cancel
Showing results for 
Search instead for 
Did you mean: 

Preferred audio rate-- 44100 or 48000?

mcc
Level 4
I have a choice, when synthesizing the audio for my app, of doing so at 44100 or 48000 hz.

Is one "better" or more natural for the Quest? That is, does the Quest have some "true" resolution that it is resampling to, such that in fact if I submit (for example) a 44.1khz sample stream to OpenSL, it will be internally resampled to (hypothetically) 48khz before hitting the DACs? Or vice versa?

Will Oculus Audio work better with one sample rate vs the other?

The Oculus Audio reference guide says "44100 and 48000 are recommended for best quality". Is this a promise that the Oculus devices are equally capable of playing either a 44.1khz or 48khz sample stream at native resolution?

(If it helps, I am developing against the Native/C++ API.)
6 REPLIES 6

EssenBlindt
Level 2
I would be very interested to know this as well !

rtathegawd84
Level 4
The difference between 44.1khz and 48khz sampling is not audible to human ears in quality. 44khz was just the standard for CD players so is still used/available today, 48khz was more used on other hardware like P.C audio and other devices. there's a lot of debate in the pro audio community about sample rates.. (I do music production myself) as above 96khz sampling there's no perceived difference in quality compared to 192khz (or more). My Audient id22 audio interface sounds better than most 192khz interfaces. In essence it comes down to the electrical components used, caps, converters, channel strips,.speakers etc etc. So to answer your question either 44-48 will be absolutely fine!!! You won't notice the difference!!

Hi RTA-- I agree that 44 and 48 sound about the same. However the problem is that if you generate audio at 44.1khz (or have 44.1khz assets), and then the device internally resamples to 48khz before playing, it will have to perform a resampling algorithm, and in my experience the human ear usually *can* hear the resampling artifacts from converting between 44.1khz and 48khz (at the limit of our perception, but enough to be annoying). This is why I am curious whether the 44.1khz output on the Quest is "natural" (IE, supported by the DAC or is it being resampled)

PITTCANNA
Level 12

Depends on how much post processing your going to do.

 

The best recommendation is to capture and do everything up and to the point of final render at the highest rate you have available, then when your in the studio you have more headroom for tweaking and adjusting levels.  After your satisfied, down sample as much as you feel comfortable with you actually listening to the end result.  But in reality 44.1 v 48 is space on the quest 2.  Meaning those who have a 64, who want to get every ounce of memory will thank you if you can compress it as much as you can.  so a 44.1 is preferable in the long run.

 

Pros and Cons of Using a 44.1 kHz Sample Rate
The main benefit of sticking with a 44.1 kHz sample rate is that the lower sample rate puts less demand on your computer. You do not need to utilize as much RAM and CPU power when using a lower sample rate.

The file sizes also tend to be smaller, which may be a factor when sharing audio files with collaborators over the Internet or saving space on your hard drive.

As humans cannot hear the difference between 44.1 kHz and 48 kHz audio, there is also less need to use the higher sample rate.

You should also consider the format that you plan to use when delivering your final mix. The standard format for audio CDs is 16-bit/44.1 kHz. If you record at a higher sample rate, the sample rate needs to be converted to 44.1 kHz.

Older conversion software would produce distortion or a loss of quality when converting integers. These same issues do not apply when converting 88.2 kHz to 44.1 kHz or 96 kHz to 48 kHz.

However, when converting from 48 kHz to 44.1 kHz, your mix may not sound as good. Basically, if you are planning to burn your music to CD, 44.1 kHz remains your best option.

Pros and Cons of Using a 48 kHz Sample Rate
While you cannot hear the difference between 44.1 kHz and 48 kHz sample rates, there are a few minor differences when it comes to editing the master.

Using a 48 kHz sample rate offers slightly more headroom for tweaking your mix. If you decide to go back and edit the master, the 48 kHz sample rate offers more flexibility, especially when working with high frequencies such as the sounds produced by cymbals and high hats.

A higher sample rate also reduces the risk of aliasing. Aliasing occurs when the different frequencies become less distinguishable due to artifacts not getting filtered out. You also get lower latency rates with higher sample rates.

Basically, a higher sample rate helps to produce a cleaner sound. However, the difference will not be noticeable in the final output.

You may also use a higher sample rate for burning audio to CDs without using 48 kHz. For example, 88.2 kHz offers twice the frequency range compared to 44.1 kHz, allowing you to reduce the risk of aliasing and higher latency rates.

Down-sampling from 88.2 kHz to 44.1 kHz is also less likely to produce distortion when using older conversion software. However, the latest conversion methods have helped to eliminate this problem. There is now no difference between down-sampling 48 kHz or 88.2 kHz to 44.1 kHz.

eVInteractive
Level 3

I think the OP wants to know what frequency the Quest's audio DSP runs at so he can author audio content at that "native" frequency.  This way there is no resampling overhead done by the device at runtime/load time - the audio can just be handed to the DSP as-is.

 

Another question along these lines is whether or not there is any hardware/parallel decompression available.

 

BetoV5
Level 2

I would also like to know if there is a prefered one for Quest. So far the only issues I've encountered with Unity are buffer related, I'm simply keeping all the sampling rate consistent across my files but that's it.