Find millimeter on Facebook

Related Articles

 

Distribute Expertise: Encoder Shootout

Dec 1, 2008 12:00 PM, By Jan Ozer

An on-the-job comparison of sub-$1,000 streaming-media encoders for the Apple Mac.


      Subscribe in NewsGator Online   Subscribe in Bloglines  

Adobe Media Encoder CS4 features batch, watch folder, and standalone operation.

Adobe Media Encoder CS4 features batch, watch folder, and standalone operation.

If you produce streaming-media files on the Apple Mac platform, you have a number of sub-$1,000 encoding options. These include the encoding tools bundled with your editing suite (whether it's Adobe Media Encoder or Apple Compressor) as well as several third-party tools, such as On2 Technologies Flix Pro (Flash only), Sorenson Media Squeeze, and Telestream Episode Pro. In this overview, I'll compare each product's codec-specific features, encoding speed, and output quality for each of the major three formats: H.264, VP6, and Windows Media Video (WMV) 9/VC-1.

If you produce streaming video files at a high enough data rate, all encoding tools look great. If you're a high-volume producer, however, bandwidth is money, and you need to know which tool retains the best quality at the lowest possible data rate. For this reason, I tested using very aggressive data rates. Specifically, I produced standard-definition tests at 640×480×30fps at a video data rate of 468kbps and an audio data rate of 32kbps. I encoded using two-pass CBR for Windows Media and two-pass VBR for H.264 and VP6. As with all tests, I checked to ensure that total file size was within 5 percent of the target, which often required multiple re-encodes.

In high-definition comparisons, I used a 720p file (1280×720×29.97fps) encoded at 800kbps in two-pass VBR, with 128kbps audio. As with the SD test file, I encoded using a prescaled and pre-de-interlaced file rather than submitting the original HDV video file (or DV for the SD trials) and allowing the encoding tools to scale and de-interlace while encoding. That way, I could isolate pure encoding quality, rather than examine an amalgam of the tools' scaling, de-interlacing, and encoding skills.

The final test involved a 1024×768×15fps screencam capture encoded to 200kbps video and 32kbps audio. Such screencams are a great training option for many software vendors and IT support folks, and it turns out that there is a huge variation in encoding quality among the various encoding tools.

With each set of tests — SD, HD, and tutorial — I grabbed and compared frames produced by each encoder. I also played each video in realtime to evaluate and compare motion quality — in particular, looking for banding and color-shifting in video backgrounds and artifacts such as mosquitoes, or faint lines hovering over sharp edges. I rated each encoder on both still and motion quality.

Since many producers still shoot in interlaced formats or need to encode older, interlaced source material, the ability to de-interlace the two fields into a sharp, cohesive frame is critical to overall encoding quality. To test each encoder's de-interlacing skills, I used a DV file containing a minefield of sharp diagonal edges, slanted guitar strings, spaghetti-thin Venetian blinds, and other de-interlacing challenges. In each encoder, I input the file; output a scaled, de-interlaced 640×480 file; and compared results.

Performance trials

To test rendering performance, I encoded a 1-minute SD file to my usual SD encoding parameters using each supported codec, and then encoded five 1-minute files to the same specifications. With each tool, I used all available techniques to boost encoding speed. For example, with Flix Pro and Squeeze, I copied and pasted the application to a new name until I had five running applications, and then I used each instance to encode a separate file. A touch of an administrative hassle, but the performance boost was more than worth it.

I used Apple's Qmaster to enable Compressor to utilize my eight-core Mac Pro loaner as a cluster of eight computers and efficiently distribute various encoding chores to all eight cores. Alas, with the Adobe encoder and Episode Pro, none of these options were available, which has the most impact on VP6 encoding times.


Continue the discussion on “Crosstalk” the Millimeter Forum.
© 2010 Penton Media, Inc.

Browse Back Issues
BROWSE ISSUES
   
Millimeter
September 2009
Millimeter
August 2009
Millimeter
July 2009
Millimeter
June 2009
Millimeter
May 2009
Millimeter
April 2009
Back to Top