DVD2SVCD: DVD format squashed vertically?
 

question:

I encoded a DVD ("Changing Lanes") with "encode as 16:9 (borders added)" -> the result is "squashed" vertically (even in 4:3 mode on the TV). I used the adaptive script and entered source anomorphic is true and des = false on the frameserve tab of DVD2SVCD.

Does this mean that the source in fact was not anamorphic (16:9) but 4:3 ??
Amazon lists the following for "changing lanes" but this is the US release. Maybe in europe the format is different ??


Format: Color, Closed-captioned, Widescreen

Hi Paulus,

It's possible that your dvd is actually a 4:3 "Letterboxed" widescreen version. If your dvd does not say "Anamorphic" on it anywhere, then that is probably what happened in your case, and then you should have used source_anamorphic=false in GripCrop. Don't you just hate it when they use letterboxing for some widescreen movies on dvds! :x What a waste of bitrate! :evil:

You can always take note of the aspect ratio of your movies in dvd-decrypter or dvd2avi before you proceed with encoding it, just to be safe. It sure stinks waiting several hours for an encode to finish and then finding out that you've got to do it all over again!...I think we've all done that at one time or another...I feel your pain! :lol: :wink:

-d&c

Hello Dazed & Confused,

I think you're right ! I'm running another encode as we speek with source ana=false.

Good suggestion about checking first in DVD2AVI. Stuppid that I didn't think of that. :oops:

This is actually my 4th encode already for this movie (first three to get the best fixed CQ value which I'm always struggling with). Now I've to start again with finding the good CQ value with "source ana=false" setting because it's way different :x

Do you have a trick for that as well :D ? I have tried TOK but for some reason I wasn't able to use is very well for file size prediction.

Paulus, the prediction is the big problem of MA script. But give a look to "CQMatic", the new tool kwag is developping to fix this point.

Hi again Paulus,

If you're not having good luck with Tok's prediction, then do what Dialhot suggested and try out CQMatic. I'm sure that we'll all be using CQMatic soon with the way Kwag's been working on it so hard! :)

When all else fails, you can always use good old Manual Filesize Prediction as described in this thread: http://www.kvcd.net/forum/viewtopic.php?t=2102 . This is the method that I still use for now because I don't mind the extra work it takes doing it this way since it usually gives me very accurate results. Once you get enough encodes under your belt, you sorta start to get a feel for estimating the right CQ levels for different movie types/lengths. So it doesn't usually take me many sampler encodes to narrow in on the proper CQ with the manual method anymore.

But hopefully Kwag's CQMatic will soon make manual-prediction a thing of the past! :wink:

-d&c