I had an AVI that was 25 fps, so I adjusted the frame rate using VDUB to make the AVI 23.976. The problem I have, is that after encoding the video using a NTSCfilm KVCD template, the .M1V is at least 5 seconds shorter then the VDUB converted AVI.
I also changed the audio by extracting it using VDUB and adjusting the framerate using Besweet and converting to mp2 in HeadAC3he. The converted mp2 and converted AVI time match. Only issue is after encoding, the movie is 5-6 seconds shorter. Any input would be nice, thanx

NEVER DO THAT ! Use "ChangeFps()" command of avisynth.

For the rest it's not a problem. Splitters always have difficulty to give the exact length of a VBR MPEG1 like we do in KVCD. Some even give a time that is half than the real one !

You're much better off using AssumeFPS(). With ChangeFPS(), the motion is not smooth as frames are added/removed. With AssumeFPS(), the playback speed is changed and no frames will be added or removed.

So you are saying that I should add "ChangeFps()" command of avisynth to the script using the original 25fps avi as source and then run thru TMPGenc with a NTSCFilm template? How do I add this command? Also, what does assumefps() do? I am using your V3 script. Thanx Dialhot

AviSource("C:\Eric\VCD\Test.avi",false)
SwapUV()
#AssumeFPS()
#ChangeFps()
BlindPP(cpu=4)
Blockbuster(method="noise",detail_min=1,detail_max =3,variance=0.1,seed=1)
ATC(2,3,5,0.5,false)
Gripcrop(528, 480, overscan=1, source_anamorphic=false)
GripSize(resizer="LanczosResize")
Undot()
TemporalSoften(2,7,7,3,2)
DCTFilter(1,1,1,1,1,1,0.5,0)
#Blockbuster(method="noise",detail_min=1,detail_ma x=10,variance=0.3,seed=5623)
GripBorders()

In doing 25->23.976 it's okay, you're right. But never do 25->29.970 this way because the speed change will be very big !

Note: with AssumeFps the sound will by pitched down by 1/8 of tone. It can be a problem if you have a musician hearing

@Eric
You should put some value in the AssumeFps (or ChangeFps) brackets, for example:
AssumeFps(23.976)
The difference is that with AssumeFps all the frames are encoded, but they will be played back slowlier. With ChangeFps the speed will be the same but at every 25 frames 1 frame will be deleted, to adjust 23.976 frames per second. If your system support PAL resolutions and framerate do not convert - it is evil.
@Dialhot
(OT)Just for small reference a get a short look in the optimal AVI-KVCD scripts and found in all except V2 GripCrop(HEIGHT, WIDTH,....) I've thought it is width,height...
Copy-paste, eh

Like this? Then I just put the script into TMPGenc? What makes this any better then changing source frame rate in VDUB and processing all frames?

AviSource("C:\Eric\VCD\Test.avi",false)
SwapUV()
AssumeFPS(23.976)
#ChangeFps()
BlindPP(cpu=4)
Blockbuster(method="noise",detail_min=1,detail_max =3,variance=0.1,seed=1)
ATC(2,3,5,0.5,false)
Gripcrop(528, 480, overscan=1, source_anamorphic=false)
GripSize(resizer="LanczosResize")
Undot()
TemporalSoften(2,7,7,3,2)
DCTFilter(1,1,1,1,1,1,0.5,0)
#Blockbuster(method="noise",detail_min=1,detail_ma x=10,variance=0.3,seed=5623)
GripBorders()

Absolutly . I had fixed the post.

@Eric
Make it both ways, and you will have a chance to tell us.

You can choose the OTA options of Besweet to convert the 25fps to 23.976fps audio speed. But if you are not in Besweet you can choose the following way....:

Change your AVS as following:


So as you see, delete the ",false" in your import-line so the audio also will be treated. Adding the ",true" to AssumeFPS() keeps the audio in sync by rising/lowering the Khz of your audio which will be served. But finally the Audio won't be at 44.100khz anymore .... but we'll fix this in HeadAC3che later ....

So, .... open this .avs in Vdub ... go to the audio tab and choose "no compression" finally choose in Vdub "Save WAV as..." to safe your well sync'ed audio to HD as WAV ...

Now as your WAV in this state doesn't run at 44.100khz anymore you load this WAV into HeadAC3che and do your mp2 encoding settings INCLUDING choosing .... "Resample Audio to: 44.100"!!!!! Maybe its called in there "Convert Audio..." well you'll recognise. And continue doing your mp2 conversion as known ...

After this "Audio Session" you go back to your .avs and insert again the "..., false" to your Import Line. Continue with your video encoding as you do it in regular.

Its also possible to add a "ResampleAudio(44100)" at the end of the .avs so you don't need to do the conversation in HeadAC3che...
BUT!: I figured out that the resampling quality of Avsiynth isn't that good in comparison to HeadAC3che.

I hope this helps...

Inc.

PS: Just converting the audio of a 25.000 FPS to 23.976 ... will let you hear (as DialHot said) a recognisable low tone audio.
You can avoid this by using Goldwaves internal tone pitch option to pitch the tone of the WAV you safed without resizing the audio .... try the quality.

Well, the kvcd encoded movie still shows a <5 sec diff. then the mp2 file. I havn't muxed them yet, but we shall see.....thanx for eveyones help.

Incredible:

Your trick for converitng PAL .avi's to NTSC Film by importing the audio into VDUB and then saving it uncompressed is, well, incredible!

I was wondering if the same trick was possible when starting with a DVD source? The .d2v file itself has no sound in it, so mpegsource() doesn't help. Is there such a function as WavSource() or something like that to import the .wav file created with DVD2AVI?

Thanks,
Icarus

Never do a wav with DVD2AVI !

Always demux the audio stream with DVD2AVI and then, if you need a wav, convert that audio stream in wav with headac3che or besweet.

Yikes! I've been using DVD2AVI wrong this whole time...

When I demux the audio with DVD2AVI, what format will the audio be in?

Regardless - do you have an answer to my first question: is there a way to open audio from a DVD rip in an avisynth script?

Thanks,
Icarus

Yes, there is a WavSource() function.

Not sure what you'd use for other audio formats though

__________________
MisterMickster

In ac3. DVD2AVI doesn't handle the DTS audio streams.

The newer versions should demux DTS as well, at least the decode fixed one that Donald Graft released.

If you use his version, you must use his fixed version of MPEG2DEC3.dll to avoid sync problems!

http://neuron2.net/fixd2v/decodefix.html

Dialhot:

While we are on the subject, may I ask why it is such a bad idea to save the .wav file from DVD2AVI? I have always done this before and have been pleased with the results.

I guess it's something I have to experience to understand - I will save 2 versions of the audio and let you know if my ears are sensitive enough to tell the difference.

Thanks,
Icarus

I'm not Dialhot but I'll give my two cents anyway.

I wouldn't trust DVD2AVI in downmixing the audio. The tool was simply not created for that. Another point is that you shouldn't do a AC3->WAV->MP2 conversion since there'll be rounding errors on the way (this was confirmed by DSPGuru, the author of BeSweet). For best quality, go straight from AC3 to MP2.

Boulder has said all what you have to know