Converting 24 bit to 16 bit for CD burning

Discussion in 'Audio Hardware' started by PJC68, Feb 22, 2019.

Thread Status:
Not open for further replies.
  1. PJC68

    PJC68 Forum Resident Thread Starter

    Location:
    Liverpool UK
    Hi folks
    I recently bought a sound devices mixpre 3M and a few good mic,s for recording, it records in 24 bit only and the resolutions are 44.1khz, 48khz and 96khz, this is a recording device in a box because i dont want to use DAW on a computer only for burning discs, heres the 3m below
    MixPre-3M
    I was going to buy a Tascam 24 track studio but i got convinced the 3m has better mic pre amps, plus a lot easier to use, portable and can be transported easily between locations.
    The trouble is i want to put all my recordings on to compact disc , CD,s are 16 bit 44.1khz so i will now have to learn how to convert it down to 16 bit to burn cd,s :realmad:, i emailed sound devices and asked them would they be doing 16 bit in a firmware update soon and they said they would look in to it, i said the mixpre 3 (non M) has 16/24 bit and musicians still use cd,s for demos/listening etc (i await in hope)
    I have a 26 piece orchestra to record in 4 weeks and will be recording in stereo (NOS) and was wondering what resolution should i record in (44.1 48 or 96) as i dont know if a 16 bit update will come out for the 3M
    I have downloaded r8brain free converting software but i have read it needs to be dithered down to 16 bit from 24 bit as well as converting, i dont want to lose sound quality through messing around as the pre amps in the 3m are top notch and sound devices are used in Hollywood films
    Sample rate converter tool software - r8brain | Voxengo
    All advice would be helpful as i bought this recorder to avoid using a computer for recording, any of the points i have made above could anyone point me in the right direction
    Please remember keep it very simple because i am mostly clueless and bought this box to avoid computer DAW
    Thanks in advance
     
  2. Strat-Mangler

    Strat-Mangler Personal Survival Daily Record-Breaker

    Location:
    Toronto
    Some applications do it automatically for you, such as Nero Burning ROM. No need to learn anything.
     
    tmtomh likes this.
  3. harby

    harby Forum Resident

    Location:
    Portland, OR, USA
    It sounds like you are trying to avoid using the right tool for the job.

    - You will need to edit the individual tracks out of the "master tape", removing all the extra waiting around after you press the record button, to give useful track numbers on a disc.
    - You should record in 24 bit, so you have the ability to at least adjust the volume and normalize while avoiding audio quality loss. This will avoid the temptation to record too "hot", potentially clipping the digital input when the orchestra plays louder than you expected.
    - Almost any software has dither when downsampling, the final step after all audio adjustments have been made. You can look at free Audacity, and see the dither options. "Best Quality" "Shaped" dither is available to be used when you export or downconvert.

    For free CD-burning software, you can save individual WAV tracks for songs, and use InfraRecorder to make your audio disc. A more advanced CD mastering engineer would make a single gapless WAV file, and use CUE sheets to write the track and index data and CD-Text.
     
    tmtomh likes this.
  4. harby

    harby Forum Resident

    Location:
    Portland, OR, USA
  5. PJC68

    PJC68 Forum Resident Thread Starter

    Location:
    Liverpool UK
    The 3m only records in 24 bit, if i am going to dither down to 16 bit what frequency would you recommend (44.1 48 or 96)
    I will look at audacity and see if it,s suitable as i did want to avoid the computer
    I will render the finished file in "WAV" and this will be gapless as it is a live concert in a church, i have already asked the conductor (who i know) to play the loudest passages in a sound check so i can set the level to avoid clipping
     
  6. PJC68

    PJC68 Forum Resident Thread Starter

    Location:
    Liverpool UK
    I will be using a mix pre 3m recorder with duracel L battery for power
    2 x Line audio cm3 cardioid condenser mics on a rode stereo bar in NOS config on a c stand
    I have 2x klotz mc5000 xlr cables
    A pair of AKG headphones for listening
    I will be putting the mics,s behind the conductor about 4 feet above him pointing towards the orchestra
     
  7. PJC68

    PJC68 Forum Resident Thread Starter

    Location:
    Liverpool UK
    I just need info on how to dither down to 16 bit painlessly
    Is dithering needed
     
  8. harby

    harby Forum Resident

    Location:
    Portland, OR, USA
    Record at 44.1/24, unless you think there will be an audience for a "hi-res" version of the recording. This will save you the downconverting effort. Most modern audio interfaces perform equally well, and sometimes with better specs, at 44.1/48, as they are filtering out the ultrasonic noise from light ballasts and other sources before it ever makes it into the recording when using a sample rate that matches our hearing.

    With a clean, quiet, dynamic recording such as orchestra, you should use dither to avoid any quantization distortion from 16 bit audio. It extends the capability of CD audio to its full potential.
     
  9. PJC68

    PJC68 Forum Resident Thread Starter

    Location:
    Liverpool UK
    Thanks for that, i was thinking 44.1 would be nearer to 16 bit 44,1 so will record 24/44.1
     
    StimpyWan likes this.
  10. harby

    harby Forum Resident

    Location:
    Portland, OR, USA
    You also might consider what you'd use for a "safety" recorder. I don't know if I'd trust 28 people's work to a single SD card -- even event photographers know to spread out their photos over many cards. The MixPre-3M has a 3.5mm stereo output, where you can go through the interface and route "MIX LR" to this jack. Even a standalone CD-Music recorder (and the correct CD-Music discs) would give me some assurance.
     
  11. PJC68

    PJC68 Forum Resident Thread Starter

    Location:
    Liverpool UK
    I have a tascam DR40 that does 16/24 bit and has 2 x xlr connectors
    But i wanted to use the sound devices
    Anyone use good dither by goodhertz as a standalone
     
  12. JohnO

    JohnO Senior Member

    Location:
    Washington, DC
    (I do not disagree with anything noted above, but this is my take and what I would do and do do.)

    If you are burning audio CDs as the final result, and if you are doing any editing of the recordings (as you probably should): I would record at 96/24. Your editing program will most likely convert the files to 96/32 if you make any edits, and you then save or export those as 96/32 WAV files. Then you can sort out your edited WAV files of 96/32, and feed them into Nero or Ashampoo Burning Studio (even the free version of this) for burning an audio CD. Those programs will automatically, painlessly and transparently convert/dither them to the 44.1/16 needed for the audio CDs. This keeps it "simple" and good. And you will still have the 96/32 edited files for any future use.

    I think dithering the edited 96/32 (which still is basically 96/24 inside the file) down to 44.1/16 would be better than dithering 48/24 or 44.1/24 down to 44.1/16. But, really, you would probably never hear such a difference.

    You can get some $500 or $1000 or $10,000 super dithering program, but I doubt that you'd hear any difference between those and Nero or Ashampoo. They are excellent now.

    Just as importantly, if you are recording direct to an SD card, be SURE to test that card out completely with a 96/24 recording of as long or longer than you will be recording, something where you would hear a skip, and listen to that completely to note if there are any skips. Fill the whole card with recordings and listen to all of the recordings, then delete those recordings and use that tested card. Some cards could have a bad memory spot which will cause a skip in the live recording. It has happened to me. Such a card goes in the trash instantly, after I copy off whatever recording it did get.
     
    Last edited: Feb 22, 2019
    enfield likes this.
  13. klockwerk

    klockwerk Forum Resident

    Location:
    Ohio USA
    If the target rate you want when you are finished is 16/44, then don't use 48khz. The sample rate dithering at 96 to 44 will sound good, and there is no dithering (sample RATE) from 44 to 44. But dithering from one rate to a another rate that is not a multiple of the target rate introduces some minor artifacting, though it is minimal when the rates are far apart. I'd just go with 16/44 (or 24/44) if that is indeed your final target.
     
  14. PJC68

    PJC68 Forum Resident Thread Starter

    Location:
    Liverpool UK
    Thanks for the replies guys, been busy last couple of days
    I have read that r8brain pro does flat dithering and down sample at the same time
    This would be ideal and a very easy solution for a newb like me
    All i want to do is record, edit, in the mixpre 3m and down sample/ dither and record to cd
    Is r8brain pro lite a little old now or still relevant, is there a better alternative standalone dither/crc software (nothing complicated)
     
  15. harby

    harby Forum Resident

    Location:
    Portland, OR, USA
    Sox is a free and open source resampler, and is quite advanced, but it is a command-line utility. One must be able to run a single command prompt line with options, which may be beyond the most computer-phobic. It also has advanced noise shaping options to be used with dithered downsampling, so the more informed can make their own decision:

    [​IMG]

    Here is a spectrogram showing why we want to dither (which is adding a little bit of our own controlled noise to replace the digital quantization noise). 4kHz sine wave at -90dBFS, 16-bit, 44100 with no dither:
    [​IMG]
    The tone at 4KHz has distortions created at frequency intervals above it. Without dither, the amount and nature of distortion depends on the input signal.

    Instead, we dither, and when we use true white noise, we end up with no distortions from the tone, just a uniform background noise (the deep blue that covers the whole spectrograph):

    [​IMG]

    With noise-shaped dither, we instead re-assign the same amount of dithering noise in a way that reflects the Fletcher-Munson threshold of hearing curve, moving the highest noise density to less-objectionable high frequencies (that many can't even hear):

    [​IMG]


    Only the most naive software wouldn't use some form of dither when downsampling. If the phrase DAW scares you, consider Audacity as a "sound file editor". You can open your audio file, see the waveform, listen to and edit out the extraneous passages, change the sample bit depth, and save your new file.

    [​IMG]
     
  16. TheVU

    TheVU Forum Resident

    Will you be synchronizing this Audio to a video?
     
  17. PJC68

    PJC68 Forum Resident Thread Starter

    Location:
    Liverpool UK
    Thanks for that
    Even though it is going over the top of my head i will read up on it (my one braincell overloading)
    I have downloaded Audacity but even that is a learning curve till i get in to it
    I have read somewhere flat dithering is better as noise shaping brings other problems
    I bought the 3m as i wanted to avoid computer music
     
  18. PJC68

    PJC68 Forum Resident Thread Starter

    Location:
    Liverpool UK
    No, just for recording music
    I think you need "timecode" for that
     
Thread Status:
Not open for further replies.

Share This Page

molar-endocrine