OPV Vs 2-pass

[techreactor]

What would yield better Quality encode, OPV OR 2-Pass encoding using HCE ???

This is assuming other parameters are the same and time is not a constraint.

[nwg]

I got undersized about 500MB doing OPV. I always do 2 pass now.

I would only do OPV (if it worked ok) when doing less reduction as possible. If reduction was 40-50% then I would do 2 pass eveytime.

[UncasMS]

hc's opv mode produced normal file sizes for me:

http://forum.digital-digest.com/showthread.php?t=55813

cce 2.5 + 2.7 werent any different in my test

and the quality of tzhe opv mode was very good as well

c.f. this: http://forum.digital-digest.com/showthread.php?t=57569

[blutach]

If time is not a constraint, then why not 2 pass (Best mode)? With OPV, you need to figure out your CQ index or something, don't you? There's a little app for doing this I saw.

Regards

[techreactor]

If time is not a constraint, then why not 2 pass (Best mode)?

Hi blu,

That was a test condition so as not to get ppl into unnecessary discussion over time and deviate from the goal, i.e. Good Quality. Although most of us know that OPV would be faster.

I saw a comment from Kassandro(i think it was his forum) wherein he mentioned that use of 2-Pass mode in any encoder, negates the quality gain/enhancements done by the filters on the video. So I even if I get "near" quality levels, I would prefer OPV.

With OPV, you need to figure out your CQ index or something, don't you? There's a little app for doing this I saw.

Yes, its called GetCQ, I think I posted it in one of my posts recently along with the steps.

[jdobbs]

I saw a comment from Kassandro(i think it was his forum) wherein he mentioned that use of 2-Pass mode in any encoder, negates the quality gain/enhancements done by the filters on the video.

I'm not sure what logic that could possibly be based upon? The two encode pretty much the same way. The difference is that in two-pass encoding the necessary quantization is applied in order to reach a consistent size/ average bitrate. In one-pass you choose the quality level (generally quantization in CCE, fully quantization in most others) and the necessary bitrate is applied to reach that level.

Generally speaking you will get better quality from two pass because it better utilizes the existing space -- but sometimes the difference is so small that it's undetectable.