1. What are the main roadblocks to its everyday use across the TV band?
The biggest issue we have in 5.1 today is education, getting the workforce that really knows what putting 5.1 together means and creating workflow tools to get the job done. Technically, mixing desks now have 5.1 capability, most of the gear has it as an option — you can click a button or turn on a switch, configure it in a particular way, and you’re there. But it comes down to the person setting up their microphones, setting up their mix in the console. Back at the station, once it comes in, how is it treated? We have to get all of those people in that chain to understand what surround is, to get people to use it the way it was intended and arrive at a consistent, quality product.
And the reason consistency in production is crucial is that there is no consistency among the consumer electronics manufacturers. Every CE manufacturer handles surround in a different way. Many of the TVs you buy come configured in a two-channel enhanced sound mode, which actually renders the sound worse than even a bad mix. It makes a good mix bad and a bad mix worse.
How do you start that education process? At individual networks? At the school level?
That is probably the biggest reason we created the DTV Audio Group. It’s people who have been involved in surround sound since the beginning, many back into the early '90s. Soon after NAB we will have some training courses available so broadcasters and freelance engineers can learn about 5.1 and how to deal with it in the workflows, so they’re not just going by trial and error and trying to survive the show. They actually understand what they are trying to achieve before they get there.
You have to understand, the economy forced many of the broadcasters to cut their training and in-house staff, using freelance staff to fill these positions. There is no real apprenticeship or training program for them to go through at the network. They are basically learning on the job.
To do good surround you have to plan it. You have to have an idea in your head before you even grab the first microphone to pick up the sound you want to capture. That’s what we’re trying to change. We’re trying to give engineers and production companies that knowledge so they don’t have to learn by trial and error. We have people who have been doing this for 20 to 30 years, so learn from them.
Does the same apply to loudness?
Dolby dialnorm is part of the ATSC format, which is something we as broadcasters need to pay attention to. In the past, there was the thought that we could use active dialnorm and the person originating the content would set this number and all would be well because that dialnorm setting is passed on through the chain, eventually makes it to the consumer gear and is matched to the level that the consumer selects on their unit. We live in a far from perfect world, so what ends up is that metadata is rarely set or is not passed on or is misunderstood and is messed up. If you go to the Dolby website and look at the different implementations for each of the broadcast networks, they all do it differently. How can you have a consistent product when everyone does it differently and everyone is shooting at different standards?
Fortunately, in 2009 the Advanced Television Systems Committee got together and said this is a problem. They recently published RPA85, which sets a dialnorm number and says, “If you don’t have the mechanism for setting and maintaining dialnorm, here’s the way you can achieve it.” It’s a recommended practice; those that are capable of using Active and know how to use it effectively, they can continue to use it. For those that don’t, there is a static number to shoot for, which is minus-24. Again, it’s a means of trying to gain a little consistency across the different hands that the signal has to pass through as it makes its way to the consumer.
Assuming engineers learn the techniques, you still have to rely on the networks and individual stations to deliver. And their track record has not been stellar.
To be fair, much of that is because it’s very expensive to take an existing station and turn it into a full digital, surround station. Many television stations were built in the analog days, and while some of them have made the transition to digital, many are still riding on analog infrastructure. Cabling is still old twisted-pair copper, and a lot of the infrastructure doesn’t really support a high-quality 5.1 workflow. So it was easy for management to make a decision that it was too big and complex so they will just go with stereo and upmix everything.
And now here comes Internet delivery. Does this complicate matters?
It certainly is complicating things, and most of that is because material that suits broadcast doesn’t necessarily suit online. Especially longer format. We do still have a lot of work to be done on the distribution of surround as computer files. Many of the streaming players at this moment don’t support 5.1 discrete surround, so what you see out there is typically synthesized programs that upmix from the two-channel. But hopefully in the not too distant future there will be a little more collaboration, some work done to create true 5.1 encoder for streaming online. The connected TV is here. At some point, it won’t matter to the viewer. You just go and get the material you want whether it’s streamed from the Internet or delivered to you over satellite or terrestrial broadcast. It’s just another program that you’re watching on your big screen.