Another Audio synch question

[Steve]

I've been editing for a number of years, first with VS6 then VS7 and for the last year MSP7. I've had audio synch problems from time to time and was always able to get around them. Out of frustration I normally do not use smart render which gets me around the problem. Reading the boards I would think the problem lies in my system and not the MSP7 (sp3)software.

I've finally boiled my current audio sych problem down to the following and was hoping someone could lend a hand to complete the analysis of where a conflict may be.

First off, the audio synch ONLY occurs during timeline previewing a smart rendered section of a piece of a project. Full rendering to a file works fine with smart render enabled. Instant preview plays correctly also. Full rendering without smart rendering is also fine. Never problems with final DVD output either.

To define the problem further, I took a 10 second MPEG-2 6000 VBR lower field first file with 48000 KHZ 16bit stero clip and performed my testing. No transitions, just the file in the timeline. I determined that when I did a smart render on the timeline two temp mpg files are created in the preview directory. One file is for audio and one for video only.

I then made copies of those preview files, loaded them into a timeline and found that the mpg for the video was four frames longer than the original video. Upon further inspection I found that in the mpg containing only the video there were some duplicated frames making the total clip longer than the original clip. The audio file was one frame shorter than the original mpg. By comparing the wav format it would seem that the last frame was dropped.

In the test the project file settings matched exactly the test mpg. I ran the same tests for an AVI and had no problems.

I've reloaded MSP7, tried different video drivers, scanned the registry for anything unusual in Ulead fields but to no avail. I have Cool-3D Production Studio and Photoimpact 8 as well, so you can see that I am a fan of Ulead products.

Any suggestions will be appreciated. I plan on upgrading my computer and will be getting MSP8 and would like to have this resolved beforehand.

Thanks,
Steve

HP Pavillion 7965
P4 1.7
768 mb mem
dedicated 120 gig 7200 rpm drive for editing, 30 gig free
nvivia, tnt model 64 video card
nvidia 4.4.03 driver (after falling back during problem testing)

[heinz-oz]

What are your project settings, capture format, source, how do you connect the source to your PC?

[Steve]

Heinz,

I use MSP7 Capture from a TRV-70 mini DV via firewire in AVI Type 1 format. After capturing a number of 20 minute clips I then use the Editor's batch convert utility for the AVI files with the following settings:
MPEG files
24 Bits, 720 x 480, 29.97 fps
Lower Field First
(DVD-NTSC), 4:3
Video data rate: Variable (Max. 6000 kbps)
Audio data rate: 224 kbps
MPEG audio layer 2, 48 KHz, Stereo

I then create a project with the following settings:
NTSC drop frame (29.97 fps)
Overlay track(s):3
MPEG files
24 Bits, 720 x 480, 29.97 fps
Lower Field First
(DVD-NTSC), 4:3
Video data rate: Variable (Max. 6000 kbps)
Audio data rate: 224 kbps
MPEG audio layer 2, 48 KHz, Stereo

I do not think it is the mpg file or project settings. Smart render does take place and I've tried as many other combinations of project settings that I could. For what it's worth, I would think that an old DLL or CODEC is getting in the way. It's only timeline smart render that has a problem. All other rendering works correctly. Please note the specific comment in my original post that the number of frames in the smart render files is not correct.

For the test, I took the 20 minute mpg, selected 10 seconds and created another mpg file so that it would be easier to work with.

Thanks,
Steve

[Devil]

I'm sorry, but what you are doing alarms me. MPEG is supposed to be a distribution write once format, not a working one. You should do all your editing/assembling etc. in DV AVI, with your project settings thus. Only when you have finished, should you convert from the timeline or a rendered file to MPEG ONCE. I think there is every chance that this will cure your problem. If it doesn't, try a 6000 kbit/s CBR encode, if it will fit.

BTW, during encoding to MPEG, you may consider encoding the audio to Dolby Digital 2/0 at 192 kbit/s, which is usual for NTSC DVDs.

[heinz-oz]

Why do you convert your AVI files to mpeg2 prior to editing in MSP? The mpeg2 format is not ideally suited for editing like DV AVI is. It's a distribution format and, when editing this, can result in all sorts of strange happenings, including working ok once in a while, it all depends.
I would do my editing in DV AVI format and only at the end create my DVD compliant mpeg2 files.

[heinz-oz]

Devil beat me to it but is of the same opinion, give it a try.

[Devil]

You Ozzes think you are ahead of us because of the accident of time zones, but you are really behind!

[heinz-oz]

In that case, devil, watch your back

[Steve]

Appreciate all of the interest, time zone issues and all.

Reason I use MPG format is twofold, one is that the project that I am working on involves twenty hours of footage from a live event. Numerous angles of the same scene. Between background audience noise, poor lighting, etc I can pick and choose the best pieces. With that much footage I do not have the hard drive space to edit in AVI. (Next time I'll have to most likely only load parts of the project at a time)

Second reason is that the program has an option for smart render in mpg. If it should not be used then the option should be turned off. What gets me is that smart render correctly works when creating an output file, but not while running in the timeline and I thought that if I could understand the difference in the process I could figure this out.

Note to Devil, I also tried 6000 CBR with the same effect. As a matter of fact I've decided that going foward 6000 VBR will most likely be my standard anyway. There is lot's of sword fighting action scenes and I thought that I would need 8000 but after reviewing I do see 6000 working fine.

Based on everyone's input, it seems that my computer upgrade is going to be needed sooner that I thought. Now I have a real reason for including a TB disk setup!

Enough talk of audio synch, now where is that copy of MSP8 that I am so eagerly waiting for.......

Steve

[Devil]

Because mpg is not an editing file system, smart render is not there for profound editing of a plurality of sources: it is there only for simple cuts and nothing else.

I seriously suggest you will need much more disk space to achieve what you want, because you are certainly losing quality, as well as synch, by going through two levels of encoding.

MPEG was designed primarily for mastering a video once and playing it back many times on many platforms. To minimize the cost of MPEG hardware decoders, MPEG was designed to be assymetrical with the encoding process requiring about 100 x the computing power of the decoding process.

...

If capturing video for editing, you can use baseline JPEG. Using JPEG requires that the system be able to transfer data and access the hard disk at bit rates of about 4 Mbps for SIF resolution. Once the editing is done, the result can be converted into MPEG for maximum compression

(Keith Jack)

You will be recommended to upgrade your hardware if you will go for MSP8. I can't discuss hardware requirements (NDA) but good, modern, computers give a distinct advantage with its finer features.