iPod touch Video encoding 500kbps vs. 1000kbps H.264.

[jmpage2]

I've been doing some more video encoding tests recently, using H.264 in Handbrake on my PC.

Just tested the most demanding video I've encoded yet. LOTR Return of the King Extended Edition, which is split across 2 DVDs with no special features on those discs, giving this title a very high video bitrate which makes it a good test.

Compared chapter 18 which is a great test as there's lots of smoke and other things which will push an encoder to the limit.

Definitely in this particular example there were notable differences between 500kbps single pass H.264 and 1000kbps two pass H.264.

In the lower bitrate example there was some obvious artifacting in several areas of this chapter, even around static things like actor's faces. This was probably caused by the presence of smoke, etc, in the scene.

The 500kbps was still watchable but the 1000kbps example looked far more accepteable for more discerning viewers.

 

[Multimedia]

I Agree With Your Assessment

I always use 1Kbps for HD content and 800kbps for SD. In the case of a movie I would always use 1Kbps unless it's SD when I would simply target 700MB so it'll fit on a CD. 20 minute cartoons like Simpsons and South Park I target for 100MB which is a little more than 500kbps.

I always use 2 pass. And I maximize the dimensions to just under iPod limits so they will play well on big screens too.

544x400 SD 4x3
624x352 HD 16x9

 

[BigCanoe]

Do you choose one of the ipod's screen dimensions and encode to that? Or do you make it larger? Do you always make videos a certain width and height to play on ipod only?

 

[aidanpendragon]

I've been doing some more video encoding tests recently, using H.264 in Handbrake on my PC.

Just tested the most demanding video I've encoded yet. LOTR Return of the King Extended Edition, which is split across 2 DVDs with no special features on those discs, giving this title a very high video bitrate which makes it a good test.

Compared chapter 18 which is a great test as there's lots of smoke and other things which will push an encoder to the limit.

Definitely in this particular example there were notable differences between 500kbps single pass H.264 and 1000kbps two pass H.264.

In the lower bitrate example there was some obvious artifacting in several areas of this chapter, even around static things like actor's faces. This was probably caused by the presence of smoke, etc, in the scene.

The 500kbps was still watchable but the 1000kbps example looked far more accepteable for more discerning viewers.

Unfortunately, this is an apple-and-oranges comparison: two-pass should always look (a little to a lot) better than one-pass. A better test is to run 500 and 1000kbps each as one- and two-pass and compare all 4. Maybe 500kbps @ two-pass itself looks "far more accepteable for more discerning viewers"?

 

[clevin]

discerning viewesrs shouldn't be watching a movie on a 3.5" low quality screen anyway. I think the concern is misplaced, unless you plan to watch same video file on computer or tv too.

 

[madmaxmedia]

Or do a 2-pass 500 and a 1-pass 1000, to see if the extra pass makes up much of the difference in bitrate.

Thanks for the info though.