Hello,

I guess my questions can be best answered in the mastering section - please feel free to move it elsewhere.

I record and mix Pop, Rock, Singer/Songwriter.
At the moment I use an older Aurora 16 converter.

My ears tell me that it performs better at 96 kHz, well at least it sounds more open, so it is different for sure. (Bad converter design?)

I read a lot of threads about IMD, Aliasing, Ultrasonic noise, Dithering and so on.

But still I‘m not quite sure what to make out of it concernic my actual mixing work flow.

Usually I dither to 24 bit even at playback with NotJustAnother Dither, and to 16 bit when creatinf the final file. It does sound less veiled and has more depth. I feel this more than hearing it.

But what to do with all the internal oversampling options in different plugins while mixing? Can I just leave all saturation, limiting plugins etc. at a high oversampling factor? Or is it best to switch it off when working with 96 kHz recordings? What logic is best here? How to find the „sweet-spot“ of specific plugins?

I also heard non-linear processing creates ultrasonic trash, which can have effects on the audible range. How to get rid of it? A ultrasonic filter after every plugin in the mix? Or only on the master bus? Which plugin is best for this purpose?
Tokyo Dawn Ultrasonic Filter? Anything better? Is this problem also true for analog recordings, as converters have LPFs already? Will digital processing create new ultrasonics in recorded audio when working in 96 kHz or is the problem only found in digital synthesizers and virtual instrument?

Many questions - but actually it‘s only one problems.

How to get rid of aliasing, IMD, ultrasonic trash while mixing, and still have the advantage of improved processing @96 kHz? I want to get the best results possible in the digital domain. I do not need superhuman sound fidelity. I don‘t record super detailed classical music. But my setup should still not have any ugly sound artifacts and masking effects that are not supposed to be there. Even if most people won‘t hear it conciously.

Thank you in advance for your help.

Well, there's "best practice", and there's what you learn from experience. A combination of the two usually leads to better sounding recordings.

In other words, research a lot, put that knowledge into practice, observe results, adjust methods, rinse and repeat.

And at the end of the day, if you can't hear a difference, is there one?

That's what I do.

I always try to improve - first of all as a musician and an artist, but also in my understanding of the gear I use and the tools.

It's just funny, that there are now common best practices, a certain standard way to do things in the digital domain - that I know of.

If I can not hear it, I don't have to change a thing of course.

I do notice a difference between 44.1 and 96 on MY converters. 96 sounds better.

I do notice a difference when switching on dithering during playback, and on the different outputs when going out to my summing unit. (24 bit!!)
(Sounds like some of the nasty digital artifacts are gone, little bit more refned and detailed.)

I do notice some plugins add a coldness to the sound, or when distortion/saturation sounds 2D and ugly.

But often it's hard to judge, whether artifacts are the result of wrong/bad mixing decisions or an accumulative effect of shortcomings of digital software.

So I was hoping to get practical tips, what you guy use and how to utilize it to compensate for those weak points of digital processing.

I’ve always wondered how people can hear “effects” of 24 bit dither on converters with enob of 18 to 20 bits...
Something must be wrong with gain structure somewhere...

...maybe

I can hear 24 bit output with dither to 16 bit on vs off. 24 has more depth and definition, of course. 24 dithered vs undithered - doubt anyone can. Too low a level.

96 sounds better on a couple of converters I have worked with, both ADC and DAC. So just upsampling a CD rip and listening to it at 96 kHz does have sense.

96 should sound better on properly designed converters... think about it!

As for up sampling a CD, it shouldn’t sound any different as the data is still the same... just a lot more zeros in the code if it’s transfered digitally... now if you rip it from the DS converter into your system then it will definitely sound different.

But it does due to the DAC filtering done differently in terms of bandpass range etc, so probably (most of) distortions move to above audible range. It is a minuscule difference, but still.

You may or may not need oversampling to be applied while at 96khz. If a plugin you use, for example, is sounding correct at 44.1khz with 2x oversampling, it is likely to sound just as correct at 96khz with no oversampling, assuming that the oversampling filters are not shaping the sound in a significant way.