Harmony Central Forums
Announcement Announcement Module
Collapse
No announcement yet.

64-bit sound. Are you using it?

Page Title Module
Move Remove Collapse







X
Conversation Detail Module
Collapse
  • Filter
  • Time
  • Show
Clear All
new posts

  • 64-bit sound. Are you using it?

    Hey guys,

    Are you using 64-bit sound?    What can you tell me about it?    What are the implications and/or promises of 64-bit sound?      Something to get excited about?


    ras

     

     

    <div class="signaturecontainer"> <br />
    <font color="blue"><b><font color="olive"><font color="sienna"><font color="purple">Every paint-stroke takes you farther and farther away from your initial concept. And you have to be thankful for that. </font> </font></font><font color="olive">Wayne Thiebaud</font></b></font><br />
    <br />
    <br />
    <b><font color="#808000"><font color="blue"><a href="http://www.facebook.com/#!/rasputin1963/info" target="_blank">Friend me on FACEBOOK!</a> </font></font></b><br />
    <font color="#808000"> <br />
    <br />
    </font></div>

  • #2

    The only you gain with 64 bit is more precision in the math(hence better sound but inaudible to the above average person, Superman maybe), with 32bit float the headroom is way more than any capable mixer needs as the sound floor is pushed so low below hearing to be sufficient for quality sound and the dynamic range is wide enough for a Mac truck to drive through(I'm joking of course, but it is wide).

    I remember the days of 16 bit 32hz, if it's your only tool you learn to work with it my friend.

    Comment


    • WRGKMC
      WRGKMC commented
      Editing a comment

      Are you talking about a operating system running at 64 bit or a sample rate?

      A 64 bit operating system allows you to run more memory. Its has zero to do with sound quality. Digital recording deals with sample rates. If you record at say 24/96 on a 32 bit operating system, its going to be exactly the same on a 64 bit systems. That sampling comes from the interface converters, not the computer operating system. All a 64 bit system does is move those ones and zeros around within the computer more efficiently. Those ones and zeros are not sound particles, they are mathematical measurements only. The converters deconstruct the analog waves on the way in, and reconstruct the signals back into an analog waveform on the way out.

      Where you may benefit is if you are running high end plugins. There may be less mathematical errors if you have more temporary memory to run those plugins. The binary data is able to back up into that memory while its being processed by the CPU and there's less chance of bits being lost or poorly processed because they are being rushed through the mathematical computations that occur. The symptoms that usually occur are dropouts, and those dropouts usually occur before you actually hear anything.

      Sometimes you can push a plugin to the verge of dropping out and hear graininess, shudders etc, but most of the data will be complete. It all gets into protocol and error correction and all that code a computer runs on, but just having a 64 bit system isn't going to give you better sound quality. A wave file recorded on a 16 bit system will sound the same on a 32 or 64 bit system. The 64 bit system should run smoother and faster, have less dropouts, and let you work it harder, but none of that relates to sound quality unless you are trying to run newer plugins on older systems that don't have the resources to run them. In that case they just fail to process the math on the binary samples.

      Besides, the DAW program has practically nothing to do with the tracking portion anyway. The data stream is simply routed straight to the drive with very little CPU processing occurring. Its only when you're mixing all your audio tools come into play. Having the extra memory to buffer all the data you may be processing whole mixing is a key to working the daw program more efficiently, but it doesn't make the data sound better because data has no audible sound.


  • #3

    rasputin1963 wrote:

    Are you using 64-bit sound?    What can you tell me about it?     


    What can YOU tell ME about it? What's 64-bit sound? Are you talking about 16-bit audio processed in a computer that can deal with 64-bit words? Or has someone convinced you that if you buy this expensive snake oil filled A/D converter you'll get 64-bit resolution in your samples?

    --
    "Today's production equipment is IT-based and cannot be operated without a passing knowledge of computing, although it seems that it can be operated without a passing knowledge of audio." - John Watkinson, Resolution Magazine, October 2006
    Drop by http://mikeriversaudio.wordpress.com now and then

    Comment


    • #4

      I'll assume we're talking about running a DAW with 64 bit internal DSP (as opposed to but not exclusive of a 64 bit computer running a 64 bit OS). 


      I suspect those with the most complex mixes should presumably benefit most. 


      I think you use Sonar?


       


      The versions of Sonar I've used (up through 8.5) from around 4 on have had the ability to use 64 bit DSP all the time, or only on rendering. For a while I was using the 64 bit render on  my very modest P4 and 32 bit everything else (best of both worlds, you get the 64 bit processing when you need it during the offline render process, but 'light' 32 bit DSP processing for monitoring while tracking and mixing, so as not to tax underpowered machines.


      FWIW, I've never noticed a difference -- but it's a very rare mix of mine that has more than 20 or 25  tracks  going at once.


       


      (I've recently bought a new box with 64 bit W7, but I still don't have a PCIex Firewire card for the new machine, so haven't really done any audio work on it. In fact, I've been doing all my listening from a SB X-Fi card going by lightpipe to the MOTU. It doesn't sound as good as the MOTU, not surprisingly, but the sound is considerably better than the noisy (hum) analog out of the X-Fi, surprisingly good considering the chip resamples everything on the fly to 48 kHz and then sends it over the lightpipe at 96. In addition to hum, the audio out of the SB (which was salvaged from a buddy's old desktop box)  has somewhat degraded HF response and apparent distortion (less clarity and delicacy in cymbal sounds,k reverb tails) but it's still better than the Connexant mobo audio interface (except for the hum... hmm... maybe that's not so better. Audible hum vs slightly even more degraded high end. Imagine what I might be hearing if my ears weren't 62 years old and my hearing didn't disappear over 10 or 11 kHz. The mind boggles.)



      music and social stuff | The Forgotify Files | A Year of Songs | mutant pop on facebook | roots acoustic on facebook


      The chorus seems a little weak... I think it needs more lasers.

      Comment


      • UstadKhanAli
        UstadKhanAli commented
        Editing a comment

        I have no idea whether this will help or not, but we noticed a difference when playing back sessions that had 4 tracks that were 24-bit vs. 32-bit float, playing back the same song.  

         

        My friend brought a drum loop, spacy synthesizer, and synthesized bass that were recorded 24-bit and 32-bit float, and then we recorded acoustic guitar at 24-bit and 32-bit float, so everything was the same except for the recorded acoustic guitar, which otherwise used the same signal chain, same guitar, same player, same everything except the performance, obviously.  The 32-bit float seemed a bit more spacious to our ears.  We heard it right away, but kept flipping back and forth between the two sessions.  

        We were happy about the nicer sound of the audio in a way, but alternately disappointed because we would need to chew up a lot more HD space.  

        I don't know if this helps your question at all since you asked about 64-bit, but there certainly was an audible difference between 24-bit and 32-bit float from our admittedly unscientific experiment.


    • #5

      The underlying issue with the misunderstandings in this thread is the use of the term "Sound" or "Audio."  There is no 64-bit audio or even 32-bit audio.  We have a conversion of analog audio signal to 24-bit digital (Or 20-bit whatever) and anything after that is a conversion of the conversion. 

       

      But no, I don't use 64-bit for anything.  Not necessary.  Besides, I'm waiting for 512-bit.  I figure instead of fu-king around every couple years with incremental changes just so people can make you feel your gear is obsolete and they can sell you new sh-t, why not wait for something REALLY BIG!

      "Everybody loves you when you're six foot in the ground."
      ~John Lennon

      Comment


      • WRGKMC
        WRGKMC commented
        Editing a comment

        If Rasputin is running Sonar, He may have been asking about a setting in the audio properties called 64 Bit Double Precision Engine. I've always assumes it was a setting used when running Sonar on a 64 bit Operating System but I guess I'm wrong. This feature can run on both 32 and 64bit operating systems 

        The Cakewalk site says this.    

          By Selecting this check box enables the 64-bit Double Precision Engine in SONAR throughout the entire signal path, including dithering and plug-ins.

        The 64-bit Double Precision Engine provides greater resolution, meaning more accurate audio reproduction and more headroom. You'll especially notice the benefits of the 64-bit Double Precision Engine when working in large projects containing many audio tracks and plugs-ins. Your chances of clipping will be significantly reduced.  

        The 64-bit signal path includes plug-ins and buses. SONAR sends and receives 64-bit data to and from all plug-ins that accept 64-bit data. If a plug-in requires 32-bit data, SONAR will automatically handle the conversation.   Best of all, you can take advantage of this audio engine on both 32-bit and 64-bit PCs giving you unmatched audio quality with spacious amounts of headroom and foot room through extended dynamic range. SONAR X1 also features native support for 64-bit floating point audio files, allowing you to import, stream, and render tracks and mixes at the highest quality available in the industry.          

         

        My guess is using this feature means there will be less rounding errors when converting from a floating point file (which is what Sonar natively works in) to a fixed point format. I never messed with the setting because I assumed it made the program 64 bit compatible, but I guess I was wrong. I'll have to try it and see if there's any benefit. Maybe it will help a highly processed mix, I don't know. It has the check box under audio properties that can also be selected as 64bit depth. Again, I haven't messed with it. When I mix down the file simply needs some mastering and its done. I'm not sure my mastering plugins will run with a 64bit float, but I'll likely do an A/B comparison and study it to see if there's any worth while benefits.

        I don't use a whole lot of plugins specifically to avoid the processing losses that occur. If I can get a better edge I'll use it, if not, it really doesn't matter much. The question would also be file space. If the mixdown size of a file is much larger, I'd have to weigh the benefits vs. the real estate. I'm guessing the size difference wouldn't be much if Any. I've mixed down in 32 bit float vs. 24 bit and saw no great difference is size so I'm guessing 64bit float would be the same deal. Honestly I can hear any significant difference between 24 bit and 32 bit float so I doubt I'll hear any difference using 64. If its simply a dynamic range setting then I already have sufficient headroom for anything I do in 24 bit, but I'm open to any improvements that simply require a box in an existing program to be selected.


      • UstadKhanAli
        UstadKhanAli commented
        Editing a comment

        Beck wrote:

        The underlying issue with the misunderstandings in this thread is the use of the term "Sound" or "Audio."  There is no 64-bit audio or even 32-bit audio.  We have a conversion of analog audio signal to 24-bit digital (Or 20-bit whatever) and anything after that is a conversion of the conversion. 

         

        But no, I don't use 64-bit for anything.  Not necessary.  Besides, I'm waiting for 512-bit.  I figure instead of fu-king around every couple years with incremental changes just so people can make you feel your gear is obsolete and they can sell you new sh-t, why not wait for something REALLY BIG!


        Like Holographic Sound (TM), which will sound far deeper, richer, more expansive, and more realistic than analog, and certainly more than digital.  I for one cannot wait.


      • learjeff
        learjeff commented
        Editing a comment

        Beck wrote:

        The underlying issue with the misunderstandings in this thread is the use of the term "Sound" or "Audio."  There is no 64-bit audio or even 32-bit audio.  We have a conversion of analog audio signal to 24-bit digital (Or 20-bit whatever) and anything after that is a conversion of the conversion. 



        32-bit and 64-bit audio formats do exist.  They're options for a .WAV file format, for example.

        There just aren't any 32-bit or 64-bit soundcards.


    • #6

      The big advantage of 64-bit systems is their greatly expanded access to RAM. For those of us who use orchestral libraries, that can make a world of difference.

      More here:

      64-bit Computing: Should Musicians Move To A 64-bit OS?

      (The article linked to above is about operating systems, but its principles apply to 64-bit DAWs as well.)

      Best,

      Geoff

      Enthusiasm powers the world.

      Comment


      • blue2blue
        blue2blue commented
        Editing a comment

        Geoff Grace wrote:

        The big advantage of 64-bit systems is their greatly expanded access to RAM. For those of us who use orchestral libraries, that can make a world of difference.


        More here:


        64-bit Computing: Should Musicians Move To A 64-bit OS?


        (The article linked to above is about operating systems, but its principles apply to 64-bit DAWs as well.)


        Best,


        Geoff




        And we start back around.


        I thought we sorted this all out back in 2004 or so  when Cakewalk introduced BOTH their 64 bit audio engine AND a version of the DAW designed to make use of 64 bit OS's running on 64 bit hardware. Two separate things.


        In Sonar, if you were/are runnig the 32 bit OS version you can still use the 64 bit audio engine (for 64 bit mathematic precision in DSP -- cature is still going to be 24 bit [or possibly lower] but DSP during playback and/or rendering can be done using the 64 bit DSP engine as the user desires.


         


        And, of course, all covered above. smile

        Attached Files

    • #7

      For those interested this AES video covers the subject of Audio Dither and perception:

       

      Great material my friends.

      Comment

      Working...
      X