24-bit 48kHz is the norm in film post-production here in LA, with new recordings and foley coming in at 96k or 192k. I'm recording at 96k most of the time and occasionally work with 192k samples when creating new sound design.

But what about 1-bit? Seems the recorders are coming down in price and becoming attainable. What's the best mic match for a 1-bit rig? How best to archive the recordings? And which workstations will let you work with such a file?

No one that I work with currently records at 1-bit, although I can imagine the incredible flexibility you could have with that resolution.

5 Answers 5


I used to have a Korg DSD recorder. Sounded amazing. As far as mics go, any mic that sounds great on a normal recorder will sound great on a 1 bit recorder - often times better...

I never found a good solution to working with these files though. They aren't particularly large files, but I never found a DAW that supported it, so I ended up converting the files to wavs or aiffs before manipulating anyways. (speaking of which, I have an awesome DSD recording that I'd love to listen to again but can't convert because I don't have my software anymore! Does anyone have Korg's Audiogate?!) Granted they sounded slightly different, but that may have had as much to do with the machine and it's preamps as it did the 1-bit recording.

If, say, Pro Tools or Logic would support 1-bit (don't know if Nuendo supports it), I would totally get one of those recorders again. You're right, there would be incredible flexibility with the resolution involved...

That said, try to stay away from the Korg MR-1 if you can. I went through 3 or 4 of them (all under warranty - hard drive kept dying on me).


I don't mean to water down the discussion but the idea of 1-bit sounds totally like someone needed a new and proprietary technology for licensing purposes. I am sure AES 5935 (the opposition's reasoning) is well conservative (and proven controversial) but I still cannot find a serious flaw with traditional PCM 192kHz/24bit.

Also, PCM does not require noise-shaping. PDM can't do without it. Adding noise? I thought we were trying to get rid of it in the first place.

At this time I'm much more concerned with what the best match for a 96k-192k rig is. For sheer Nyquist mischief that's a mic that can handle up to 96kHz frequency range. Mmm.. new and intricate sounds =)

  • Hah, indeed: Imagine future mics coming with a high-freq cut/rolloff switch to reduce ultrasound interference. Probably will happen in our lifetime... Commented Mar 18, 2010 at 17:50
  • 1
    @NoiseJockey - might not be required, since you can expect ADCs to feature anti-aliasing / low-pass filters already.
    – georgi
    Commented Mar 18, 2010 at 19:14
  • 1
    Adding noise is good when it reduces audible distortion. Especially in noise shaping, where the added noise is ultrasonic.
    – endolith
    Commented Mar 20, 2010 at 13:52

EDIT: Turns out, I totally misunderstood your question. It's 1 bit DSD you where talking about (thanks Kibibu for pointing that out). So my answer is totally void. I'll leave it here, just in case anyone else got confused.

Old and wrong answer:

Do you maybe mean 32-bit on field recorders? 1-bit is extremely low and would mean the resolution of the sound is very bad (or great if you're into chip music). The higher the bit depth, the higher the signal to noise ratio and the better the dynamic range. So if we record at 24 bit, we can record a little softer (safer to avoid clipping) and then bring it up in post without increasing too much noise.

Sample Rate is time based, and means how many times per second will the sound be sampled or recorded onto the recorder (eg: 96000 times/second or 96kHz). In the 'real' world, sound is continuous but digital processes information in chunks. To digitize the sound, it needs to be changed into discrete-time signal, these chunks of sounds give the illusion of a continuous signal, this is much like film running through a camera - this process is called quantization.

Bit depth is the resolution of your recording. In other words, how many bits are recorded per sample. If our sample rate is 48000/sec (48kHz) and our bit depth is 24-bit, then 1/48000th of a second is one sample and each one of these samples has got 24 bits of information.

Does this make sense? Or is it even more confusing? I know I didn't answer your question at all, but if you meant 32-bit, then I'm pretty sure we will soon have it on our recorders.

If you want to hear what 1-bit sounds like, check out this little 1-bit synth I made at a handmade music event.

  • 1
    1-bit is a different recording technology with a phenomenally high sampling rate. It's used on SACD, for example.
    – kibibu
    Commented Mar 16, 2010 at 10:06
  • Really? Wow, never heard of it. Good to know... Commented Mar 16, 2010 at 10:21
  • Awww, i thought it was a satirical question! Well. I suppose it's still a good question in it's educative property. But I was a pretty big fan of it upon first misunderstanding.
    – MtL
    Commented Mar 16, 2010 at 16:01
  • Yeah, I couldn't tell if it was a joke or not until I was reading through the first answer.
    – Mark C
    Commented Apr 16, 2010 at 22:06

Your 24-bit 48kHz uses the same oversampling sigma-delta modulation internally. Korg provides "AudioGate" software to convert 1-bit recordings to normal formats, which is what the 24-bit/96 kHz converter does internally anyway, so I don't see the point.


The only software i can think of that can use DSD/ support 1bit is Pyramix, which also has some very useful functions for working with video. Ive never used the DSD abilities of it so i cant comment on it really, but considering what else it can do im sure its worth a look.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.