Question(s) about 24 bit and 16 bit

You could send them a PM, I'm sure they would be more than happy to help you! I'll download Audacity later and see if I can help you.

In edit > preferences > quality > default sample format > leave at 32 bit float.

In edit > preferences > quality > default sample rate > leave at 44.1 kHz

In edit > preferences > file formats > uncompressed export format > leave at 16 bit

with the wav file in the window, on the file pulldown menu where you see the downward facing triangle > set sample format > leave at 32 bit float.

with the wav file in the window, on the file pulldown menu where you see the downward facing triangle > set rate > 44.1

In the bottom left hand corner, where it says "project rate", set at 44.1

Then go to file > export as wav

I just converted one single 24/48 wav file to 16/44.1, and there was no silence at the beginning or end of the file, and it sounds great. I used the high quality sample rate settings and triangle dither for both real time and high quality. I think 32 bit float precision is just the internal processing setting of Audacity. The higher the depth, the more precise it is. Hope this helps.

[2000_man]

What version of Audacity are you using Tubular? I had downloaded the 1.3.3-beta version because it could handle flac files. I'm going to uninstall it and go with the last workable version 1.2.6 and see if that works.

[2000_man]

I believe I have found the culprit. There is a bug in Audacity as explained in this thread . Read under "Extra notes about burning to CDs" specifically #2 where it states:

2- Make sure your Project Rate (Hz), as shown in the Project Rate button bottom left of the Audacity screen, matches the rate showing in the Track Panel (where the mute/solo buttons are). If you don't do this, and you're exporting WAV or AIFF files, silences may be added at the end of tracks and the labels may not export in the correct position. This is due to a bug when resampling is done between the Project and Track Rate upon export. If necessary you can select your track by clicking in the Track Panel then Project > Quick Mix, which will resample the track to the Project Rate. Then simply delete the excess silence generated at the end of the track and the export will have no added silence. In 1.3.3, click Tracks > Resample, and in the box that pops up, enter the rate showing in the Project Rate button.

My problem is that if I do as you were saying in "with the wav file in the window, on the file pulldown menu where you see the downward facing triangle > set rate > 44.1"

then the track sounds slow as molasses. I have to only change the project rate to 44100 and I keep the set rate -> 96000 in the downward facing triangle. Since they both need to be the same because of the bug that will result in silence, it won't work. So, I have to delete the excess silence in which they make it sound easy, but I have to magnify the track to the nth degree to see the last bit of sound to trim the silence off.

Slow as molasses is right! I just tried a 24/96 wav and got the same results as you. I guess I didn't notice the slowdown of going from 48 > 44.1 when I did it earlier, but I did notice it when I did 48 > 44.1 again. Also 48 > 44.1 slows it down less than going 96 >44.1. There must be some setting that we are missing. Either that or Audacity is a piece of shit for resampling!! You shouldn't have to edit out the silence at the end of a track for every wav!! Not to worry, there is a much better free sample rate converter here:

http://www.voxengo.com/downloads/?highlight=r8brain

Get the Win GUI version. Use the very high quality setting for best results. It takes quite a while to convert, but there is a batch mode, so you can input all your files and start it, then come back in like an hour and it will be done. You can still use Audacity to dither and convert 24/96 > 16/96. Just keep the sample rate the same. I'm not sure what the best dither setting is: triangle, rectangle, or shaped. r8brain will convert bit depths for you but I don't think it dithers, so it would be good to keep Audacity for dithering. I took a 24/96 wav, then dithered and converted to 16/96 with Audacity, then resampled to 16/44.1 with r8brain and the results are great. No slowdown, and no silence was added at the beginning or end of the track.

Correction, r8brain also dithers:
"Fixed bug in implementation of dithering. Now dithering also incorporates a slight noise-shaping."

http://www.voxengo.com/product/r8brain/

So if all you need to do is to convert 24/48 and 24/96 files to CD-quality 16/44.1, all you need is r8brain!

[2000_man]

r8brain looks like the holy grail to a novice like myself. Thanks! I just wish I had found it about a week ago

I found a really good forum over here when I was searching on what to do in Audacity. It covers just about every scenario possible. Sounds like triangle dithering was the way to go. I thing shaped is the default. When editing my wav files, I dithered with shape and then with triangle when I took out the excess silence. According to the tutorial, I should've gone triangle and then none. Anyway, I made a copy of the 16 bit converted from the 24 bit that I did in Audacity and I now have no pops, no silence and everything changes seamlessly. So, now I know what to do (and not to do) in Audacity and r8brain. I'm going to make a copy using r8brain and compare the two.

I learned some new things playing around with Audacity though that I didn't know before. Searching and reading the forums over there, r8brain seems to be pretty well accepted among the masses if all you are doing is a simple conversion. Audacity is more robust and you can play around with the dithering, but for my needs r8brain should do the trick from now on. And you don't have to edit out the silence at the end.