Xvid encoding: Single Pass or 2 Pass?

[Saito_Hajime]

I'm currenly converting a rmvb file to avi using virtualdub through avisynth. Now comes the compression and encoding of the video. My question is.......which encoding is better when it comes to keeping the quality of the video and at the same time, having a reasonable file size?

[celtic_druid]

Well with those constraints; 2 pass. 1 pass fixed quant is fine for quality, you just have no real control over filesize.

[Saito_Hajime]

Well with those constraints; 2 pass. 1 pass fixed quant is fine for quality, you just have no real control over filesize.

Uhmm.....there is 2 pass (first pass) and 2 pass (second pass), which one should I choose? About constraints, does that mean single pass doesn't need any changes to it's settings and still keep it's quality?

[celtic_druid]

Both. That is the point of 2 passes. It requires 2 passes.

[jagabo]

If you don't need a file of a specific size then use single pass target quantizer mode. It's twice as fast as 2-pass VBR, and you don't need to figure out what bitrate will give you the quality you want (the appropriate bitrate will vary depending on the frame rate, frame size, and nature of the video). Pick the quality you want and the encoder will use whatever bitrate is required to get that quality for each frame, not a bit more, not a bit less.

With a target quantizer of 2 the output will be nearly identical to the input, even if you look at enlarged still frames. With large frame sizes the bitrate might exceed what some set-top Divx/DVD players can handle.

At 3 you will see a little macrblocking if you look at enlarged still frames. You won't notice them at normal playback speed.

At 4 or more you will get more macroblocking and it may become visible at normal playback speeds.

As celtic_druid pointed out, with target quantizer mode you don't know exactly how big the final file will be. On the other hand, with CBR or 2-pass VBR you don't know what the quality will be. In the end, if you convert a video using target quantizer, then 2-pass VBR set to the same file size, the two will look nearly identical.

If your intention is to fit a full movie on one CD it's easier to use 2-pass VBR. You calculate what bitrate is required to fit the movie into 700 MB and encode in two passes.

If you're putting lots of 22 minute TV shows on DVDs then encode in target quantizer mode at the quality level you want. Then put as many files as will fit on each DVD.

[Saito_Hajime]

I see. Thanks for the help guys . I guess I'll just go for single pass since it's easier, faster and keeps the same quality of the original video. The rmvb file is a 1 hour tv show and the size is 385MB, so I guess it won't be that big if I just convert it to avi using a single pass.