Twopass-2nd Encoding. Higher Quality & Smallest Size - "hoping knowing the way"

[MercuryZ]

Hi,

i use Twopass-1st for file size 491 MB [D-CX 1280x720 WMV9 24fps HDTV ] = AVI .... the result was 812 MB!

see my way below:

Main Settings

Profile @ Level: Advanced Simple @ L5
Quantization Type: H.263

Max consecutive BVOPs: 2
Quantizer ratio: 1.50
Quantizer offset: 1.00

removing the Tick from Packed bitstream

______________________________________________

1nd Pass --> more

remove the tick from Discard first pass


Tick Full quality first pass
___________________________________________________

Zone Option

Start frame #:

Weight:
Quantizer:

which of tose should i tick it?

BVOP sensitivity:

_______________________________________________________

Quality preset: (User defined)

more

1-Motion

Motion search precision: 6 - Ultra High
VHQ mode: 4 - Wide Search

Tick Use chroma motion

frame drop ratio:
Maximum I-frame interval:

2-Quantization

Quantizer restrications: {2,31,2,31,31}

Tick Trellis quantization

_________________________________________________

That's was my way to encoding with [ Twopass-1st ]

i hope i can found the best way for encoding using XviD .. resulting high quality and small size. i use Twopass-2nd, i think its the better way. So, i want some one help me
with Twopass-2nd Options.

Video's Propoties:

[ 491 MB .. AVI .. 120fps ]

as i know .. the Target Size = File Size * 1024 .. it will be

491 * 1024 = 502784 kb

thats all i know

i hope someone will help my here

Mercury

:D

[Blue_MiSfit]

Welcome to Doom9!

Ok I'm a little confused. So you have a WMV9 source at 1280x720, and you want to re-encode it to XviD? That's fine.

It sounds like you are doing a full quality first pass - and not doing the second pass. A full quality first pass is generally not necessary, and there's no point in doing a first pass if you don't also do a second pass :D. Also what's this about 120fps?? Your source is 24, why change it??

Reset everything to defaults, and setup a batch job doing the first pass followed by the second pass (where you specify the target file size or bitrate that you want to hit). Make sure all settings are consistent in both passes. Leave everything at defaults, and try that.

Once we've confirmed that you can actually hit a target filesize then you can turn on advanced features like multiple b-frames, higher VHQ, qpel, AQ, CQMs, etc...

Also, how are you encoding? Through VirtualDub? xvid_encraw in MeGUI? How are you feeding the WMV into Xvid? AviSynth's DirectShowSource?

More info! :D

~MiSfit

[MercuryZ]

thanks dude.

i use VirtualDuMod.

i'll try the defaults setting bys changing only [ Target Filesize ]

and i'll tell u the result.

( did u have a good or perfect way to encode ? and i want the file small as we can and high )

Mercury

:D

[Blue_MiSfit]

Well, it's hard to suggest settings when I haven't seen the source :)

I've found that some of the common advanced settings usually provide nice improvements.

I almost always use Quarter-Pixel, Adaptive Quantization, at least VHQ1, sometimes 4, and usually some sort of custom quantization matrix. But you'd best get really comfortable with XviD and comparing different clips before you start messing with CQMs :)

You never know how small you can get the video while maintaining high quality. The term 'high quality' is very subjective. What looks like crap to me might look great to you... and vice versa. Sometimes you can fit a whole 2.5 hour movie in ~700MB, sometimes it takes me ~2200MB (1/2 of a DVD+R).

Honestly though, most of the 'big' changes come into play with AviSynth filters. They can really work magic.

Let us know how it goes, and don't give up! :D

~MiSfit

[badshah]

Honestly though, most of the 'big' changes come into play with AviSynth filters. They can really work magic.


he is using Vdub and so do I.... do u mean to say AviSynth filters can do magic and Vdub filters can't ?

[Brother John]

Generally speaking any filtering can work magic, but magic both ways. That's why filtering is an even more complicated subject than encoder setup.

As for VDub vs. AviSynth filtering: Filters tend to be developed for AviSynth rather than VDub and AviSynth has that scripting language that in connection with some filter packages (who might think of masktools :)) can work additional magic. So AVS filtering is a lot more flexible than VDub's. Quality depends on the actual filter. I suppose generally AviSynth has the higher quality filters.