Transcoding 101 Roundtable, MIX 2011 and NAB 2011

A bit of shameless promotion:

On Thursday, March 24th, I’ll be participating in a roundtable web event titled “Transcoding Methodology 101” hosted by Streaming Media. Joining me will be Kevin Louden (Telestream), Charlie Good (Wowza) and Jon Robbins (Rhozet). The live event is scheduled to start at 11 am Pacific Time.

You can register for the web event here:


Microsoft’s annual web developer/designer conference MIX is taking place the same week as the annual NAB show this year – also in Las Vegas. MIX 2011 is taking place April 12-14 at the Mandalay Bay hotel, while NAB 2011 is happening April 11-14 at the Las Vegas Convention Center. Microsoft’s media platform will be represented at both events, and I will be present at both too.

Here is the current list of MIX sessions:

I will be presenting a session titled “Introducing Microsoft Media Platform” in which I will give an overview of Microsoft’s media technologies and frameworks and recent releases. Other noteworthy media related sessions will be:

“Introducing MMP Content Manager” by Steven R. Woodward (Microsoft)
“MMP Video Editor” by Jason Suess (Microsoft)
“MMP Player Framework: Past, Present, Future” by Tim Greenfield (Vertigo Software)
“Introducing MMP Audience Insight” by Eric Schmidt (Microsoft)
“Behind the Scenes of Channel 9 Live at MIX” by Nic Fillingham (Microsoft)
“5 Things You Need To Know To Start Using <video> and <audio> Today” by Nigel Parker (Microsoft)

It’s not too late to register for MIX 2011! Visit and register today!

About Alex Zambelli

Alex is a Senior Product Manager at Hulu in Seattle, WA. Prior to his current job he was a Product Manager at iStreamPlanet (Turner) and Technical Evangelist for Microsoft Media Platform at Microsoft Corporation. He specializes in video streaming, adaptive HTTP streaming, video compression, and video processing best practices.
This entry was posted in Expression Encoder, Internet Information Services, Silverlight, Smooth Streaming and tagged , , , , . Bookmark the permalink.

2 Responses to Transcoding 101 Roundtable, MIX 2011 and NAB 2011

  1. David Ronca says:

    On the “Transcoding Methodology 101? you mentioned that 2-second GOP size was found to be optimal for adaptive streaming. The Expression Encoder has an option for variable GOP sizes aligned across multiple bit rates. Have you found that 2-second is better than aligned variable bitrate?

    • Hi David,
      Expression Encoder (and any other encoding products based on the Smooth Streaming Encoder SDK) does offer the option of using adaptive GOP in 2-pass VBR mode, even for Smooth Streaming output. This is possible because that particular SDK is multi-bitrate aware so it knows to align GOPs across multiple bitstreams. Speaking strictly from a quality perspective, using GOPs longer than 2 seconds and starting GOPs on scene changes is likely to produce better quality video. However, from a client heuristics perspective it does make network performance less predictable because the client can no longer accurately estimate the upcoming number or frequency of HTTP requests, or the average size of the upcoming chunks. It’s not an impossible problem to work around, but it does make the heuristics more complex and therefore less reliable. For that reason, as well as because of the fact that not all encoders support aligned variable GOPs, we tend to err on the side of simplicity.

      In my opinion not enough testing has be done across the industry to actually figure out which GOP/fragment legnth works best on what networks and in what conditions. Does one GOP length work better than another one on WiFi vs LAN, DSL vs cable, 3G vs 4G, Internet vs intranet, etc? I wish I knew the answer myself.