Results 1 to 5 of 5

Thread: Test result Q: going native...

  1. #1
    Join Date
    Sep 2001
    Posts
    2,328

    Test result Q: going native...

    Now that the results are in, it seems we can say that the dogma of PT mix bus inferiority has at least been shaken. Granted that very experienced folks like Mixerman at r.a.p. got it right (of course it could have been a good guess, but I'll give his ears the benefit of the doubt)...but still, there was enough educated diversity in the guessing that it seems reasonable to say the differences were pretty subtle.

    Now, my question: to what extent can the results of this exercise be applied to native DAWs like Digital Performer?

    I realize that this question can't be answered reliably without repeating Lynn's experiment under identical conditions (except with DP substituted for PT). But I'm inviting shameless and irresponsible speculation here...MOTU advocates claim that DP's mix bus is actually superior to PT's because of the 32-bit floating point math it employs. What's the house consensus on this?

  2. #2
    Join Date
    Aug 2001
    Location
    Fort Wayne, IN, USA
    Posts
    27,246
    I have some friends who are internationally known and well respected engineers that claim that DP and PARIS both sound superior to PT in their opinions. I have not tried either of those options so I will not speculate. That's all I know.

    I still have the source files (.wav) and could note the settings if anyone was curious to find out with the same material I used on the PT vs. Neve CD.

    It's always interesting to put these commonly held beliefs (myths sometimes) to a more objective and equal test. I learn a lot each time.


    Originally posted by Brian Middleton:
    <STRONG>Now, my question: to what extent can the results of this exercise be applied to native DAWs like Digital Performer?

    I realize that this question can't be answered reliably without repeating Lynn's experiment under identical conditions (except with DP substituted for PT). But I'm inviting shameless and irresponsible speculation here...MOTU advocates claim that DP's mix bus is actually superior to PT's because of the 32-bit floating point math it employs. What's the house consensus on this?</STRONG>
    Lynn Fuston
    3D Audio

    Making beautiful music SEEM easy since 1979.

  3. #3
    Join Date
    Sep 2001
    Location
    SWEDEN
    Posts
    178
    Originally posted by 3D Audio Inc.:
    <STRONG>I still have the source files (.wav) and could note the settings if anyone was curious to find out with the same material I used on the PT vs. Neve CD.</STRONG>
    I would love to do the test with Logic Audio and Nuendo.

    Any chance of getting the files via Internet?


    Cheers
    Frederick
    Ninja Production
    www.ninjaproduction.com

  4. #4
    Join Date
    Aug 2001
    Location
    Fort Wayne, IN, USA
    Posts
    27,246
    Not likely. 48 files at 24-bit 48K. It would take a while to download.

    I'm not committed to this idea yet, but I'm putting some feelers out for another piece (more progressive) to see if I could have two high quality sources.

    I'm still considering it. There's definitely a good idea in there. Same files, different platforms, no changes just duplicating the mix that is already there. It would have to have no reverb or any plugins to be valid. We could do PT, PTHD, DP, Paris, Nuendo, and Logic, right? That would be enough to confuse just about anyone.

    I would love to do it all locally to have control over the mixes and also to use the same clock. I think there should be a version that is clocked internally and the same exact mix clocked externally. That would give a more level playing field.

    Hmmm. Could be revealing. I like the sound of this experiment.

    Give me some more ideas, but let's not let it get out of hand. If it gets too complicated with too many variables, then it will never happen.

    Originally posted by Frederick Norén:
    <STRONG>

    I would love to do the test with Logic Audio and Nuendo.

    Any chance of getting the files via Internet?


    Cheers
    Frederick
    Ninja Production
    www.ninjaproduction.com</STRONG>
    Lynn Fuston
    3D Audio

    Making beautiful music SEEM easy since 1979.

  5. #5
    Join Date
    Apr 2002
    Location
    Dallas, Texas
    Posts
    4,751
    First of all, I would like to thank those who have so thoughtfully contributed very high quality input to this forum. It's nice to see audio professionals share. I am not a pro at this (I run a not-for-profit studio as a hobby targeted as less commercially viable music), but I aspire to create a quality product.

    I do have some input on the ProTools vs. native discussion. Though I am not a trained engineer, I know just enough about DSP and digital audio -- espeically processing it -- that I think I can lend some insight into the claim that certain native products sound better than ProTools.

    It is my opinion that a product which uses completly 32-bit floating point processing architecture throughout the software will indeed sound better than ProTools when doing processing, but not necessarily mixing. I have done listening tests between the 32-bit float mode and 24-bit integer mode of my software (Nuendo) and can confirm, again, in my opinion, this is true.

    The reason it will not make a difference for mixing (and when I say mix, I mean it in the "Add the signals together" sense, not in the "production mix with effect to make a finished product" sense) is because to mix two signals together one simply adds the vector of samples together. Provided there are no overs, this is very simple operation. And, because the sampling is done in a 16 or 24-bit integer format anyway, nothing is lost.

    Now processing is a whole new story. Every time a plug-in is used in either system, a buffer of data is passed in a 24-bit or 32-bit vector between the plug-in and mix bus of the software. Something as simple as changing the gain of the signal requires multiplication to the vector which will most often result in a large number of non-integer quantites. Though processing within the plug-ins themselves tends to be of the highest quality, the issue is that the plug-in must then quantize (that is, turn all those non-integers back into integers) the signal to the resolution of the mix bus.

    In the case of Pro Tools, this is 24-bits. In the case of Nuendo, this is 32-bit float. What is SO important about the use of floating point quantites is it allows the retention of the resolution necessary due to plug-in processing through the mix bus. When the signal is soft, any bits not required to represent the amplitude (loudness) of the signal can be used to increase the effective resolution (bit-depth) of the recording!

    That means that 32-bit floating software suffers from less quantization distortion than 24-bit integer, and the result will be a better sounding final dither to 16-bit integer formats such as CD. The more plug-ins you use, the more fader moves (or adjustments in signal gain which do not result in whole number coefficients), the more a difference 32-bit software will have.

    I'd love to hear any feedback, confirmation or otherwise, from those out there really in the know.

    --Jim
    Jim Dugger
    Poorhouse Productions

    At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?

Similar Threads

  1. Native Instruments Hitler Humor
    By Piyono in forum The Old Yellow Board
    Replies: 44
    Last Post: 12-27-2009, 09:57 PM
  2. Another test CD
    By lowdbrent in forum The Old Yellow Board
    Replies: 6
    Last Post: 03-08-2005, 06:14 AM
  3. a test for CD burner integrity
    By grekim in forum The Old Yellow Board
    Replies: 1
    Last Post: 10-08-2004, 05:03 PM
  4. Difference Signals
    By 3daudioinc in forum Awesome DAW-SUM Comparison
    Replies: 222
    Last Post: 12-24-2003, 04:01 AM
  5. Avalon blind test
    By stefan s in forum Microphones and Preamps
    Replies: 15
    Last Post: 08-04-2003, 08:49 PM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •