BBC Media Solutions Consulting Engineer Steve Daly talks on the workflow planning and technical system design for the world-feed delivery of Live Earth, which involved generating 22 hours of live continuous tapeless broadcasting to 125 countries.

Following our major contribution to the Live 8 concerts in 2005, we secured the contract to manage the World Feed operation for the 2007 Live Earth Global Concerts.

For Live 8, we were able to use our existing fast-turnaround editing facilities, but Live Earth was far more complex and presented a number of challenges which required a new innovative approach.

The event comprised a series of 12 music concerts across all seven continents, with the final output being a 22 hour continuous HD broadcast to 135 countries with an estimated audience reach of 2 billion people.

For an operation of this size and with the requirement to be fully HD, we decided the best approach was a custom-build, in the same way that we would for a major sporting event.

Using our newly-built HD studio, Studio Eight, to provide audio and video presentation facilities, we housed our main server, editorial checking and playout area in the studio lighting gallery and built a highlights editing operation in the adjacent conference rooms.

Key to a complex operation such as this was workflow planning. Based on our experience on live events, we chose to use EVS servers to provide core HD media storage and for playout and turnaround functions. Because of the production team’s need for extremely fast turnaround of highlights and bumpers we decided to use in-server EVS editing capability. Whilst we regularly adopt this approach for sports highlights, it is unusual for us to employ EVS for complex artistic pieces, for which we would normally use conventional non-linear systems - such as Avid and Final Cut Pro.

To facilitate this new approach we designed and delivered a training programme to a pool of conventional non-linear editors to enhance their skill sets in this area and ensure they were comfortable with the tools and workflows.

With a huge amount of content, continually changing running orders and language issues an efficient live-logging solution was essential. We chose to convert our normal fast-turnaround area into a live logging area, dedicating a room to each concert with communications to the venues and coordination centres. We sourced logging staff fluent in the native language of the concert, or used translators as appropriate to aid the logging process.

We looked at a range of software-based logging packages, however none met the particular demands of this event so instead we turned to BBC developers to customise existing web-based running-order management software for use as a live logging tool. Data entered in our logging rooms using existing IT infrastructure was available immediately to directors, video editors, play-in operators, script supervisors and producers.

Additional data, such as TX durations, edit requirements and compliance approval, could be added by authorised staff. A filtered subset of this data was published via the web to be available to the broadcasters in the 135 countries who were involved in the event.

We also provided access to a live running-order management system to facilitate real-time calculations of running order timings to allow advertising and sponsor compliance to be continually assessed. A ‘next 5’ running order was also published to the web to augment a satellite comms circuit and telephone dial-in we provided, giving countdowns to advert breaks.

Although the capabilities of the EVS in-server editing dealt with most edit requirements, some still required more complex tools, so we provided connectivity from our EVS server system to our existing HD Village via newly run fibres, extending control of servers and logging data feeds to these areas. Material requiring editing could be sent to an editing queue and then published directly back to the EVS servers when complete, with edit information and durations added to the logging system to aid running order management.

For an extremely high-profile broadcast such as this, it was crucial that the system was resilient. The design of our EVS network involved multiple redundant ingests of key feeds and a tiered series of backup plans to cope with various failure modes. We allowed time during the latter stages of the build for validation and load-testing by EVS engineers. As proof of this, the system actually coped with hardware failure on the main TX server shortly before the show with no impact to transmission.

The evolving nature of the programme, with concerts being added and dropped throughout the final days leading up to the event, required great flexibility and adaptability in the system design. For example, in addition to the multiple main outputs (some provided via our Snell and Wilcox HD Alchemist standards converters) and planned backhaul and MSN streaming outputs, we were unexpectedly called upon to deliver a number of other live feeds to world broadcasters on the day, something that our contingency system capacity allowed us to provide.

Immediately after the main 22 hour transmission we were required to transmit a 90 minutes highlights package and once again turned to a combination of live presentation and in-server editing to deliver this, including the need to transmit clean and dirty versions of this edit.

We were responsible for coordinating the global communications operation, including the integration of hired event communications systems into existing studio and post production facilities, and liaising with the companies providing circuits and satellite bookings.

With limited time for planning, procurement and delivery we were required to design systems and workflows with great flexibility and resilience within tight cost constraints.

In the lead-up to the programme we were also responsible for final editing of insert packages and graphics elements sourced from a variety of formats, for preloading onto our EVS playout network, across multiple servers for resilience. On the day we became involved in delivering additional editing and repackaging for the Wembley concert operation.

We were responsible for a number of other functions such as building and operating an HD record-bank to preserve all the concert footage and programme outputs. These tape recordings formed part of our contingency plans with tape-based concert footage always being cued and available for immediate live transmission to cover for recovery time required in case of server failure.

We were involved in numerous other aspects of the project such as sourcing operational and production staff, coordinating runners and providing catering and accommodation for production and operational staff.

Our experience gained on this project assisted the adoption of new workflows on other events and helped steer the planning and delivery of other large-scale operations such as the 2008 Beijing Olympics.

back to top