Learning Center
Plans & pricing Sign in
Sign Out

Telestream Episode Engine


Telestream Episode Engine is reviewed.

More Info
  • pg 1
     Telestream Episode Engine                                                                                         By Jan Ozer

         For VP6 and H.264 encoding, Telestream’s Episode
     Engine is fast and produces equal or better quality than                                                                         Figure 1. The Episode Engine
     most other streaming encoders. For those producing                                                                               Admin panel encodes a job via
     shiny optical discs, MPEG-2 performance and quality is                                                                           Split-and-Stitch.

     also quite good. However, Windows Media producers
     should look elsewhere because Engine’s encoding
     speed is slow and the output quality is subpar.
         Episode Engine runs only on Macs and consists of two
     applications and a configuration window that you access
     from the System Preferences window. You create all
     presets in a standard copy of Episode (which is included)
     and then monitor and create encoding jobs in the
     Episode Engine Admin panel.
         Episode Engine comes in two versions: standard              relevant devices such as the iPod and iTunes—and most
     ($3,950) and pro ($8,450). Part of the difference relates to    relevant broadcast formats, though you’ll need the pro
     output formats, with the pro version providing support for      version to output many of these formats. For more details,
     high-end formats such as DNxHD, DVCProHD, XDCam                 check out the Format Support document at www.
     HD, and GFX. The standard version can input most of   
     these but can’t output them.                                       Speaking of presets, I’ve got a lot of minor issues with
         In addition, the pro version adds several features, such    Episode’s settings in general. The language is overly
     as Split-and-Stitch, a High Availability option, and SNMP       complex and obscure to most users, and many settings
     support for monitoring the system remotely over a network.      are flat-out suboptimal. For example, most templates add
     Briefly, Split-and-Stitch is a technique for very fast file     letterboxes to convert SD source material to 4:3 output,
     rendering during which Episode Engine divides each file         which is seldom the desired technique, and key frame
     into multiple components, encodes each separately, and          settings always opt for key frames at scene changes only,
     then stitches them back together (Figure 1).                    whereas most producers use periodic key frames and
         High Availability relates to running multiple Engines in    key frames at scene changes (Figure 2).
     a cluster. That is, you can combine multiple Engines into          These flaws are much more serious in the Episode
     a cluster controlled by a “master” on one of the encoding       desktop product because the typical user may not be
     stations. Once you go beyond five or six computers in a         (and shouldn’t have to be) a compression expert and
     cluster, Telestream recommends that you use a                   should be able to trust the company to choose the most
     dedicated master that doesn’t compress and doesn’t              usable settings. In the context of a product such as
     charge a license fee for that server.                           Episode Engine, which an experienced compressionist
         The master allocates work throughout the encoding           will set up and most users will interface with via watch
     cluster, and if any encoding station fails, the master          folders, the problems aren’t nearly as significant.
     reallocates all work to the remaining servers. However, if         The other noteworthy usability issue is the bifurcation
     the master fails, you’re out of luck. With the High             of server-related controls between the administrative
     Availability option, you get an extra master on a cluster, so   application and the controls in the preference panel.
     if the first master fails, the second takes over.               Maybe it’s a Mac/Windows thing, but if you have an
To top