Is there any reason for high resolution audio files?

September 6, 2021
 by Paul McGowan

14 comments on “Is there any reason for high resolution audio files?”

  1. Paul points out that high frequency content is always lost on down sampling. That lost information is gone for ever, and cannot be recovered later by up sampling the digital data.

    There are, of course, legitimate reasons for up sampling audio data, but recovering high frequency content is not one of them.

    Up sampling and down sampling both involve the use of a digital low pass filter. If we were up sampling CD data, we’d need a LPF at about 20 kHz to calculated the interpolated data points. It’s hard to get away from that filter.

    1. Good afternoon Mark!
      Perhaps both you and Paul can answer this question for me.
      First, let me start by telling you, that I am a blind man.
      Paul already knows this about me.
      Someone told me, that Apple Music, only starts out at 16bits.
      And the sample rate, is 44.1KHZ.
      I don’t know how true that is, because I’ve never been able to download any music from there.
      But how I get music from Apple Music, is, I plug my iPad in to my computer, via its headphone jack.
      I use Wave Pad to make and edit my recordings.
      Then I use Diamond Cut 10 to add other sound affects to it like equalization and fat bass.
      I also use Audacity to add just a touch more of bass and treble.
      And I use a little bit of reverberation to make it sound like I’m setting in a large room listening to it.
      But the wave files go in at 32bits, 192KHZ.
      But when I get threw with them in Audacity, they come out in 64bit floating wave files, with a sample rate of 396KHZ.
      Even if I didn’t do the sonic treatment to these files, to me, they sound better coming out, then they did going in.
      But the person I told this to, isn’t so sure.
      He tells me, “there’s no way that you can do that.”
      But yet, I’m both seeing and hearing this stuff happen every time I do it.
      What am I picking up on, that other people aren’t?

      1. John, that’s a very interesting question. If I understand correctly, you’re telling us that your processing chain (iPad, PC, WavePad, Diamond Cut, Audacity) improves on the sound of the original file, even if the EQ and reverb is turned off. We know that up sampling can’t uncover lost information, but there’s some change happening that sounds good to you. It’s hard to guess what that change might be, but it reminds me that as a teenager I had a Revox A77 valve/tube tape recorder, where the replayed music always sounded better to me than the original. Objectively, the tape recorder (even half-track at 15 ips) couldn’t improve on my poor quality source, but it certainly sounded great.

        1. Good morning Mark!
          I don’t know how to explain that either.
          All I know, is that this is what I both see and hear every time I do it.
          But I’m also thinking about cutting a record album on myself.
          I’m almost 50 years old.
          Granted that I got paid to do this for other people that either had or started their singing career.
          But now, I want to do a few, while I still have a chance to do it before I get too old to care about it anymore.
          I heard Paul talking about upgrading to a new Lanova recording work station.
          And this is where both he and Gus will be able to produce quod DSD files.
          I’m looking at a Merging Technologies system that can do the same thing.
          But the only thing that’s tripping me up about it, is the software that has to be used with it.
          But sense I have to start out on a PC, the software has to be JAWS user friendly.
          And when I make the jump from a PC to a Mack, the software also has to be Voiceover user friendly as well.
          If you can remember what I told you all yesterday, I’m a blind man.
          So I have to use text to speech screen reading software in order to be able to do anything with any kind of computers.
          But what I wish to do, is lay down the tracks on tape first.
          Then mix them all down to just one big stereo track.
          And then, record directly to DSD.
          But as of right now, I still don’t have a clew about how to go about doing all of it.
          I found an open reel tape recorder that can do four tracks, I haven’t picked it up just yet.
          But after I do that, figuring out the next step is the hardest part of all of this.

          1. John,
            Just start using GarageBand on a mac. You can get 24/44 which I can tell you is good enough. Once you get experience with GarageBand, you can decide for yourself how much more you want to go.
            The benefit of recording to PCM is software availability and ease of use, especially for editing.

            1. Good morning CTA!
              For what it’s worth, I’m a veteran when it comes to making recordings on computers.
              And as for garage Band, I tried downloading that to this Windows system.
              The problem I had with it, I couldn’t install it.
              It gets me to the install wizard, but JAWS hangs it up on me right there.
              But as for putting together an album on myself, I’d much rather do that in DSD.
              It will get me and my would be fans a whole lot closer to what I’m hearing in my own studio.
              A lot of records made today, won’t do this right here.
              I want you to be engulfed in my music.
              I want it to be rapped around you.
              But you have to get it right the first time.
              PCM falls some what short of doing that.

              1. GarageBand is, I think, a purely mac environment software.
                I also think that you over think (pardon the repetition) the sound.
                You will not hear a difference between 24/44 and DSD in a controlled situation. It is in your mind.
                Focus on ease of use, especially given that you state you have experience recording.

                Your fans will hear in your studio what you hear through your monitors. This is what matters to you, what comes out of the monitors is the “truth” that will be heard. It will never be what you hear while playing. Because if you move around the room, the sound will change, while the sound from the monitors will stay stable.

                1. Hi again CTA!
                  It’s not just for Mack’s only, it’s for Windows too.
                  Just Google GarageBand for Windows, and you’ll see that, you can download it for a Windows operating system too.

        2. Mark,
          I wonder which model of ReVox A77 you had because I cannot recall it having any tubes.
          I have a B77 model that definitely has no tubes but it is not as old as the A77.
          Are you sure of the “valves”?
          Also, I never managed to get it to sound better than the vinyl from where I was recording music. I still have some Ampex tapes from a live concert. But I can’t compare those to the real thing.

          1. You’re absolutely right. I checked the schematic, and the A77 is 100% transistors. Not sure how I could have remembered so poorly. Obviously getting it mixed up with some other kit.

  2. I wonder if Paul can comment on the use of 32-bit floating point for audio. The IEEE floating point format consists of a sign bit, 23 amplitude bits, and eight exponent bits. The resolution between two similar size signals is, on average, about the same as 23 bit linear. On the face of it, there is not much precision to be gained using floating point. However, floating point does maintain the high precision for smaller signals, which is not the case for linear PCM. I suspect that there might be a benefit in the sense that quantization noise is masked by the large signal size and scales with smaller signals.

    The biggest benefit of 32-bit floating point is inside a digital audio workstation where it removes all worries about dynamic range, clipping and noise floor. That benefit is not extended to delivery of audio to listeners.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram