can you cut video to a precise specific length?

[efc1978]

Hello,
I'm adding a rare complete and uncut superior audio soundtrack to video which is shorter than the actual audio eg. between songs the video cuts out band interraction with the audience etc., but the superior audio is full length and uncut, so I want to add black/blank video to extend the video so the rare and superior audio can be left full length and uncut.
So I need to cut the blank/black video which I will insert between songs to specific lengths, eg. I want to add in between songs blank/black video of specific lengths of 15.7 seconds, and 23.6 seconds etc.
I have tried cutting some blank/black .VOB.s with programmes like Boilsoft Video Splitter, but it seems I can not cut to exact lengths I specify?, for example if I cut and save a section I want to keep from 00 to 27.4 seconds the output file will be various tenths of a second different eg. 27.1 seconds instead of 27.4 seconds....,
Is there a way to cut .VOB (which I'm working with) or any other video format precisely to an exact tenth of a second you specify?
Thanks

[hech54]

VideoReDo can get pretty specific while minimizing re-encoding(Smart Rendering).....you can view the timeline and frame by frame(so to speak).
I'd imagine if you want to get more precise you need to re-encode after converting to some sort of lossless AVI file....lossless AVI theoretically should give you the most precise options.

[efc1978]

Thanks I'll give it a try, though I just saw it says trialware is $96, hopefully I don't have to pay for something I'm not even sure will be exactly what I need, could you please recommend programmes I could download for free?, though I presume free programmes won't be as accurate as the one you have recommended...
So from your answer I take it that with most video cutting programmes you often can not specify to an exact 1/10th of a second you want the video to cut to?, and even though it allows you to enter a number of 3 decimal places it still will often cut some tenths of a second different to the actual length you specify?
I'm keeping the video lossless, but as the parts I want to cut are going to be blank/black screen filler gaps between songs it doesn't matter if they have to go through a lossy process!, but they will need to be cut to very accurate lengths.
Thanks

[jagabo]

If this is the only video you need to cut you should be able to complete the job during VideoRedo's free trial period. It gives you frame accurate cuts. For PAL video that gives you 1/25 second precision.

The non frame accurate editors you refer to only cut on key frames and don't reencode so you lose no quality. The benefit of VideoRedo is that it's a smart editor. It allows cuts anywhere but it only reencodes cut GOPs. If you have to reencode the entire video anyway you can use any dumb editor like VirtualDub and AviDemux.

[efc1978]

If this is the only video you need to cut you should be able to complete the job during VideoRedo's free trial period. It gives you frame accurate cuts. For PAL video that gives you 1/25 second precision.

The non frame accurate editors you refer to only cut on key frames and don't reencode so you lose no quality. The benefit of VideoRedo is that it's a smart editor. It allows cuts anywhere but it only reencodes cut GOPs. If you have to reencode the entire video anyway you can use any dumb editor like VirtualDub and AviDemux.

Thanks for the info, yes it's the only video I'll need to edit this way, however I'll need to add the black/blank video screen fillers between each song, so I guess I'll need to make/cut about 15, plus most likely I'll have to do a few trial/error runs to get things right, but yeah will probably need to make/cut around 15 of various lengths, hopefully the trial version will be able to cover it! I can keep the main video source lossless as long as the blank/black filler video I'm cutting can be VOB's.
Thanks for the info.

[manono]

This is for DVD (right?) so you can create your own blank frames for whatever length you like in AviSynth. Demux the DVD video by songs/cells using PGCDemux (I'll assume the DVD has the songs broken up into chapters/cells/songs), and then join them all back together (video M2Vs together with black frame M2Vs) along with the complete (DVD compliant) audio in Muxman.

The scripts used (and which can be opened in most MPEG-2 encoders) might be something like:

BlankClip(length=1031, width=720, height=576, fps=25, pixel_type="YV12")

The 1031 is the number of frames which you can figure out easily by knowing the length of each black-frame segment. Change it to match the length of each black-frame segment. The blank frames will be black by default, although you can make them any color under the rainbow, just about, or change colors at will. Test each script by opening it in VDub to make sure of no mistakes. Go File->File Information to make sure it's the length you want. The already existing video won't be reencoded.

http://avisynth.nl/index.php/BlankClip

[efc1978]

This is for DVD (right?) so you can create your own blank frames for whatever length you like in AviSynth. Demux the DVD video by songs/cells using PGCDemux (I'll assume the DVD has the songs broken up into chapters/cells/songs), and then join them all back together (video M2Vs together with black frame M2Vs) along with the complete (DVD compliant) audio in Muxman.

The scripts used (and which can be opened in most MPEG-2 encoders) might be something like:

BlankClip(length=1031, width=720, height=576, fps=25, pixel_type="YV12")

The 1031 is the number of frames which you can figure out easily by knowing the length of each black-frame segment. Change it to match the length of each black-frame segment. The blank frames will be black by default, although you can make them any color under the rainbow, just about, or change colors at will. Test each script by opening it in VDub to make sure of no mistakes. Go File->File Information to make sure it's the length you want. The already existing video won't be reencoded.

http://avisynth.nl/index.php/BlankClip

Thanks will also check that out...
My intentions were already to edit the main DVD using PGCDemux and Muxman, I'll look into this method you've suggested also when it comes to working with the black frames.
Thanks for the info.

[efc1978]

This is for DVD (right?) so you can create your own blank frames for whatever length you like in AviSynth. Demux the DVD video by songs/cells using PGCDemux (I'll assume the DVD has the songs broken up into chapters/cells/songs), and then join them all back together (video M2Vs together with black frame M2Vs) along with the complete (DVD compliant) audio in Muxman.

The scripts used (and which can be opened in most MPEG-2 encoders) might be something like:

BlankClip(length=1031, width=720, height=576, fps=25, pixel_type="YV12")

The 1031 is the number of frames which you can figure out easily by knowing the length of each black-frame segment. Change it to match the length of each black-frame segment. The blank frames will be black by default, although you can make them any color under the rainbow, just about, or change colors at will. Test each script by opening it in VDub to make sure of no mistakes. Go File->File Information to make sure it's the length you want. The already existing video won't be reencoded.

http://avisynth.nl/index.php/BlankClip

Hi,
I've downloaded AVIsynth and Virtualdub.
I have then opened notepad and cut and paste in the secific command you wrote eg.
BlankClip(length=1031, width=720, height=576, fps=25, pixel_type="YV12")
Then "saved as" using the name blankvideo.avs.
But when I open VirtualDub and try to open with Open Video File it says AVI Import Filter error (80040154).
Also when I try to open with Run Script it says Error during script execution at line 1, column 10 : Variable "Blank Clip" not found and shows the command info you typed which I cut and pasted into Notepad.
I also tried above with a video pathway to an .AVI in the script, but got the same results...
Can you please point me in the right direction from here....
However when I open the Notepad .avs in TMPGEnc Video Mastering Works 5 it plays in the preview box a 41 second blank video!, which I guess is partial success....., but I'm thinking it should be opening in VirtualDub?
However I have no way to export that as an .m2v (you mentioned I could combine a blank .m2v with my main video .m2v I'll get from my DVD .VOB's).
I also tried above with a video pathway to an .AVI in the script, but got the same results...
When I went to open Virtualdub and open the script through File-File Information as you suggested the File Information part is blanked out, so I can't click on it....
I know this is probably all very simple, can you please point me in the right direction from here....
This is all new to me and am stuck here at the moment!
Thanks

[manono]

It should open in VDub, yes, with File->Open Video File, and the fact that it opens in TMPGEnc means AviSynth is installed correctly and the script is good. I also just tested that exact same script in VDub myself, just to be sure, so I suppose VDub must not have been installed correctly at your end. The reason I mentioned using it was so the script could be tested before encoding. It throws a useful error message if something's wrong with the script, where other applications or programs might not.

My guess is that you didn't install the frameserver support. It's easy. Here's a simple little guide that should still work after all these years:

https://forum.videohelp.com/threads/37390-How-to-frameserve-%28dvd2avi-VirtualDub-Avisynth-VFAPI%29

A little down from the top you want the 'Installation hints' and the stuff about Auxsetup.exe.

However I have no way to export that as an .m2v (you mentioned I could combine a blank .m2v with my main video .m2v I'll get from my DVD .VOB's).

If the script is good and for the right length, then any MPEG-2 encoder I've ever used should also accept the script and encode the black screen for the length you want. Won't TMPGEnc encode it for you? If not, why not? A colorspace issue? I chose YV12 because some encoders want that. If not TMPGEnc, HCEnc surely will encode it. Or CCE. Or Procoder. I also tested using CCE to encode the 41 second black screen, and it worked as I expected.

I haven't used TMPGEnc for a very long time and can't help you with it, although others here can.

Once you get your black screen M2V/MPVs made, open the first video in Muxman and use the 'Add' button to add the successive black videos and song videos. When all done with the videos add the audio. I suppose it's too much expect for it to work perfectly the first time, because even one black screen M2V off by a tenth of a second will throw off the audio synch for everything afterwards. But if you're careful it should come out alright in the end. Heck, if you get adventuresome, instead of boring black screens you could insert a red screen. Or use ImageSource to encode an M2V of a photograph.

http://avisynth.nl/index.php/ImageSource

Another thing, at 25 frames per second each frame lasts for 40 ms, so it'll be difficult to get them to exactly match the in between song stuff. If one black screen video is a bit long, maybe make the next one a bit short so it averages out in the end to keep the synch. Good luck.

[efc1978]

It should open in VDub, yes, with File->Open Video File, and the fact that it opens in TMPGEnc means AviSynth is installed correctly and the script is good. I also just tested that exact same script in VDub myself, just to be sure, so I suppose VDub must not have been installed correctly at your end. The reason I mentioned using it was so the script could be tested before encoding. It throws a useful error message if something's wrong with the script, where other applications or programs might not.

My guess is that you didn't install the frameserver support. It's easy. Here's a simple little guide that should still work after all these years:

https://forum.videohelp.com/threads/37390-How-to-frameserve-%28dvd2avi-VirtualDub-Avisynth-VFAPI%29

A little down from the top you want the 'Installation hints' and the stuff about Auxsetup.exe.

However I have no way to export that as an .m2v (you mentioned I could combine a blank .m2v with my main video .m2v I'll get from my DVD .VOB's).

If the script is good and for the right length, then any MPEG-2 encoder I've ever used should also accept the script and encode the black screen for the length you want. Won't TMPGEnc encode it for you? If not, why not? A colorspace issue? I chose YV12 because some encoders want that. If not TMPGEnc, HCEnc surely will encode it. Or CCE. Or Procoder. I also tested using CCE to encode the 41 second black screen, and it worked as I expected.

I haven't used TMPGEnc for a very long time and can't help you with it, although others here can.

Once you get your black screen M2V/MPVs made, open the first video in Muxman and use the 'Add' button to add the successive black videos and song videos. When all done with the videos add the audio. I suppose it's too much expect for it to work perfectly the first time, because even one black screen M2V off by a tenth of a second will throw off the audio synch for everything afterwards. But if you're careful it should come out alright in the end. Heck, if you get adventuresome, instead of boring black screens you could insert a red screen. Or use ImageSource to encode an M2V of a photograph.

http://avisynth.nl/index.php/ImageSource

Another thing, at 25 frames per second each frame lasts for 40 ms, so it'll be difficult to get them to exactly match the in between song stuff. If one black screen video is a bit long, maybe make the next one a bit short so it averages out in the end to keep the synch. Good luck.

Thanks for the info, I'm sure I'll eventually get it sorted out, but unfortunately I don't think I'll get a chance to work on this until after the weekend...
Yeah I know it will be hard to get precisely accurate, but even if I have to end up cutting small parts of the audio will be OK, I don't think 1 tenth of a second of crowd cheering between songs will matter that much!
mmm I looked and I can't see how to export as .m2v with TMPGEnc, there's a bunch of possibilities but not .m2v.
Thanks again.

[jagabo]

But when I open VirtualDub and try to open with Open Video File it says AVI Import Filter error (80040154).

You must use 32 bit VirtualDub with 32 bit AviSynth, or 64 bit VirtualDub with 64 bit AviSynth. You can't use a mix the two.

[efc1978]

But when I open VirtualDub and try to open with Open Video File it says AVI Import Filter error (80040154).

You must use 32 bit VirtualDub with 32 bit AviSynth, or 64 bit VirtualDub with 64 bit AviSynth. You can't use a mix the two.

That's what this error means?, I always download 64bit when possible, I'll have to take a look when I'm back at my PC.
Thanks

[smrpix]

Annother opinion:
The minute you move away from pure smart-rendering, why not just use a real NLE?

Download the trial of Premiere Pro,
import the vobs directly,
add your good audio on a new track,
split and shift the original audio and video to align them with the new audio -- use the waveforms to check sync,
delete the old audio
export as mpeg for dvd.
Use the output as a source for authoring in AVStoDVD (no need to re-render.)

The advantage is that you can see what you're doing, the quality is no worse than frameserving and combining, and you avoid math and (probable) sync issues. You also avoid installing incompatible versions of plug-ins.

[efc1978]

Annother opinion:
The minute you move away from pure smart-rendering, why not just use a real NLE?

Download the trial of Premiere Pro,
import the vobs directly,
add your good audio on a new track,
split and shift the original audio and video to align them with the new audio -- use the waveforms to check sync,
delete the old audio
export as mpeg for dvd.
Use the output as a source for authoring in AVStoDVD (no need to re-render.)

The advantage is that you can see what you're doing, the quality is no worse than frameserving and combining, and you avoid math and (probable) sync issues. You also avoid installing incompatible versions of plug-ins.

Hi, I'm not familiar with some of the shortened method names used sometimes, but I have worked on a few projects before simply synching/aligning new to old audio times, but I don't understand the way you're describing, particularly how I would do that when I add in the new blank/black DVD frames to make up for missing video, I'm not sure if you didn't understand my project/goal description, or probably I just don't have the knowledge to understand exactly what you are saying!
Thanks

[jagabo]

But when I open VirtualDub and try to open with Open Video File it says AVI Import Filter error (80040154).

You must use 32 bit VirtualDub with 32 bit AviSynth, or 64 bit VirtualDub with 64 bit AviSynth. You can't use a mix the two.

That's what this error means?

No, I think it's a generic "couldn't open file" error code. But that's what happens if you don't have AviSynth installed. And if you are using 64 bit VirtualDub and only have 32 bit AviSynth installed, or vice versa, that's what happens. The 64 bit and 32 bit video systems are completely separate and don't know each other exists.

[smrpix]

You don't need to add anything. Sync up your clips to your new audio like this:


[Attachment 32260 - Click to enlarge]

Then delete the original audio like this:


[Attachment 32261 - Click to enlarge]

Then render out to mpeg2 dvd (as Premiere calls it.) The blank spaces in the video timeline are "automatically" rendered as black.

[efc1978]

But when I open VirtualDub and try to open with Open Video File it says AVI Import Filter error (80040154).

You must use 32 bit VirtualDub with 32 bit AviSynth, or 64 bit VirtualDub with 64 bit AviSynth. You can't use a mix the two.

That's what this error means?

No, I think it's a generic "couldn't open file" error code. But that's what happens if you don't have AviSynth installed. And if you are using 64 bit VirtualDub and only have 32 bit AviSynth installed, or vice versa, that's what happens. The 64 bit and 32 bit video systems are completely separate and don't know each other exists.

OK will download both again when I get back to this after the weekend, like I said I always download 64bit when I see the option, so not sure how I would have ended up with one 32bit and another 64 but I'll check it.
Thanks

[efc1978]

You don't need to add anything. Sync up your clips to your new audio like this:


[Attachment 32260 - Click to enlarge]

Then delete the original audio like this:


[Attachment 32261 - Click to enlarge]

Then render out to mpeg2 dvd (as Premiere calls it.) The blank spaces in the video timeline are "automatically" rendered as black.

Thanks will check this out also, new to video editing (but pretty experienced in various audio editing), feel a bit flooded at the moment with info about stuff I've never worked with before so sorry if I'm misunderstanding suggestions/advice/help.
Thanks

[Cornucopia]

There is an "elephant in the room" that needs to be addressed one way or another: Timebase.

PAL video is 25FPS for 25p material = 1/25th sec per frame = 40mSec.

For MUSICIANS and their synchronization awareness, a de-synchronization is noticeable after + or - 10-15mSec. Particularly with fast attack/transient instrumentation such as percussion or guitars.

Therefore, it might not be possible to get exact enough clean synchronization by shifting the video (since it cannot be shifted any less than its own atomic timebase). This means a de-synchronization can be up to + or - 20mSec (1/2 a frame's worth), still more than what might be noticeable to musicians.

Now, there are some mitigating factors:
You could be using Interlaced material, in which case it might be possible to shift by 1 field (20mSec, for a de-synch of up to 10mSec either way), but this brings in complications in dealing with interlaced line matchup.
You could be using 50p material, in which case the timebase is a true full 1/50th sec or 20mSec, which would bring the de-synch (+ or - 10mSec) within the unnoticeable range for even most musicians.
Your audience might not be musicians, so they may not be as accutely aware of desynchronization, so you have more de-synch leeway.

However, if you want things to be DEAD ON EXACT, the only way to truly do this is NOT ONLY to shift your video, but to make cuts to your audio and shift them as well, since the audio's timebase (at 48kHz) is 1/480000th of a sec or ~0.021mSec. This could/should be done in a good DAW (my fave is ProTools) that has video capability. There are ways to make the shifts seamless (I have done this many times), so it's not such a worrisome thing to "violate" the good audio stream.

Ultimately it is up to you (and your audience) as to what is acceptable or not, but I felt the topic needed to be fully explored.

Scott

[efc1978]

There is an "elephant in the room" that needs to be addressed one way or another: Timebase.

PAL video is 25FPS for 25p material = 1/25th sec per frame = 40mSec.

For MUSICIANS and their synchronization awareness, a de-synchronization is noticeable after + or - 10-15mSec. Particularly with fast attack/transient instrumentation such as percussion or guitars.

Therefore, it might not be possible to get exact enough clean synchronization by shifting the video (since it cannot be shifted any less than its own atomic timebase). This means a de-synchronization can be up to + or - 20mSec (1/2 a frame's worth), still more than what might be noticeable to musicians.

Now, there are some mitigating factors:
You could be using Interlaced material, in which case it might be possible to shift by 1 field (20mSec, for a de-synch of up to 10mSec either way), but this brings in complications in dealing with interlaced line matchup.
You could be using 50p material, in which case the timebase is a true full 1/50th sec or 20mSec, which would bring the de-synch (+ or - 10mSec) within the unnoticeable range for even most musicians.
Your audience might not be musicians, so they may not be as accutely aware of desynchronization, so you have more de-synch leeway.

However, if you want things to be DEAD ON EXACT, the only way to truly do this is NOT ONLY to shift your video, but to make cuts to your audio and shift them as well, since the audio's timebase (at 48kHz) is 1/480000th of a sec or ~0.021mSec. This could/should be done in a good DAW (my fave is ProTools) that has video capability. There are ways to make the shifts seamless (I have done this many times), so it's not such a worrisome thing to "violate" the good audio stream.

Ultimately it is up to you (and your audience) as to what is acceptable or not, but I felt the topic needed to be fully explored.

Scott

I'm working with PAL video 25 frames per second.
Kind of following everything but not sure as I've never worked much with video....
Usually when synching one audio speed to another audio speed, for example if I take 1 particular audio file I can pretty much match the 'speed' (tempo) first to last snare starting and ending by 0secs.01 which is pretty accurate, seeing as video can edit to 1 frame PAL .04 per second, it won't be noticeably off speed to most people if I cut and start to synch the audio to video at the start of each song, this should only need a .04 per second re synch between each song!?
It's 3am and I'm exhausted so I probably don't even make sense at this point of trying to process it all!, but I feel I'm on the right track!
Thanks.

[efc1978]

You don't need to add anything. Sync up your clips to your new audio like this:


[Attachment 32260 - Click to enlarge]

Then delete the original audio like this:


[Attachment 32261 - Click to enlarge]

Then render out to mpeg2 dvd (as Premiere calls it.) The blank spaces in the video timeline are "automatically" rendered as black.

Comparing to the other methods suggested this looks too simple to be true!?, so this will simply just add the black video frames for parts that have no audio?
I don't understand some of the terminology you have used, is this a lossless process?, as I want to keep it lossless...
Also it says Premiere Pro costs $600, a little expensive for 1 project!, will I be able to complete this whole project using the trial version?, is the time length the trial lasts the only limitation?, or limited to a certain number of edits allowed or something like that?
Thanks

[manono]

...is this a lossless process?

Of course it isn't lossless. "Then render out to mpeg2 dvd" means reencoding it.

I think Cornucopia brought up a good point. Your DVD source is speeded up PAL 25fps. You haven't said anything about your good audio but the chances are real good the songs in it won't match up with the video for your DVD, even if they're both from the same concert. Keeping the video lossless might mean having to speed up your good audio to match. Or have you already tested this out?

I'm less worried about possible synch problems, although it may require a lot of time and effort to get it right.

[efc1978]

...is this a lossless process?

Of course it isn't lossless. "Then render out to mpeg2 dvd" means reencoding it.

I think Cornucopia brought up a good point. Your DVD source is speeded up PAL 25fps. You haven't said anything about your good audio but the chances are real good the songs in it won't match up with the video for your DVD, even if they're both from the same concert. Keeping the video lossless might mean having to speed up your good audio to match. Or have you already tested this out?

I'm less worried about possible synch problems, although it may require a lot of time and effort to get it right.

Oh well if it's lossy then I won't be using that process, but thanks for the suggestion... I should have known it was lossy but I'm a bit overwhelmed with info at the moment, thinking of the process it makes sense though yes.
The audio is multitrack to DAT, and I have highly suspected all along that most probably I'll need to speed change the audio of each song very slightly, but I'm sure any change will be so tiny the pitch will still sound correct.
I do not know exactly what percentage PAL speed is off but I know it's likely going to be a bit different to DAT.
*strange, I just read PAL is sped up 4%, I have to say this is hard to believe, there's no way some of the music videos I have in PAL sound 4% off speed (80 cents off pitch).

[manono]

I have to say this is hard to believe...

Believe it. Unless the audio on PAL DVDs has been pitch-corrected, it's a semi-tone too high and drives musicians and people with perfect pitch up a wall.

[smrpix]

As I said in my earlier post, the moment you move away from smart rendering, you are re-encoding, incurring loss, whether you do it the "simple" way or the more convoluted way.

The loss with Premiere directly to mpeg2 is quite minimal. First generation you have to look pretty hard to see it.

If you want, you can export from Premiere to uncompressed, but you'll still need to re-encode the output to mpeg2 to make a new dvd. If you're using Vdub, TMPGEnc, Avisynth, CCE, etc., you are still incurring loss.

Or if you want to go the smart rendering route, (See post #2,) you can use Premiere to quickly and precisely determine the exact lengths of black you need to create for combining in VideoReDo (or TMPGEnc Smart Renderer.)

As for PAL speedup. See if it's really an issue before stressing over it too much. You will know pretty quickly if you have it and there are both easy and hard ways to fix that too.

(BTW, I'm not a hack for Adobe, you can do this in Vegas or Edius just as easily. Also, Premiere is not $600, it's $20/month with a one month free trial.)

[efc1978]

As I said in my earlier post, the moment you move away from smart rendering, you are re-encoding, incurring loss, whether you do it the "simple" way or the more convoluted way.

The loss with Premiere directly to mpeg2 is quite minimal. First generation you have to look pretty hard to see it.

If you want, you can export from Premiere to uncompressed, but you'll still need to re-encode the output to mpeg2 to make a new dvd. If you're using Vdub, TMPGEnc, Avisynth, CCE, etc., you are still incurring loss.

Or if you want to go the smart rendering route, (See post #2,) you can use Premiere to quickly and precisely determine the exact lengths of black you need to create for combining in VideoReDo (or TMPGEnc Smart Renderer.)

As for PAL speedup. See if it's really an issue before stressing over it too much. You will know pretty quickly if you have it and there are both easy and hard ways to fix that too.

(BTW, I'm not a hack for Adobe, you can do this in Vegas or Edius just as easily. Also, Premiere is not $600, it's $20/month with a one month free trial.)

Hi,
Thanks for the info.
I believe though I am keeping the video in this project lossless by taking my source DVD .VOB's and using PGCDemux to split the main video from the audio, so I can then use the .m2v's to combine the video with the new audio in Muxman, which will keep the video lossless. As for the blank/black vide frames I need to create, it doesn't matter if that has been through a lossy process as it's just 'black'.

[efc1978]

I have to say this is hard to believe...

Believe it. Unless the audio on PAL DVDs has been pitch-corrected, it's a semi-tone too high and drives musicians and people with perfect pitch up a wall.

I'm not understanding how I've read so many times on the internet ALL PAL is sped up 4%...
I would have noticed on my music DVD's if something is sped up 4%, that's a fact, it would be very obvious to me...
OK I just took a music DVD filmed in 1992, a particular song on the DVD is 3min:04secs.6, I then checked the same song on the official Compact Disc release and it's the same, I then checked the same song from the same show made with a DAT recorder in the audience, and it's exactly the same, all 3 are the same time length to 1 tenth of a second, so how can ALL PAL be sped up 4%.

A second example, I just took 2 PAL DVD's purchased here in Australia, one is a transfer of a 1991 release VHS, the other a different titled 2001 DVD release, the video clip of the song on both DVD's is exactly the same length 3min:44secs.8, the same song on the Compact Disc release is 3min:44.5, just 3 tenths of a second different, a lot of musicians wouldn't even think that was much, and it's far from 4% different, the small difference may also be because of the process of mastering the audio CD and the different sourced video clip on the DVD, or many other factors.
Even if the DVD had been pitch corrected the length would be different to the CD, and surely the video on the DVD could not have been sped to the correct length on the CD.., and surely the CD 'album' has not been incorrectly sped up to match the supposed 4% fast PAL DVD.

Really I don't understand why I have read so many times ALL PAL stuff is 4% fast....., it's not.

[racer-x]

It is 4% faster if it was NTSC source converted to PAL.

As for your problem, if you want to keep it as lossless as posible, you'll need to use a smart-render editor like VideoRedu. Vegas can also smart render mpeg-2 sources, so that is another option for you.

[jagabo]

It is 4% faster if it was NTSC source converted to PAL.

More precisely, if it was a film source. Interlaced NTSC is more likely to have gone through a field blending with no run time change.

[smrpix]

There are a number of different ways PAL audio is handled depending on the acquisition method. Earlier, I suggested if you don't see a problem, don't worry about it. This stuff is useful, often necessary to know -- but at times folks around here, myself included can be "too" helpful -- which can be confusing.