96K vs. 48K vs. 44.1K

[daeron80]

Originally Posted by DrFord

Just pulling up old mixes and re-bouncing them made them 10x better.

Now that is a surprise. Do you have any before-and-afters you would be willing to post, where the only difference is the clock at mixdown? Because, theoretically, if you didn't change anything, it shouldn't make any difference, other than that you would be hearing the same bitstream more clearly. I could understand if you mean, "I could hear my mix more clearly, so I was able to clean it up and make it sound better." And certainly, old mixes will instantly sound better (if they're good mixes) when heard through better converters. But if no changes were made to the mix (and if nothing was being sent out to analog processors), the bounce-to-disk file should be the same, regardless of converter quality.

[daeron80]

Originally Posted by Sven62

why use 24 if it all ends up 16?

Try it both ways and you'll hear why. If you do any Photoshop type stuff, even only as a hobby, you already know how much better a heavily compressed jpg looks if you start with a clean, high-res file than if you start with a low res file. In digital audio, each bit doubles amplitude resolution, so 24-bit is modeling the waveform 256 times more precisely than 16-bit. That means that the plug-ins and mixer have that much more information about the signal. So they can make tremendously more accurate calculations, yielding a much truer 16-bit mix. The rounding errors will be far less egregious.

[DrFord]

Dave-
You are correct sir.. yet again. What I didn't specify was that I would contribute the improvement to is a better clock source from my Big Ben, which does contribute to in the box mixes. I don't often bounce out of my Rosetta 200, as I don't do mastering myself any more than I have to.

To elaborate since changing converters the most dramatic improvement has been in the recording, as my signals have gotten hugely better since the new converters.

Monitoring has also gotten a huge improvement as with a better outbound converter my mixes immediately sound better, in that I was able to hear more clearly, which then translates to mixing better.

[lcressy]

After reading this I did an experiment. I made a new session at 96k, and imported the session data in from a song I was working on, and WOW, what a difference! Thanks for the good info on this thread. I didn't realize that I could actually work at both rates, I thought it needed to be one or the other. I'm sure it is better to be able to start at 96k, but I too have the digimax and am limited to tracking at 48k, so this is a major improvement. Thanks!

[daeron80]

Originally Posted by DrFord

I would contribute the improvement to is a better clock source from my Big Ben, which does contribute to in the box mixes. I don't often bounce out of my Rosetta 200, as I don't do mastering myself any more than I have to.

Oh, I see, going out analog. Does it help to have both the Big Ben and the Rosetta? I would have thought the Rosetta had the same clock.

[DrFord]

The Rosetta does not have the C7 clock, but the ADx series does.

I will also admit that my monitoring may have improved from the Rosetta 200 while my ITB mix may have stayed the same. So the mix sounded better when it was bouncing, but the bounce may not have sounded better... weird but I think you know what I mean!

[Sven62]

Originally Posted by lcressy

After reading this I did an experiment. I made a new session at 96k, and imported the session data in from a song I was working on, and WOW, what a difference! Thanks for the good info on this thread. I didn't realize that I could actually work at both rates, I thought it needed to be one or the other. I'm sure it is better to be able to start at 96k, but I too have the digimax and am limited to tracking at 48k, so this is a major improvement. Thanks!

How can you import audio recorded at 48K into a 96K session without having speed issues? i.e. the imported track playing back really slowly.

Or maybe I'm confusing when I accidentally bring up a 44.1K session in 48K. I guess that's different.

[Sven62]

Originally Posted by lcressy

I too have the digimax and am limited to tracking at 48k

Hey Icressy.... do you know of a way to make the Digimax start up at the same clock speed each time? Mine usually starts up in 44.1 and that's what I almost always use, but sometimes it starts in 48K and I don't notice until it's too late.

[daeron80]

Originally Posted by Sven62

How can you import audio recorded at 48K into a 96K session without having speed issues?

In the import dialog, use the Convert button instead of the Add button. Then you'll see a check box for Apply SRC (Sample Rate Conversion). The correct source sample rate should appear automatically. Set the quality level to Best or Tweakhead. For some confounded reason, it defaults to Lowest quality. I think it used to remember which setting you used last but it doesn't in 7.3. :breathing deeply and counting to 10: OK, I'm better now.

[DrFord]

Another good way to understand why things would sound better in a better environment, will be to read the thread in the Mac HD / TDM forum talking about Stereo Dithered Mixer versus Stereo Mixer.

The topic is a bit different, but the outcome is the same. Where as 24 bit and 48 bit is a similar analogy to 44.1 and 96k.

Now I'm not saying these are in any way the same thing. And this isn't even taking into account your clock source (which produces a stable capture clock and math rate which all samples are captured by) and your AD / DA converters (which do the capturing... they all sound different and are different).

Because of plug ins I have and like to use, namely my LE Reverb from McDSP, I have to work at 44.1k still, but comparing my recordings with my Big Ben and my AD-16x to my old 002 recordings is night and day. Once I get a better Reverb that works at 96k I will be permanently switching to 96k.