Nord User Forum

Wondering if it’s redundant to use 24bit instead of 16 , unless it’s a noticeable difference for Ns2

Not really. The only real difference is the noise floor, which jumps from "low enough to be well below the hardware's noise threshold" to "low enough to be ignored mathematically, as well".

analogika wrote:Not really. The only real difference is the noise floor, which jumps from "low enough to be well below the hardware's noise threshold" to "low enough to be ignored mathematically, as well".

Interesting to read that, thanks!

Dexibell are pushing their products on this 24 bit sampling, I did wonder what difference it would make if any.

Its not quite as simple as just noise floor. It can also impacts dynamic range. It really depends on how its structured and what the overall signal chain is (both input to output).

As an example, Alesis upgraded the DACs in their QS series keyboards from 18-bit to 24-bit and it did make a difference. Its subtle but definitely cleaner/clearer (and I'm not talking about noise).

But its an entire path that depends a lot on implementation so 24-bit all by itself is not really all that meaningful.

Better converters sounding better really doesn’t have much to do with the sample depth, though.

analogika wrote:Better converters sounding better really doesn’t have much to do with the sample depth, though.

But sample depth does have something to do with headroom and dynamic range

It might be interesting to read this test, if you're not already familiar with it. May bring some additional perspective:

http://archimago.blogspot.com/2014/06/2 ... i.html?m=1

The conclusion reads "In a naturalistic survey of 140 respondents using high quality musical samples sourced from high-resolution 24/96 digital audio collected over 2 months, there was no evidence that 24-bit audio could be appreciably differentiated from the same music dithered down to 16-bits using a basic algorithm (Adobe Audition 3, flat triangular dither, 0.5 bits)."

baekgaard wrote:It might be interesting to read this test, if you're not already familiar with it. May bring some additional perspective:

http://archimago.blogspot.com/2014/06/2 ... i.html?m=1

The conclusion reads "In a naturalistic survey of 140 respondents using high quality musical samples sourced from high-resolution 24/96 digital audio collected over 2 months, there was no evidence that 24-bit audio could be appreciably differentiated from the same music dithered down to 16-bits using a basic algorithm (Adobe Audition 3, flat triangular dither, 0.5 bits)."

This is the tail wagging the dog. They started with 24/96 samples. Try the same experiment with 16 bit source material and lets see what you get.

Just the initial sampling process will be different, because your headroom is significantly more restricted in a 16-bit environment.

This is no different than shooting photos. I shoot RAW because it gives me the largest dynamic range and I can push the files around. I don't deliver 14-bit RAW photos, I deliver 8-bit Jpegs. The final images look great, but I wouldn't want to shoot 8-bit Jpegs as my starting point.

As I stated earlier in the thread, there is an entire signal chain involved in this. Having a 24-bit audio engine gives you a lot more headroom. so a single sample probably won't make any noticeable difference, but when you start adding things up, it does have an affect. The NS3 has 6 engines and has to mix a lot of signals before it gets to the output. If your audio chain is 16-bit, you'll run out of headroom a lot sooner than if the audio chain is 24-bit or higher. DAW's actually run their mix engines at much higher bit depths specifically to address this issue.

24-bit samples and DACs don't magically make everything better, but it can make a difference if implemented correctly.

The theoretical dynamic range of 16 bits is 96 dB.

There probably isn’t a sampler in existence with hardware that has a -96 dB noise floor.

A higher-quality DAC circuit will produce a better sound. Just adding more bits to the sampler unit won’t.

That said, I revise my initial opinion, since higher bitrates do make a difference in processing — and samples are usually processed by dynamics engines and effects processors before they hit the DAC.