Jump to content

Recommended Posts

christt   

Hello All,

I would like to know if recording at 96khz and exporting to 48khz  resulting in a better sounding file.

Is it like in video that 4k exported to 1080p looks better than 1080p natively.

 

Thank you in advance.

Share this post


Link to post
Share on other sites
christt   

For recording  I would like to know for every type such as dialogue, foley, music, etc

For purpose for online, tv, computer.

Share this post


Link to post
Share on other sites

The conversion from 96kHz to 48kHz is a mathematical process that can be done with as much precision as you desire, given the prcessing power and time available on a PC.   So the resulting file will have the same "qualities" as the bottom 24kHz of the signal if it was recorded at 48kHz. 

So the only way for a recording at 96kHz that is converted to 48kHz to be better than a recording at straight 48kHz is if the 96kHz converter has better specs (S/N, Dynamic Range, linearity, clock jitter) than the 48kHz one.    A lot of the time, you're looking at the same converter just running it at different frequencies, so the specs will be mostly the same.

Since all common ADC converter chips are delta-sigma types, outputting at 96kHz can actually result in the chip doing less decimation stages, and possibly having better specs at the higher rate than the lower one.

So in theory, recording at 96kHz and doing a high quality down-sample to 48kHz can result in a better recording, but only if the ADC (and analog front end) actually have better specs to start with.  The downside of having to do the extra work on the computer usually puts this process out of the realm of day-to-day work, but can provide a little extra quality if it's a concert recording, for example.

The comparison with video @ 4K shouldn't be made, because the capacity of a 48kHz 16 bit recording is already great enough to cover the human ear in terms of dynamic range and hearing range, so it's already down to the subjective level.  (24 bit recording is the standard because you don't want to compromise both S/N and Dynamic range if you are recording quiety, and you don't want to dither that early on in the signal chain).

Video 2D downsampling and motion compensated frame rate adjustment are much heavier processes where the trade offs in processing time and cost are still a large part of the decision.

Share this post


Link to post
Share on other sites
christt   
On 5/11/2017 at 2:05 PM, Tom Duffy said:

The conversion from 96kHz to 48kHz is a mathematical process that can be done with as much precision as you desire, given the prcessing power and time available on a PC.   So the resulting file will have the same "qualities" as the bottom 24kHz of the signal if it was recorded at 48kHz. 

So the only way for a recording at 96kHz that is converted to 48kHz to be better than a recording at straight 48kHz is if the 96kHz converter has better specs (S/N, Dynamic Range, linearity, clock jitter) than the 48kHz one.    A lot of the time, you're looking at the same converter just running it at different frequencies, so the specs will be mostly the same.

Since all common ADC converter chips are delta-sigma types, outputting at 96kHz can actually result in the chip doing less decimation stages, and possibly having better specs at the higher rate than the lower one.

So in theory, recording at 96kHz and doing a high quality down-sample to 48kHz can result in a better recording, but only if the ADC (and analog front end) actually have better specs to start with.  The downside of having to do the extra work on the computer usually puts this process out of the realm of day-to-day work, but can provide a little extra quality if it's a concert recording, for example.

The comparison with video @ 4K shouldn't be made, because the capacity of a 48kHz 16 bit recording is already great enough to cover the human ear in terms of dynamic range and hearing range, so it's already down to the subjective level.  (24 bit recording is the standard because you don't want to compromise both S/N and Dynamic range if you are recording quiety, and you don't want to dither that early on in the signal chain).

Video 2D downsampling and motion compensated frame rate adjustment are much heavier processes where the trade offs in processing time and cost are still a large part of the decision.

Thank you for the input, very clear.

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×