Buyers' Guide to Encoding Appliances 2019
Software encoding, especially in the cloud, is all the rage these days. Not only has cloud-based encoding reached feature-set parity with older local (on-premises) hardware-based encoders, but the speed of data networks now enables uploading uncompressed or slightly compressed video files to cloud-based transcoders.
Get instant access to our 2019 Sourcebook. Register for free to download the entire issue right now!
Yet there are still very viable reasons to consider using hardware-based encoding, a number of which we will cover in this Buyers’ Guide. First, we’ll take into consideration the extremes of hardware-based encoding: ultra-portable and ultra-dense.
Ultra-Portable Encoding:The ultra-portable encoding category contains two types of encoders: studio production and field production. In both instances, the encoder has a very small form factor, whether it be a dedicated single-board computer (SBC) or a purpose-made encoder in a ruggedized casing.
Ultra-Dense Encoding:The ultra-dense category focuses primarily on rackmounted encoding, whether it be in a production environment (where the output of each camera is recorded simultaneously in a common codec, format, and resolution) or in a data center (where content is received as an IP stream and then re-encoded or repackaged for streaming delivery).
With these categories in mind, what are some of the factors that Streaming Media readers should consider when comparing encoding appliances with each other or comparing them against software-based encoding solutions?
One argument for cloud-based encoding is the ability to forego capital expenses (CapEx) and only use what computing is needed as part of ongoing operations (OpEx). But one of the last costly CapEx seems to be the most overlooked: equipment to meet bandwidth requirements.
Over the past few years, when my lab has run comparison tests of cloud-versus-hardware encoding— centered on uploading a full-sized video asset for cloud-based encoding, versus simply transcoding the same asset on a local device—we’ve often run into the issue of customers having bandwidth that’s limited enough that it skews the results more toward the continued use of on-premises encoders than cloud-based transcoding.
While it might be easy to dismiss this by noting that theoretical bandwidth and actual bandwidth differ, too often we’ve found that our clients haven’t allocated the CapEx to keep the DSL router or cable modem up to modern specs.
Recent advancements in cable provider data networks, including the expanding use of fiber-to-the-premises, mean that it’s now time to consider upgrading on-premises devices to meet the Data Over Cable Service Interface Specifications (DOCSIS) 3.1 level of connectivity. While the DOCSIS 3.0 specification has been around since 2006, it’s been only since 2016 that full-duplex DOCSIS 3.1 has enabled data rates higher than 1Gbps (gigabit per second) in both upstream and downstream configurations. And while 1Gbps may sound like overkill for your local venue, the actual data rate is often 1/10 or less of the stated throughput, in no small part based on the way cable service providers use a shared network between units in multi-tenant office or housing complexes.
Since most cable providers don’t offer DOCSIS 3.1 equipment, it falls to venue owners or content creators to upgrade their own on-premises hardware. But doing so may significantly change the dynamics around hardware encoding appliances for live events, including the ability to push out one or more ultra-high-definition (UHD) 4K signals.
The Need for Speed
What if software-based encoding isn’t fast enough for your needs? An example of this from an ultra-dense encoding standpoint is the need to convert multiple UHD 4K signals into streaming formats with web-friendly data rates.
Using a standard laptop or even a robust desktop computer with a single software package provides a cost-effective solution. But that approach will take an inordinate amount of time to convert even just one UHD 4K signal to one streaming output of a particular data rate, resolution, frame rate, and color depth.
On the flip side, the use of blade computing with software is certainly feasible, but the cost associated with populating the multi-blade server with processors, RAM, storage, and software can be daunting. Then there’s the form factor of a blade server, which necessitates finding a place to put the server in a data rack, ideally outside the earshot of the production environment.
What alternatives exist in the middle ground? Haivision announced one such alternative, the KB Max, at the National Association of Broadcasters (NAB) show last year. The KB Max box functions like an either-or solution, offering either localized transcoding “by encoding full adaptive-bitrate cascades in HEVC or H.264” for up to four simultaneous 1080p encodes, or acting as a transmitter of “high-quality single bitrate streams up to 4K UHD to all major platforms for cloud transcoding.”
Many of today's live video encoding solutions require extensive compute resources, limiting the ability of live streaming business models to economically scale. This article will introduce a new real-time video encoding solution, combining the performance of System-on-Chip (SoC) encoding, with innovations from NVMe-based cloud infrastructure, which together provides an economical and high quality solution to deliver encoding at scale for live video streaming.
Encoding and transcoding are at the heart of every OTT and online video workflow. The first article in this three-part series gives an overview of the technologies and a look at three major players in the space: Harmonic, AWS Elemental, and Telestream.
Now widely available, per-title encoding makes whatever codec publishers are already using more efficient by creating a custom optimized encoding ladder.
Companies and Suppliers Mentioned