Beatmaking and Electronic music at 96kHz vs. 48kHz vs. 44.1kHz

[epic1beatz]

Hey all, just curious. Does anyone make beats at 96kHz and is there a noticeable difference? I'm speaking strictly of electronic music, no recording using mics.
I use 48kHz and it seems to be a better sound.

Thanks

 

[KonKossKang]

analogue= no samplerate]infinite hypothetical samplerate.

Higher the samplerate, the smoother the waveforms are I think.

44.1 is standard though. [I prefer 48khz myself]
but basically the higher the samplerate, the less latency/better sound quality.

 

[krushing]

analogue= no samplerate]infinite hypothetical samplerate.

Higher the samplerate, the smoother the waveforms are I think.

44.1 is standard though. [I prefer 48khz myself]
but basically the higher the samplerate, the less latency/better sound quality.

I think it's confusing to even think about analogue as having "infinite sample rate", because while it's kind of like that, the fundamental difference between analog and digital is that the former is continuous, without discrete values for points on a timeline, whereas the latter is exactly that - discrete samples, taken at intervals (ie. the sample rate). Of course, that's just the beginning and to leave it at that will easily lead into the false impression about those steppy waveforms and even "better sound quality". I've linked to this article a gazillion times, but I still think it's one of the best ones to somewhat concisely explain how digital audio (and its conversion back to audible analog sound) works: Digital Problems, Practical Solutions |

 

[KonKossKang]

That is a confusing topic. More research probable.

 

[laurend]

96 KHz is theoreticaly better than 44 KHz. But 44 KHz is currently the standard sampling rate for music production. 48 KHz is for video or movie stuffs.
If you have enought CPU storage and bandwidth, 96 KHz and 32 bits is the best production format. 44 KHz is OK for music distribution.

https://www.maximalsound.com/mastering/Good_Practices_for_Mastering.pdf

 

[crimsonhawk47]

http://lavryengineering.com/pdfs/lavry-sampling-theory.pdf
24bit vs 16bit, the myth exploded!

In Short, sampling is not like frames in a film. 44.1khz samples what is in our hearing range, and 24 bit Quantization (with dither) recreates the waveform perfectly, other than white noise that is so ridiculously below the level of your signal that it's quieter than whatever noise is generated by your interface. Understand that sampling, and the Nyquist-Shannon Theorem, is not a "theory." It was not made to explain digital audio, digital audio was built from the theorem.

 

[KonKossKang]

...dither having a hand in recreating a waveform perfectly?
wow.

 

[krushing]

...dither having a hand in recreating a waveform perfectly?
wow.

This is actually a pretty nice analogy - what dithering does to images:

 

[CVLT CORP.]

...dither having a hand in recreating a waveform perfectly?
wow.

dither will mask distortion generated from bit depth reduction, but at the cost of a louder noise floor.

Anyway, the quality of sampling rate and bit depth of your productions should be strictly dependent on the consumer...cause if people will listen to your music in mp3 (and they will do unless they are audiophiles), then it's pointless to work at 32 bit and 96khz.

 

[waltzmastering]

Here's a decent article that Dan Lavry wrote a few years back about optimal sample rates.
http://www.lavryengineering.com/pdfs/lavry-white-paper-the_optimal_sample_rate_for_quality_audio.pdf

 

[epic1beatz]

Awesome discussion and thanks for all the replies. I'm currently rolling with 48kHz and it seems to be good sounding/works for my computer memory.

 

[Synapsis]

analogue= no samplerate]infinite hypothetical samplerate.

Higher the samplerate, the smoother the waveforms are I think.

44.1 is standard though. [I prefer 48khz myself]
but basically the higher the samplerate, the less latency/better sound quality.

higher sample rate, more information density... 44.1 is cd quality... you have to take into consideration the nyquist filter... record at high sample rates, especially with digital recording... then dither down to 44.1 16 bit

 

[DarkRed]

higher sample rate, more information density... 44.1 is cd quality... you have to take into consideration the nyquist filter... record at high sample rates, especially with digital recording... then dither down to 44.1 16 bit

Yes, in the software domain a very important choice is to record at as high sample rate and bit depth as possible, so that when the sound hits hardware again, it does so at optimum quality. Lowering the sample rate and bit depth from max should only be done when you do the final prints to specific playback formats at the end of mastering, it is better to record and never dither. All of the digital junk should not touch the signal.