AVCHD Footage to SD (AVI for Premiere)

[poisondeathray]

You can do it all in Premiere too, that's definitely an easier option and workflow. HDV is 1440x1080i 4:3 PAR, so mixing AVCHD 1920x1080i footage can cause some problems as well. But at least both are TFF, so you don't have the field reversal business.

The quality of the resize and deinterlace using Adobe (or vegas for that matter) isn't as good as using other higher quality methods. This has been proven time and time again here and on other forums.

99.9% of the time a DVD made from downscaled HD material will look better than one made shot from the same camera in SD mode.

We are just providing options; it's up to the end user to choose what sacrifices in terms of quality or workflow s/he is willing to make. Some users might not be able to "see" the differences but I assure you 100% , that there are clear differences.

[foochuck]

Where can I get the UT video codec you referenced below? That is neither lagarith or huffyuv, correct?

Now open that .avs file in virtualdub. Select video=>fast recompress, video=>compression (then select either lagarith, or huffyuv or ut video codec or some other lossless format, you may have to install them if you already haven't. I highly recommend UT video codec for lossless if you have a quadcore, its many times faster than lagarith and almost twice as fast as huffyuv-mt for decoding), then file=>save as avi. If you leave audio as direct stream copy, it will encode the audio as wav. If you wanted to disable this, just go audio=>no audio. Then you could import the native audio separately into premiere (you might demux it using tsmuxer for example)

[poisondeathray]

UT Video Codec
http://forum.doom9.org/showthread.php?t=143624

One "bug" is that it has to be set up in the configuration each time you use it to encode (it doesn't remember or save the settings). If you have a quad, set it to 4. If you have a i7 with HT, set decode to 8 threads etc...

[foochuck]

After opening the script in Virtual Dub I received this error:

Avisynth open failure: Script error: there is no function named "JDL_ReverseFieldDominance"

Did I forget a plug-in?

[poisondeathray]

Yep, they are from stickboy's page. It's listed on a previous post

Get jdl-util.avsi and jdl-interlace.avsi from the url below, put in avisynth plugins directory (just like you did with nnedi2.dll, yadifmod.dll) , and use the JDL_ReverseFieldDominance() function. The 1st and last field will be screwed up unless you pass a bobbed frame JDL_ReverseFieldDominance(bobfirstlast=true)

http://avisynth.org/stickboy/

[foochuck]

Got it. I'm saving my first clip now trying out the lagarith codec.

One question about UT video codec - I'm using an intel core i7 - you said it would run faster on this computer - however my other computer isn't quite as fast - for editing purposes, would it be better to stick with lagarith or were you just referring to the decoding process?

[poisondeathray]

Got it. I'm saving my first clip now trying out the lagarith codec.

One question about UT video codec - I'm using an intel core i7 - you said it would run faster on this computer - however my other computer isn't quite as fast - for editing purposes, would it be better to stick with lagarith or were you just referring to the decoding process?

For editing, you need fast decode speed. Lagarith, although commonly used, is fricking slow. Edits aren't very snappy even with a fast PC. (although you might be ok since it's only SD)

Look at the decode times for UT vs . lagarith. It's 8x faster on a quad, but not as good compression. If you want best compression for YV12 formats, FFV1 (using ffdshow) is slightly better than lagarith

[foochuck]

Wow this first clip looks like it's going to be over 2GB - the original m2ts file was 320MB!

[poisondeathray]

Yes, that's typical for lossless compression. Now think of how big an HD version would be...

If you wanted, you could encode to DV as an intermediate, but suffer some generational quality loss.

[foochuck]

I'm going to try out a DV Codec as well, I might be able to live with that.

[foochuck]

If I go with a DV codec, everything would be the same, I will just need to choose my DV codec when I decode the footage, right?

[poisondeathray]

If I go with a DV codec, everything would be the same, I will just need to choose my DV codec when I decode the footage, right?

You mean ENcode right?

So instead of lagarith, you would use your DV encoder (e.g. cedocida)

[foochuck]

lol...Encode...right.

[foochuck]

I just finished encoding this test clip. After playing the AVI file it looks like there is some artifacting going on - as if the footage is still interlaced. It didn't look like progressive footage (even after interpreting it correctly in Premiere). Is that just a loss in quality?

[poisondeathray]

I just finished encoding this test clip. After playing the AVI file it looks like there is some artifacting going on - as if the footage is still interlaced. It didn't look like progressive footage (even after interpreting it correctly in Premiere). Is that just a loss in quality?

You're testing this in premiere?

What were your sequence settings and conform settings?

Can you post a screenshot of what you mean? ( I tested this same procedure yesterday)

[foochuck]

Also, I left the audio on Direct Stream Copy, but no wav file.

[poisondeathray]

what model is your camcorder? can you use mediainfo on one of your avchd clips, and post the text information here?

and what mode did you shoot in? I assumed it was 1080i30

[foochuck]

General
ID : 0
Complete name : E:\video projects\footage camera 2\20090912134401.m2ts
Format : BDAV
Format/Info : Blu-ray Video
File size : 320 MiB
Duration : 5mn 16s
Overall bit rate : 8 499 Kbps
Maximum Overall bit rate : 18.0 Mbps

Video
ID : 4113 (0x1011)
Menu ID : 1 (0x1)
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4.0
Format settings, CABAC : Yes
Format settings, ReFrames : 2 frames
Duration : 5mn 16s
Bit rate : 7 895 Kbps
Width : 1 440 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate : 29.970 fps
Resolution : 24 bits
Colorimetry : 4:2:0
Scan type : Interlaced
Scan order : Top Field First
Bits/(Pixel*Frame) : 0.169
Stream size : 298 MiB (93%)

Audio
ID : 4352 (0x1100)
Menu ID : 1 (0x1)
Format : AC-3
Format/Info : Audio Coding 3
Duration : 5mn 16s
Bit rate mode : Constant
Bit rate : 256 Kbps
Channel(s) : 2 channels
Channel positions : L R
Sampling rate : 48.0 KHz
Video delay : -67ms
Stream size : 9.65 MiB (3%)

Text
ID : 4608 (0x1200)
Menu ID : 1 (0x1)
Format : PGS
Duration : 5mn 15s
Video delay : -67ms

[poisondeathray]

For the audio, do you have an AC3 decoder installed? e.g. AC3Filter ?

Otherwise you can use the demux (using tsmuxer) and reimport the native AC3 audio into Premiere

The shooting mode is important. If you don't know if it was 1080i30 (i.e. 1080i60) , vs some other mode, it could explain what you are seeing. If you are unsure of what you have, cut a sample using tsmuxer and post to a free hosting site eg.. mediafire.com

[foochuck]

The footage is: 1440 x 1080 (1.33) - Upper Field First - 29.97 fps

[poisondeathray]

Where are you getting that info from? I know what it says, but often you have different content than what the container signals. e.g. 24p in 30i container. It will read the same specs, but the actual content is different. Because whatever program you are using is reading only the header information

[foochuck]

NeoScene gives me the ability to convert my footage to AVI and to a progressive format. If I convert the footage to progressive, would it be easier to resize my footage?

[poisondeathray]

NeoScene gives me the ability to convert my footage to AVI and to a progressive format. If I convert the footage to progressive, would it be easier to resize my footage?

That would be deinterlacing it. And yes, there are no "special" issues when resizing progressive footage

But you would be losing 1/2 the fields. That's 1/2 the temporal information.

But I thought you wanted a SD interlaced AVI, comparable to your DV-AVI footage so you could use it together?

[foochuck]

You are correct, I do want SD interlaced AVI files so that I can edit them in the same Premiere Pro (CS3) project to go with my existing DV-AVI footage.

[poisondeathray]

You are correct, I do want SD interlaced AVI files so that I can edit them in the same Premiere Pro (CS3) project to go with my existing DV-AVI footage.

Then you don't want to single rate deinterlace it , which is what cineform is doing by converting to progressive. It's converting 60 fields per second (assuming that's what you really have, not 24p in 60i) to 30 frames per second progressive. And that won't match the DV footage (which is 60 fields per second, BFF).

You want to bob it up to 60p then resize it, then reconstruct the interlacing with lower field first (just like the DV footage) - which is what the avs script is essentially doing

You have to provide the information requested if you want more specific help...

[foochuck]

I'm gonna double check the camera model / shooting mode...

[poisondeathray]

It would be better to post a short sample. That way we can be sure exactly what you have, and what you need to do.

[foochuck]

I have a stupid question. Which of the following is a standard DV-AVI codec (from a virtual dub screen cap of video compression)? I'm guessing just Microsoft Video 1. I'm going to convert a short clip to that format.

[poisondeathray]

None of them.

I would install cedocida if you want to use DV-AVI

(But personally I would use lossless compression, especially if you are going though all the trouble of doing this properly through avisynth...)

[foochuck]

Here is a very short sample clip that was encoded to lagarith :

http://www.foochuck.com/video/video-test-1.avi

As you can see the clip does not look progressive - it also looks like it's skipping frames.

If you'd like, I can cut a short clip of the original m2ts file. You mentioned a software I can use to grab a small segment of the m2ts file - could you repost that software / link?