Is 2 or 3 pass in CCE really necessary?

[Commander XJL]

Is it really necessary to use 2 or 3 pass in CCE? In TMPG when I use 2 pass I don't see a difference than when I use 1 pass. So is 2 or 3 pass needed?

[mikeathome]

Originally posted by Commander XJL
Is it really necessary to use 2 or 3 pass in CCE? In TMPG when I use 2 pass I don't see a difference than when I use 1 pass. So is 2 or 3 pass needed?

Hi,

we had have exactly the same discussion already. In general, NO, it isen't if your movie is a 1:2.35 anamorph or even letterbox and length ~100mins and below. In this case you'll fit a 1passVBR Q:30 (Q:40) re-encode on a DVD-R which gives very good quality hardly to top with a multipass.

This is my personal opinion based on ~100 backups.

mike

[Mazze]

But how about those projects where you squeze 2 movies on one disc. At bitrate as low as 1700kbit/s are there any vissible diffrences between a 1,2 or 3-pass encoding. Havn't had the time to try it out myself as this surely is a time consuming process, thought that you seems to have tried it already and might have the answer?

[easy2Bcheesy]

Not sure about fitting two movies on a DVD-R! But generally, in my experience with MPEG2 encoding and CCE, getting the Q level right is far more important to the naked eye than the amount of passes you do.

[jdobbs]

I hate to disagree, but... In virtually all situations multiple pass encoding will create a superior picture at a given bitrate. Decreasing the Q increases the bitrate, so yes it provides a better picture than if you used more quantization. But when you trying to get the very best picture out of the same space -- multiple passes will always give you the best picture possible.

Single pass variable bitrate encoding spreads the available bitrate across a very narrow window. Multiple pass allows you to take the total number of bits available for the entire encode and allocate it exactly as needed across the entire movie. You could have an entire high-demand section that is encoded at 8000Mbs (rare) while another large section of blank screen uses almost zero.

Now with that said... when you have way more bandwidth than you need (like a short movie in a large DVD-R) -- a very low Q factor could result in a picture that is almost indistiquishable from 3 passes, not because its as good a method, but because there was just so much damn space that you can't tell the difference. BUT... if you didn't have that luxury, you could get an equally good picture in a smaller space using multiple pass!

I personally like the Robshot multiple pass method. It gives a picture in two passes that is hard to beat.

[Commander XJL]

were is this robshot method at, I want to try it

[jdobbs]

Well I just noticed that www.robshot.com (where it used to be) is no longer availabe... but good old Doom9 has a copy in his guides called "Getting the most out of CCE"