Technical questions on DV bitrate and sampling rate

[captain_tinker]

Folks,
I have been doing home movies for a few years now on DVD, and am slowly learning a lot more about it while trying to get them to look better and sound better. I used to use Sonic and Ulead when I first started, and had some decent results. I have now started to try capturing straight to DV avi using WinDV and then converting the file to mpg2 using TMPGEnc, and saving the files as m2v, and mp2. I have then been using ffmpeggui to convert the mp2 to ac3 and using TMPGEnc DVD Author to create the DVD's using the ac3 sound.

My question has to do mainly with the sound from the DV file. What format is it in, what bitrate is it at by default at capture time? I know that on the camera itself I have set the sound to be 16 bit. I was looking at the following entry in Wikipedia:

http://en.wikipedia.org/wiki/MiniDV

It says "DV allows either 2 digital audio channels (usually stereo) at 16 bit resolution and 48 kHz sampling rate." Ok, so far so good. I then look at ffmpeggui, and it asks about the bitrate. The default is 192, and the sampling rate is 48000 (48 kHz I assume).

So where does the bitrate of 192 kbps fit in regards to the original sound file from the DV? What bitrate is it "captured" at? Is setting it down to 192 compressing it too much, and would setting it higher during the conversion in TMPGEnc just make the file larger without giving it any appreciable increase in sound quality? I ask this because there is an audio bitrate option in TMPGEnc, where you can set it all the way up to 384 kbps. I am just wanting to know if the original file from the DV tape already has a bitrate, and what it is, so I can match it and keep it the same from beginning to end. Thanks!

-capT

[stars]

Hi the sound is usually recorded in 16b and 48kHz (as you say) and have a bitrate of 1532k (something).

When you convert the sound to .ac3 and use 192kbps you will get a very good result, but if you go higher
the result will be much better, but you cannot hear it. If you dont know where in the sound spectrum to listen.

If you use the 128kbps option you will clearly hear som "hissing" in the higher spectrum..

I usally use 192kbps its no need to go higher if you are only using 2 channels.

Its better to convert the wav stream (which is in the DV capture) directly to .ac3, not extract it as a mp2 file.

You can convert the wave file with Quenc V0.70 works great...

stars...

[gadgetguy]

What Stars said, except I usually encode audio to 384kbps, but that's just me. Try encoding some samples at different bitrates and see how they sound to you. Try to use a good sample featuring highs, lows and midranges for your experiments. QuEnc works great for both Audio and Video.

[captain_tinker]

Stars,
Thanks for the tip, I had forgotten I can just extract the wav from the avi, I will try that and see how it goes. I think I read you can do that with VirtualDub. I have that, so I will do that. Something that I think would be really interesting to do would be to make the sound file into 5.1 channel ac3. That would be neat. But seeing it is just a home video of my kids, I don't really need surround sound. I can get that from them all day as it is. ;-D

-capT

[redwudz]

I use pretty much the same method as you. After I transfer the DV with WinDV, I drop it into VirtualDub Mod, edit, filter if needed, extract the sound to WAV using full processing, then convert it to AC3 with ffmpeggui. I frameserve the edited video only to TMPGEnc encoder. Frameserving eliminates the in between edited file and saves space on your hard drive. Then when I author with TDA, I add the AC3 back in.

I like VirtualDub Mod as the audio handling just seems simpler than the regular VD. I use the Panasonic DV codec so VDM can read the file. Virtual Dub also has many filters available, http://neuron2.net/weblog/index.php

Extracting to WAV with VDM: https://www.videohelp.com/forum/viewtopic.php?t=280779

Frameserving with VD: https://www.videohelp.com/virtualdubframeserve.htm (Same for VDM)