Who is? And what is your setup?

I fear of spamming again my harddrives with the doubled size of the double size of 48kHz...


I did 192 recordings for a extreme pitched down Vocals one time. It was just a test, on what is going on in the upper frequency spectrum. The result was something between artefacts and noise :D. So its more of a playground thingy.

Where 192k is very important for sound quality -> if you use distortion or amp simulation. There, the higher frequencies fold back on the audible ones (i can't give you a source for this, it was my Sound Design trainer who told me this) -> 192k makes sense.

  • And to the 24 bit -> 24 bit recording a must in my opinion. Then i let the mastering engineer dither to 16 bit for maximum quality! Dec 10 '13 at 11:56
  • 24bit - 16bit, no problem. this is well hearable, specially with airy things. What you mean is fourier-analysis, on which a sampling or synthesize-processor works. It means that you can fold every sound into a spectrum of sine-waves with a lot of their specific harmonics. Quick explained: The more harmonics you have, the more exact is the reproduced sound. The high, unhearable harmonics influence the lower and their lower harmonics and so on. But what's realy sampled, after 96khz - ?? What is technically possible without having only noise? Dec 11 '13 at 7:27
  • Distortion is indeed the strongest need case for big Nyqvist headroom... although, fortunately, an electric guitar signal doesn't actually contain very much above 10 kHz, since both the pickups and single-size speakers act as pretty good low pass filters. Distortion aliasing is thus more of an issue for master exciters etc.. Nov 26 '14 at 11:50

I very rarely go outside of 24 bit 96k recording. For the most part, nothing interesting happens up above 48kHz anyway. I'll go to 192 for things like contact mics, electrostatic recordings, or metallic things like bells. Other than that 96k works great for me.

  • Are you going to 192k if you want to pitch it later? Is there a difference between 96k and 192 for you if you work later with the sounds? Dec 9 '13 at 20:16
  • yep, its just for pitch flexibility.
    – Rene
    Dec 9 '13 at 22:13

I'll record 24/48 for ambiences, 24/96 for most fx that I'm going to manipulate later.

  • Is there a reason why you are not going to record everything in 96k? Have you tried 192? Dec 9 '13 at 20:17
  • I stick with 24/48 for general ambiences mainly because I know I'm going to do minimal processing later. Also these are usually longer recording (+5minutes) so file size is a consideration. I've tried 192 and for material I'm going to do extreme processing on I'll use it, but I find 96 a good general start for fx recording.
    – KellCole
    Dec 9 '13 at 22:07
  • For ambience, did you notice that your recorder needs a lot more electricity for working? As I have changed from 48 to 96kHz, the Sound Devices recorders needs much more power. If I'm out of nowhere and have to to be economical with battery power, it's a reason for me to save energy... Dec 10 '13 at 6:30
  • I use my Zoom for ambience recording so I haven't really noticed any drain. With the sound devices, power drain always seems to be an issue with ours.
    – KellCole
    Dec 10 '13 at 17:11

192 khz only if the sonic material, the microphone and preamp allow it. in my case that's not a lot of times. but it is interesting for 'playtime' moments. city atmospheres pitched and filtered can sound very beautiful and airy.

  • Nice idea! I'll try that out. What equipment are you using if you record 192k? Dec 9 '13 at 20:09
  • sonosax sx-r4 (up to about 100 kHz) and dpa 4060's (they go a long way up to 40 kHz) Dec 10 '13 at 19:21
  • hm... you bring me to another question. 'Up to 100kHz' - this information about frequency range came also out with high sampling rate-possibilities. But I have learnd and as far as I know, EVERY better microphone (not SM58 or so) can go up very high in frequency range. It was only limited at 44100 with old samplingrates. so no one cares. Would be interesting to find some information about this. I can imagine, that a new and very light diaphragm could make a higher f-range, but I wouldn't be surprised if a common KM184 could do absolutely the same - anyone knows?? Dec 11 '13 at 7:13
  • it's not the diaphragm that is the issue for higher frequency response, the electronics are the main difficulty. designing a 'HRF' circuit that allows 'clean' signalpaths takes great care, material, time. that's why really good microphones cost that much. same goes for the preamp/converter designs. a really good recorder (SD788/Sonosax/etc.) can go a very long way (up). A zoom h2n, not. Dec 12 '13 at 15:39

Usually (more than 90% of the time) I am recording at 24/96; when I think a sound I am going to record might sound interesting after heavy pitchshifting, I switch to 24/192. Setup is a MS rig of Sennheiser MKH 30/8040 into a Sound Devices 702.


24/96 is my standard for recording anything these days and 192kHz is a must for when I know the material will be significantly manipulated (and the recording equipment can actually handle it). Even when recording for video projects that are standard 48kHz I run a second recorder at 24/96 as a backup.

Considering how cheap storage has gotten and 24/96 using up about ~1GB per hour per track, 24/192 ~2GB per hour per track and a 1TB HD sells for about EUR 50, I don't think that storage space / price should be an excuse anymore.

It's also a matter of staying "future proof". I've already had cases where clients demanded 96kHz recordings (whether or not that was actually warranted is a different question) so it's good to just be ready for that.

  • You're right. The cost of storage shoud not be an excuse. The thing is, the file is not only saved once. It is probably saved a couple of times, it is on backups, backups of the backups, in project folders, raw-recorded and edited. I have problems to backup my raw-sounds archive, soon it's not going anymore to one disc, I need to find a new way. I dont' wanna split it, but I probably need to. And it needs time and time and time to transfer some TB. But I agree with you, to be compatible with the future in sound quality, but I'm sure that after 192k 32Bits will come. Or 48Bits?? Dec 11 '13 at 7:39
  • 2
    I don't think we will see drastic changes in recording bit-rates and -depths in the next few years. Personally I believe that 16bit 44.1 kHz is absolutely sufficient for audio playback. For recording purposes I am a big fan of 24bit and maybe even 32bit when recording tech catches up but I can't even fully take advantage of the dyn. range that 24bit gives me, except for peace of mind that even with levels at -40dB I'm still going to get great S/N ratio and don't have to worry about limiters much. 32+bit processing is of course a different topic but that doesn't concern us when recording. Dec 11 '13 at 13:44

I compared 41 khz and 192 khz. Admittedly, recording a guitar may not make a big difference but there was one. It was as if someone had increased the presence a little mixed with an exciter. This effect was barely audible but it was there. If I were to record drums, cybals, a piano or acoustic instrument, I'm sure that it would make a slight difference. Not sure about electronic music. Recording at 192 khz will probably tax my computer and hard drive. Perhaps I'll record some items as 192 and others at 96khz - yet to be tried.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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