My biggest hurdle is the same VOB->AVI->MP4 workflow.

I got the overall workflow down. I'm just trying to tighten everything up.

my only other tip is to use x264 and no cabac. There is a quality boost at performance hit for using cabac, but all my files that I have encoded with cabac off work perfect on dual core computers running vsl, no crashes whatsoever and the quality looks the same. 3 frames and your good to go.

and when I mean dual core computers, i mean any computer being mac or pc using ffdshow or quicktime.

I love Cabac, but that's pretty much from unc to mp4. I'm trying to get the best quality from interlaced vob to ivtc'd avi

nerds

+1. Haha

+1!

Glad to see so many care.
Nik, I know you know about this, you and swift were the ones who started in on the whole "pulldown removal"

I care but you guys just sound so nerdy with all the terminology...haha.

I am watching this thread intently for secret tips.

you guys get respect for improving what you do.
making the videos better for the rest of us is very appreciated.
the time and work you put into this is amazing.
your patience is far beyond what mine is.


nerds!!

you rock though, seriously.

It's not patience, I think it's persistence.

I'll keep doing it until I get it right, but then I quickly get bored and move on to the next thing.

automate DGIndex with a batch script or powershell
DGIndex -wait -argumentlist -i filename -o filename -fo 2 -om 2 -exit -hide

automate the creation of your avisynth scripts with a batch script or powershell
Create an avs template, put a place holder in for the filename. if you really want to get fancy you can parse the the DGIndex log to determine if the file is standard or widescreen.

if using MeGUI to encode, automate populating the queue with a batch script or powershell
This is tricky and requires a lot of steps. MeGUI stores queue info in several different files. Although it can be done it a batch script or powershell it was easiest for me to do in Automate (by Network Automation).

when going directly from DVD or VOB to MP4 your avisynth script should perform the following tasks (in this order)
DeInterlace *if needed
IVTC *if needed
crop *if desired or needed - usually to remove black space or artifacts from video edges
resize *resize to 640x480
crop *if the video is widescreen; (0,60,-0,-60) will result in a 640x360 resolution
ConvertToYV12 *this is the colourspace of the MPEG standard

when going from DVD or VOB to your NLE for editing and then to MP4 you need two avisynth scripts
1. Pre-NLE
DeInterlace *if needed
IVTC *if needed
crop *if desired or needed - usually to remove black space or artifacts from video edges
resize *resize to 640x480
crop *if the video is widescreen; (0,60,-0,-60) will result in a 640x360 resolution
ConvertToRGB *this is the colourspace of most NLE's (Premiere Pro, Final Cut Pro, Avid, Vegas, Final Cut Pro X er i mean iMovie Pro X, etc)
2. Post-NLE
ConvertToYV12 *this is the colourspace of the MPEG standard
***Post-NLE you could also export to uncompressed avi and encode to MP4 in your encoder of choice***

Colorspace Info
Digital Video can be stored in either RGB or YUV. RGB stores a color value for Red, Green and Blue per pixel which can be added together to create any color. YUV stores color in Luma (Brightness) aka Y and Chroma (color info) aka U/Cr and V/Cb.

YUV is sampled less per pixel than RGB because the accuracy of the chroma is less noticable. Because of this there are several subsampling variations. The most common two are YUY2 and YV12.

YUY2 samples the luma once every pixel but only samples the chroma every horizontal pair of pixels. YV12 samples the luma once every pixel but only samples the chroma every 2x2 pixel blocks.

MPEG1, 2 and 4 (x264, XviD, DivX, etc) - All use 4:2:0 YUV (YV12) color.
DVDs - Uses MPEG-2 with 4:2:0 YUV (YV12) color.
DV - Uses 4:1:1 YUV with the same number of chroma samples as MPEG-2 but in a different order.
MJPEG - Uses all kinds of YUV sampling, but 4:2:0 YUV (YV12) is the most common.

mofucking Nyce with the info.

So can't I go from vob to YV12 into sony vegas.
I did this on the weekend and vegas will accept yv12 or RGB

Revealing all teh s3cr3t5, bastard... ;)

thanks for not Nik

So this is where the l33t Video editors hang out?

I'm watching lots and understanding little.

n3rd5!!

haha

;)

Is it possible on a mac or not yet?

nope.
sorenson does it, but it's very hit or miss. I don't trust it.

all I want is to be able to have those 23.976 frames be perfect frames, no interlacing, artifacts etc.

Culprit can you explain this: I would think most new computers would handle Cabac as easy as Cavlc and you get a higher quality output.

Thanks

+1

I'm glad you guys are sharing the knowledge

Yeah I'd like to know this too.
I started iwth Cavlc at baseline, then moved to Cabac at high and now I'm at Cabac Main.

And not saying cabac at high, main or baseline. That's for the main concept stuff.

Niro, newer computers yes, but I was specifically pointing out Dual Core Computers, or older pc's & macs in compatibility and stability. If you encode a dvd -> with cabac on most pc's will stutter or crash, with it off I get flawless performance.

Newer computers now and days you get a quality boost with it on. All personal preference, but I tried many encodes and did not notice MUCH of a difference.. but there still is a difference

damnnittt.. i meant to say most dual core pcs.. but I am specifically pointing that out. Newer computers (especially with direct x 11 capable) you should not have an issue.

what about ivtc'ing stuff Culprit?

The ivtc'ing is done before the editing/encoding so that really is not an issue you know.

NO I know that, but this thread is about advanced IVTC'ing
Got any tips for decimating, de-interlacing, decombing, colorspace etc?

Nah whats been posted above seems to be the standard workflow.

there has to be different ways, some people might use tdeint or tfm, decimate vs tdecimate etc.

I've used Episode Encoder (Mac) for IVTC, and it usually seems to work quite well. Assuming the material can actually be IVTC'ed at all, of course (there's a lot of older stuff that cannot).

At scene changes you'll sometimes get some interlaced frames showing up (stuff that was shot on film, but edited on video). A bit of manual editing takes care of that.

Debuging AVISynth Scripts
Firstly, use AVSP. It's like Visual Studio for avs files. It has intellisense for built-in functions and user-defined functions.
avisynth.org

To add your your custom user-defined functions (avsi): Edit->AviSynth Function->Show Function Definition or Ctrl+D. Close the edit function info window. Click 'Import from Files', browse for AVSI and check all the functions you want to import. Now your custom functions are included in the AVSP intellisense. *Make sure 'Include user functions in autocompletion lists' is checked.

Secondly, throw a SubTitle line in your script to display the value of a variable if you are getting errors.

Also if you would like to see the video information, use the Info() function.

Decomb
Alot of external functions for this process. I use the Decomb package which is a combination of Telecide, Field Deinterlace, Decimate and IsCombed. For DVD to MP4 the only function needed is Decimate, but for other material Telecide and Field Deinterlace can be used when appropriate.

Deinterlacing
The built-in function, BOB, is ok but it uses Bicubic Resizing. The most popular external functions are TDeint and YADIF. I always use TDeint.

When rendering to mp4
STOP USING BASELINE for the AVC Profile
Use Main,
Use MAIN,

Did I mention use main,
Also use CABAC, computers can handle this now.

AVC = Main
Cabac
Ref frames = 3

Next person I see using baseline on their edits is getting put on blast.
Thanks

Format settings, CABAC : No

Here's my sample settings, they been pretty solid on every machine (pc and mac) that i've put to the test, but i don't do much of any edits anymore

General
Complete name : E:\Crates\1990\Urban\E40 ft Nate Dogg - Nah Nah (Intro Dirty) - Widescreen [LALOUD].mp4
Format : MPEG-4
Format profile : Base Media
Codec ID : isom
File size : 84.4 MiB
Duration : 3mn 32s
Overall bit rate : 3 325 Kbps
Movie name : Nah Nah (Intro Dirty)
Album : Loyalty and Betrayal
Album/Performer : E40 ft Nate Dogg
Part/Position : 0
Part/Total : 0
Performer : E40 ft Nate Dogg
Composer : DJ Culprit
Genre : Urban
Recorded date : 1999
Encoded date : UTC 2011-03-31 08:29:44
Tagged date : UTC 2011-03-31 08:29:44
Writing application : Culprit Encode Script 20110223
BPM : 100
Copyright : Jive Records
Cover : Yes
Comment : 4A
initialkey : 4A
catg : 90's Urban

Video
ID : 1
Format : AVC
Format/Info : Advanced Video Codec
Format profile : Main@L3.0
Format settings, CABAC : No
Format settings, ReFrames : 3 frames
Format settings, GOP : M=2, N=21
Codec ID : avc1
Codec ID/Info : Advanced Video Coding
Duration : 3mn 32s
Bit rate mode : Variable
Bit rate : 3 000 Kbps
Maximum bit rate : 5 512 Kbps
Width : 640 pixels
Height : 360 pixels
Display aspect ratio : 16:9
Frame rate mode : Constant
Frame rate : 23.976 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Bits/(Pixel*Frame) : 0.543
Stream size : 76.2 MiB (90%)
Title : Nah Nah (Intro Dirty)
Writing library : x264 core 114 r1913 5fd3dce
Encoding settings : cabac=0 / ref=3 / deblock=1:0:0 / analyse=0x1:0x111 / me=umh / subme=7 / psy=1 / psy_rd=1.00:0.00 / mixed_ref=1 / me_range=16 / chroma_me=1 / trellis=1 / 8x8dct=0 / cqm=0 / deadzone=21,11 / fast_pskip=1 / chroma_qp_offset=-2 / threads=6 / sliced_threads=0 / nr=0 / decimate=1 / interlaced=0 / constrained_intra=0 / bframes=3 / b_pyramid=0 / b_adapt=1 / b_bias=0 / direct=1 / weightb=1 / open_gop=0 / weightp=2 / keyint=24 / keyint_min=2 / scenecut=40 / intra_refresh=0 / rc_lookahead=24 / rc=2pass / mbtree=1 / bitrate=3000 / ratetol=1.0 / qcomp=0.60 / qpmin=0 / qpmax=69 / qpstep=4 / cplxblur=20.0 / qblur=0.5 / ip_ratio=1.40 / aq=1:1.00
Language : English
Encoded date : UTC 2011-03-31 08:29:44
Tagged date : UTC 2011-03-31 08:29:45

Audio
ID : 2
Format : AAC
Format/Info : Advanced Audio Codec
Format profile : LC
Codec ID : 40
Duration : 3mn 32s
Bit rate mode : Variable
Bit rate : 319 Kbps
Maximum bit rate : 343 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 44.1 KHz
Compression mode : Lossy
Stream size : 8.10 MiB (10%)
Title : Nah Nah (Intro Dirty)
Language : English
Encoded date : UTC 2011-03-31 08:29:45
Tagged date : UTC 2011-03-31 08:29:45

Sorry Puma, I knows no betta.

At culprit. I remember you saying no cabac a year or two ago, but why not now.

I've seen most people use it and most pools.

Cabac makes a difference when it comes to CPU utilization...

Are Puma and Culprit related? I'm just sayin!!

lmao no.

lmao i dont get it dub!

Yes CABAC causes less stress on the CPU, and makes all my videos for the PC work pretty flawless (even for older dual core pc's which is a plus), even though I have my mbp. I stick with it because I really was upset when I got my alienware computer so long ago and had to learn the hard way.

*NO CABAC" i mean

fuck man, just repeating myself a zillion times in the same damn thread.. ignore plz and happy friday

Years later, I still really can't find a way to ivtc on a mac that works as good as the OG PC way.

They claim the pull down in after effects is made for it, but it gets mixed reviews

I stick with the pc method, it's solid and easier anyways imho

Yeah, I just haven't had a PC in 5 years.
Using this old bohemith to move files to it, ivtc, then move a uncompressed file back
(Without TB or usb3)

1st world problems

bootcamp son!

I IVTC with my gig laptop using bootcamp, it's a 2011 and it works great for IVTC and editing in Vegas.
Would be faster in bootcamp, but I don't feel like booting to the windows side and I can easily edit audio in Ableton for use in Vegas.