Question for Ty

[Take Vos]

Hello Ty,

I have done some research in sample conversion and found the following gem. You virtually get a higher bit depth when recording at a higher sample rate (if the AD converter keeps its absolute accuracy at higher sample rates). In the 80’s we would have called this oversampling.

Imagine the following four samples recorded at 192 kHz.
0.00, 0.25, 0.50, 0.75

If this gets naively down converted to 96 kHz you get:
0.125, 0.625

If it gets down converted to 48 kHz you get:
0.375

Each sample rate conversion step, gains you one bit of precision.
Actually 24 bit AD converters do this internally, they sample at a high sample rate at low bit depth and then down converts the sample rate.

In principle you don’t have to down convert at all, you can send 192 kHz to the speakers and the analogue filters and mechanical limitations would do the sample rate conversion for you.

Cheers,
Take Vos
developer of Boom Recorder

[Ty Ford]

[Take Vos] “I have done some research in sample conversion and found the following gem. You virtually get a higher bit depth when recording at a higher sample rate (if the AD converter keeps its absolute accuracy at higher sample rates). In the 80’s we would have called this oversampling…..

Each sample rate conversion step, gains you one bit of precision.
Actually 24 bit AD converters do this internally, they sample at a high sample rate at low bit depth and then down converts the sample rate.

In principle you don’t have to down convert at all, you can send 192 kHz to the speakers and the analogue filters and mechanical limitations would do the sample rate conversion for you.”

Hi Take,

Right, but as I said earlier, given the limitations of human hearing and the accuracy and bandwidth of playback systems, the reason to go there is not compelling enough for me to go for it on any regular basis, other than “because we can.” If you have a VERY, VERY high end project that will benefit from it (and broadcast or DVD distribution doesn’t warrant it) then fine.

Regards,

Ty Ford

Ty Ford’s “Audio Bootcamp Field Guide” was written for video people who want better audio. Find out more at https://home.comcast.net/~tyreeford/AudioBootcamp.html
or https://www.tyford.com

[Take Vos]

Hi Ty,

I guess you are right, I mean most people think mp3 files are CD quality.
As for “because we can” argument, right now HD-DVD offers (lossless compressed) 5.1 surround 24 bit / 96 kHz. And most new consumer amplifiers can handle this now as well (it is in the HDMI standard).

And I guess high frequency recording is nicest for people creating special effect, who routinely slow down recorded audio to design interesting sounds.

Cheers,
Take

[Take Vos]

Hi Ty,

I guess you are right, I mean most people think mp3 files are CD quality.
As for “because we can” argument, right now HD-DVD offers (lossless compressed) 5.1 surround 24 bit / 96 kHz. And most new consumer amplifiers can handle this now as well (it is in the HDMI standard).

And I guess high frequency recording is nicest for people creating special effect, who routinely slow down recorded audio to design interesting sounds.

Cheers,
Take

[Toby Dalsgaard]

There seems to be an element of confusion going on in the A/V world right now about 96K audio, which no thanks to Digidesign’s marketing dept, many people consider 96k to be “Hi-Def” audio: a technical spec that needs to be married with Hi-Def picture. It’s not. 96k is 96k. It’s has nothing to do with HD picture. You can still edit in HD in 48, 44, or even 32. Frankly, unless you are a Lakeland Terrier, I don’t see why you’d need/want to capture in anything higher than 48K. I’d be very curious as to why that director feels he needs to capture on-set audio at 96k (assuming the reason is legit and not just ill-informed techno paranoia)