File formats - Why we use what we use!

[Gerk]

One more thing .... I also just tried exporting the converted MP3 as wav, and it shrunk back down (some0, but not back to the original size:

orig wav: 0:14:86
exported wav: 0:14:91

So it came back to within 5ms.  Interesting.  Going to test on other applications as well, the above one was done with Audacity.

Exdporting with my DAW (Digital Performer), gives me:

exported wav: 0:14:94

Moral of the story, MP3's are evil timing wise without a doubt LOL.

Mark

[digitaldrummer]

Gerk,

good info.  thanks for digging even deeper!

for fun, I just converted a 16/44.1 WAV to APE (extra high) and then back to WAV.  I did a binary file compare (fc.exe on WinXP) and it came back with two bytes changed, and reported that the original file was longer...

00000004: F4 1E
00000005: 21 22

I'm guessing (from the early line #'s) its something in the header though so maybe harmless.

Mike

[Letizia]

thanks Mark and Mike!

sure, i'll try wav to wma and see what it does.

i was at http://lame.sourceforge.net/ (same site Mark posted) last week... part of starting to  research sound quality of various mp3 converters, as well as other formats. somehow landed on a forum dedicated to the same...

http://www.hydrogenaudio.org/forums/index.php?act=idx

specifically...

http://www.hydrogenaudio.org/forums/index.php?s=0959a5fa4d98c958380b46def9fc3cc7&showtopic=21904

[Gerk]

Gerk,

good info.  thanks for digging even deeper!

for fun, I just converted a 16/44.1 WAV to APE (extra high) and then back to WAV.  I did a binary file compare (fc.exe on WinXP) and it came back with two bytes changed, and reported that the original file was longer...

00000004: F4 1E
00000005: 21 22

I'm guessing (from the early line #'s) its something in the header though so maybe harmless.

Mike

I've tried this before, along with a test of the original and encoded/dcoded out of phase with each other.  I used the *nix binary compare (cmp) and they came out identical for me.

Mark

[narosis]

For the mac users who want to remain compliant with the windows users and one another for that matter, I suggest MAX which can be found http://sbooth.org/Max/ from the MAX developer page :: Max can generate audio in over 20 compressed and uncompressed formats including MP3, Ogg Vorbis, FLAC, AAC, Apple Lossless, Monkey's Audio, Speex, AIFF, and WAVE.


 

[Gerk]

Wow, this looks super cool  

Max is an odd name to choose for an audio product though ... at first I thought you might have been talking about this max (Max/MSP):  http://www.cycling74.com/products/maxmsp

Mark

[narosis]

 When I first found MAX, I too thought it a strange name to utilize, especially with the popularity of cycling74 / Max/MSP! I'm a Pluggo owner / user and very familiar with cycling74, speaking of I also use PD = Pure Data. I hope that MAX performs as we expect & need it to.

[Gerk]

It looks pretty interesting so far.  Not impressed with the interface though.  It would be great if we could have designated types of encoding/decoding, or maybe a prompt to choose which type.  It seems like you have to set the preferences manually each time you want to do a different type of encoding.  Looks great for batch jobs and cd ripping though

I've had great luck using the latest CocoaMonkey so far, have never had compatability issues with anyone to date.

Mark

[narosis]

I didn't even know there was a CocoaMonkey going to google it in a sec, yet before i do let me state, "I agree with you, it would be better if you didn't have to manually adjust preference settings & could do it via the interface."

[dogbizkits]

In this topic, the main problem we're faced with (for the collaborating musician / engineer / producer) is download speed. This is what we all know makes compression utilities an absolute must for those with low bandwidth. However, it's purely a matter of personal preference as to which format one opts for when collaborating. Personally, my argument for using the native files (WAV / AIFF) stands on this basic foundation: If you're able to make the native file (WAV / AIFF) available, then do so because it's ultimately the best solution for the mix engineer. If people can only work with compressed files due to bandwidth limitations, then that's the way to go for them. I would not criticise in any way. It's just a matter of finding a common agreement / solution between one another for providing master tracks that will allow the mix to be given its best chance possible for maximum sound quality. I generally work with the native files only and steer clear of the small compressed ones because I don't belieive in "tampering" (converting) just for the sake of quicker download times. Why convert something "down" - only to convert it back "up" again ? I think that's a step best left out to avoid chances of conversion errors. Having said that, I can confirm that ZIP and RAR are exceptionally good packaging tools. I've often received a whole project as several ZIP / RAR packages (i.e., for Drums, Guitars, Keyboards etc, etc) and they are very clean when expanded. For those of you who use Cubase, Steinberg do not have any recorded issues with ZIP files in Cubase projects. Unfortunately, I've had lots of problems with APE files [bad conversions] and they've generally been more trouble than they're worth. So, we see particular groups of guys and girls working more and more with each other when they establish a common way of working. This applies equally to native file users - or compression fans. That's just a little bit of insight into the way I see things, the way I work - and why I like to work that way 

Rab 

[BassPlayer]

As you probably know I'm a big Wavpack proponent. I've send several people self extracting wavpack files and not only were there no problems but it elimitated a bunch of mouse clicks MAX now has wavpack support. Why people still use ape is beyond me.

[Gerk]

Cause I don't use MAX   LOL

The music community is traditionally really slow at adopting new stuff ... I know a lot of folk that still won't use any kind of compression (aside from zipping wavs).

Mark

[Krisus]

http://www.sharebig.com/  1.5 gig transfer, so signing in...  See what you think.

Send the whole damn thing!  Why not?  Wavepack is a good compression, plus they give you the option to keep all of the bits that have been cut in a separate file.  You can put them back together on the other end.  I used this program when it was first compiled, great piece of code.

I have never really liked MP3 anyway, but it is such a standard you just have to get used to it for now until Steinberg decides to pull his patent trick, which will happen I believe.  I've gotten way better compression ratio results with wma, no doubt.  Getting below 128 on an MP3 is really risky, but I have done the low 50's with a wma and it sounded good enough to pass.  An MP3 at the bitrate is atrocious!

To master I would prefer two things.  1.) A fully mixed wave file and 2.) A separate vocal track.  Reason?  Musicians record like musicians.  9 out of 10 times they bury the vocals.  I can dig them out, but it's a bitch!  So if I am mastering a track recorded by a musician rather than a vocalist, I always ask for that separate track.  I have developed a method to master an MP3 also.  If the mix is true to the harmonics   and is a fair representation of what they want to be, I can polish it up on the backside.

I know, I talk too much      

[NickT]

Krisus -

The Sharebig site looks like the ads are on the download side of things. Good find, I think.

We have been doing pretty well with the files in the thread. But you can do things how you see fit.

Thanks,

Nick

[Krisus]

That's cool.  I mention this from a mastering point of view.  That is how I look at recordings.  Of course I think we all would want our finished product as dynamic as possible.  I haven't used that file transfer thing yet, I just recently found it.  Have no idea if it's good or not.