How Do You Convert PAL AC3 5 Channel to NTSC?

[allanc448]

I have managed to convert a number of PAL 2 channel AVIs to NTSC with satisfactory results.
Now I have a problem with a PAL AVI that has AC3 5 channel that I am trying to convert as follows:
1 Extract audio as PAL WAV using VirtualDub
2. Convert PAL WAV to NTSC using BeSweet GUI (WAV to WAV).
3. I then convert the video as follows:
Open PAL AVI
Video / Direct Stream Copy
Frame Rate - change to 23.976
Audio - WAV Audio that is output from step 2 above
Save as AVI

The resulting videos audio seems to be synced and mostly satisfactory. However, there are a number of areas in the audio (3-5 seconds each) where you can understand the voice but it sounds like a 'hollow echo'. That is the best that I can describe it.

All advice is appreciated.

[FulciLives]

If you convert the video using the DGPulldown method then the audio does NOT have to be changed.

See this here thread ---> CLICK HERE

Basically you encode the PAL AVI at NTSC DVD resolutioin (720x480 for instance) but leave the frame rate at 25fps ... then you use DGPulldown which does a "trick" that makes the video "appear" to be standard 29.970fps NTSC

However the video still has the same running time as the original PAL video clip ... hence no reason to change the audio.

It works rather well and is a VERY nice way to do PAL to NTSC conversion!

- John "FulciLives" Coleman

[allanc448]

If you convert the video using the DGPulldown method then the audio does NOT have to be changed.

See this here thread ---> CLICK HERE

In the end, I will be watching the video on a NTSC standalone DVD player that does not play PAL.
At this point I am more concerned about the quality of the video and audio compared to the length of time to accomplish the conversion.
I am not very familiar with my old steps and basically worked thru instructions found at this site, asked questions, etc.
Are there any 'downsides' to the new method that you are recommending that I should know about?
TIA.

[FulciLives]

The new method creates a new compatable NTSC video. It will work on any NTSC only DVD player.

The whole point of using this new method is that the video duration does not change ... therefore no need to adjust the audio.

It is easy to adjust audio when you are working with standard STEREO audio but 5.1 DD can be tricky to adjust.

I suggest you try the new method as it leaves the audio alone.

- John "FulciLives" Coleman

[allanc448]

The new method creates a new compatable NTSC video. It will work on any NTSC only DVD player.

The whole point of using this new method is that the video duration does not change ... therefore no need to adjust the audio.

It is easy to adjust audio when you are working with standard STEREO audio but 5.1 DD can be tricky to adjust.

I suggest you try the new method as it leaves the audio alone.

- John "FulciLives" Coleman

I will try it.
Thank you once again.

[allanc448]

If you convert the video using the DGPulldown method then the audio does NOT have to be changed.

See this here thread ---> CLICK HERE

Is there a step-by-step guide?
A lot of the thread is over my head.

TIA.

[Abond]

I think that this is not suitable for you as you do not reencode the avi. DGPulldown is for mpeg2 files, I think it will not work with avi files.

[Cornucopia]

Abond--allanc448 said (s)he was going to watch on a standalone DVD player, so what FulciLives said is still quite valid and workable.

Scott

[Abond]

There are quite a lot standalone DVD players that play avi files. I don't know really but the first post show that he is converting the fps with DirectStreamCopy - it is possible to do it this way and the result will be avi at 23.976 without reencoding the video. That is why I thought he need avi in NTSC format.

[allanc448]

Abond--allanc448 said (s)he was going to watch on a standalone DVD player, so what FulciLives said is still quite valid and workable.

Scott

They are to played on a standalone DVD player (Pioneer) that
* does not * play AVI
Does that mean that the suggested method won't work.
BTW - allanc448 is Allan C (male).

[Abond]

Well, then DGPulldown is the way to go. The basis is very simple - encode the video at 25 fps with NTSC resolution. When ready apply DGPulldown to make 25fps - 29.97 fps. The duration of the video remains the same. Author the video with original ac3 audio and you are done.

[allanc448]

Well, then DGPulldown is the way to go. The basis is very simple - encode the video at 25 fps with NTSC resolution. When ready apply DGPulldown to make 25fps - 29.97 fps. The duration of the video remains the same. Author the video with original ac3 audio and you are done.

So, I should use VirtualDub Mod and demux the AC3.
As I said a lot of this this is still very unfamiliar to me as I try to use guides and do not really understand the TmpGenc. Could you please advise as to what specific settings I need to encode the video as per above?
Thanks in advance.

[Abond]

Usually I am working with avisynth, but maybe I will try to give you some advices and please excuse me if I am going sometimes wrong.
Until you are at VDubMod demuxing the audio, demux also the video to produce an silent avi.
Load it in TMPG as video source and select stream type ES video only output.
Load DVD pal template. Now load a template called unlock.
Settings->Video ->720x480
Video format-> Unspecified (I do notn know how it will work)
Frame rate 25
Set the bitrate (use bitrate calculator for this) Rate control 2pass VBR Encode mode Non-interlace, Motion search precision->Motion estimate search(fast)
Advanced tab->Source type non-interlace, Field order Top field first,Source aspect 1:1 VGA,Arrange method : full screen keep aspect ratio 2.
Gop structure tab-> Detct scene change checked only. Now here is a bit weird as I do not know -> we are trying to convert to NTSC, but currently are encoding mutant stream. For NTSC max frames in gop are 18, for pal are 15. Seems to leave it as 15 frames in gop should not hurt.
OK. Start.
Ah, the last if you want to shut down the comp after encoding, you are clicking not start, but file->save project. Then file->Batch encode->add the project->run. Then in the upper left corner of the main TMPG window you have a box Shutdown after encoding -> check it.
After it is done Run DGPulldow select 25->29.97. Multiplex with the ac3 in ImagoMpegMuxer. Author with what you like and burn.

[allanc448]

Usually I am working with avisynth, but maybe I will try to give you some advices and please excuse me if I am going sometimes wrong.

Hopefully, I will able to try it overnight since the TmpGenc encoding takes about 4-5 hours.
Thank you once again.

[allanc448]

Usually I am working with avisynth, but maybe I will try to give you some advices and please excuse me if I am going sometimes wrong.
Multiplex with the ac3 in ImagoMpegMuxer. Author with what you like and burn.

Sorry, but I don't understand this part.
The demux in VirtualDub provides the AC3.
Don't the steps in TmpGenc generate the M2V?
Usually, I then author and burn.
Why is ImagoMpegMux necessary?

[Abond]

Well, my advice was to use ONLY the video in TMPG, not the audio. If you are aiming for DVD ac3 5.1 will provide somewhat beter quality and of course 5.1 sound.
The other thing is that the video stream we are encoding in TMPG is not very complaint - it will be after DGPulldown. Then I don't know how TMPG will encode the audio, but if you want mp2 sound use BeSweet+BesweetGUI to reencode ac3-mp2 - mp2 quality I think will be better than TMPG audio engine.

[NamPla]

Well, then DGPulldown is the way to go. The basis is very simple - encode the video at 25 fps with NTSC resolution. When ready apply DGPulldown to make 25fps - 29.97 fps. The duration of the video remains the same. Author the video with original ac3 audio and you are done.

How does it make 25fps into 29.97fps without changing the duration?

[FulciLives]

Well, then DGPulldown is the way to go. The basis is very simple - encode the video at 25 fps with NTSC resolution. When ready apply DGPulldown to make 25fps - 29.97 fps. The duration of the video remains the same. Author the video with original ac3 audio and you are done.

How does it make 25fps into 29.97fps without changing the duration?

It's magic !!!

Go read the THREAD at doom9 if you want: CLICK HERE

So far I tried it with one PAL DVD source and the results were most excellent!

- John "FulciLives" Coleman

[NamPla]

Yeah but what does this mean?

Suppose you have this where the top line is the top field and each letter is a new picture:

A C E
B D F

If you pulldown the first frame, you get:

A A C E
B D F

...which gives a temporal sequence of:

A B A D C F E

...which is all screwed up. If you have progressive video it looks like this:

A A B C
A B C

which gives a temporal sequence of:

A A A B B C C

Oh right yes I get it now. It adds an extra frame at the start. But what happens when you get to "Z"?

[allanc448]

Well, my advice was to use ONLY the video in TMPG, not the audio. If you are aiming for DVD ac3 5.1 will provide somewhat beter quality and of course 5.1 sound.
The other thing is that the video stream we are encoding in TMPG is not very complaint - it will be after DGPulldown. Then I don't know how TMPG will encode the audio, but if you want mp2 sound use BeSweet+BesweetGUI to reencode ac3-mp2 - mp2 quality I think will be better than TMPG audio engine.

In the past, I would extract the AC3 in VirtualMod and not do anything more with it until the authoring in DVD Author GUI.

[midnightsun]

Well, then DGPulldown is the way to go. The basis is very simple - encode the video at 25 fps with NTSC resolution. When ready apply DGPulldown to make 25fps - 29.97 fps. The duration of the video remains the same. Author the video with original ac3 audio and you are done.

How does it make 25fps into 29.97fps without changing the duration?

The same way you make 23.976f/s 29.97f/s leaving the running time untouched: it inserts repeat-field flags

[jhr1948]

been looking around for this question, here and doom9. this seems like a better way to convert, however, when the source is a pal interlaced dvd, is the conventional (xesdeeni's script) still better to use? the audio stays the same xesdeeni's method when the source is interlaced.

i'm just trying to get an improved picture, changing frames can lead to slighty jerky pictures

[FulciLives]

been looking around for this question, here and doom9. this seems like a better way to convert, however, when the source is a pal interlaced dvd, is the conventional (xesdeeni's script) still better to use? the audio stays the same xesdeeni's method when the source is interlaced.

i'm just trying to get an improved picture, changing frames can lead to slighty jerky pictures

You can use the AviSynth FieldDeinterlace() command (part of the decomb package plug-in) to deinterlace as you re-encode it to 720x480 25fps thus allowing you to use the DGPulldown method. NTSC is a whole other story but deinteracing PAL (when converting to NTSC) seems to NOT be that big of a deal in terms of the quality of the image, at least in my experience.

- John "FulciLives" Coleman