44.1 kHz or 48 kHz?

[mixyguy2]

I'm just stating the fact that higher sample rates have clear mathematical benefits.

1. Except it's not a fact at all. It's an oversimplification at best, a fool's errand at worst.

2. It's also irrelevant, as MUSICAL benefits are all that matter, and aside from a modest latency benefit which oh 99% of all music produced won't notice one way or the other, recording at 96 offers no benefit, never mind even higher rates.

But hey, it's your delusion. Enjoy.

[jethrobull]

Just make music.....As good as you can.

Amen to that!

[roman.i]

Yea, sure, won't notice this, won't notice that, no one notices nothing.
Love this argument.

[Synthman2000]

"Math and industry standards already back it up."

I agree that high sample rates reduce aliasing in non linear process. That is not consequence free, there is an exchange, for other sonic ills on what you have yet to make commentary (broadband noise from extended bandwidth IMD) I see you ignore that issue.

Second, the point is not that I disagree with maths, what I am proposing is this aliasing is inaudible in almost all real world circumstances (especially with sensible use of OS plug ins) and a very good chance of all real world circumstance. The same is proposed for less steep 48kHz filter.

The reason ? Every other possible sonic parameter to which there is not the same degree of care, because it is easy to parrot something that is believed to be an absolute, but much more difficult to follow and make the correct dynamic /momentary requirements of audio production from start to end. Something I suspect you have very little experience of.

An increased electricity bill is what you are most likely gaining. (could be the difference 60W to 200W power usage of a modern CPU.)

I also saw you had written about OS plug ins and that they were no good or words to that effect. (forgive me if I am misquoting I have not the time to fully back track) that is a preposterous statement. And that premise is resting on stacking non linear processes.

The plug in choice itself and settings will have much more difference than any cumulative OS, if you can hear the differences on your monitoring that is ? I'll assume it is not mastering level until told otherwise.

Aliasing is buried by all other audio error and ill practices to the point of inaudibility and to suggest that running high SR's or even 48kHz from 44.1kHz will make audio sound noticeably better is very far from the reality.

There are far more important things to get right which audio producers fail at knowingly or unknowingly. People can freely choose their own sample rate, without being shamed into not being professional by yourself and safe in the knowledge that the choice of 44.1kHz is very much likely to be inconsequential in the real world.

[jamcat]

duplicate

[jamcat]

duplicate

[jamcat]

I agree that high sample rates reduce aliasing in non linear process. That is not consequence free, there is an exchange, for other sonic ills on what you have yet to make commentary (broadband noise from extended bandwidth IMD) I see you ignore that issue.

Actually, I commented on IMD waaaay back in this very thread in April. Here was my conclusion:

"[IMD boogeymen] seem a little misleading, not just because they rely on conditions that are not likely to occur in the wild, but also because it doesn’t seem to be a problem that is confined only to ultrasonic frequencies. IMDs can occur between frequencies that reside well within the 20 to 20k range, so neither lower samplerates nor ultrasonic filtering will solve those occurrences.

And since they occur in the analogue domain, then eliminating them isn’t relevant to our interests, if our goal is to eliminate digital artifacts to get closer to analogue."

Second, the point is not that I disagree with maths, what I am proposing is this aliasing is inaudible in almost all real world circumstances (especially with sensible use of OS plug ins) and a very good chance of all real world circumstance. The same is proposed for less steep 48kHz filter.

Aliasing is going to depend on the source material, the severity of the harmonic distortion, and the sample rate at which it is being processed. It also depends on how much other stuff is going on that might mask it. So it might not be audible in some situations, but be very audible in others. I am actually most concerned about brittle highs from phase ripple above 10kHz, due to Nyquist being too close to 20kHz.

[...] Something I suspect you have very little experience of.

Let's just dispense with the personal stuff already. You had been doing so well this time.

An increased electricity bill is what you are most likely gaining. (could be the difference 60W to 200W power usage of a modern CPU.)

No. I have an Apple Silicon Mac.
Also, I mix at 48kHz so I'm not using more CPU 99% of the time.

I also saw you had written about OS plug ins and that they were no good or words to that effect. (forgive me if I am misquoting I have not the time to fully back track) that is a preposterous statement. And that premise is resting on stacking non linear processes.

Yes, the premise is resting on stacking non-linear processes. That's what happens when you use multiple plugins in sequence. It's not a preposterous statement when you consider how many times your audio is getting downsampled back to 44.1 on the way out.

There are far more important things to get right which audio producers fail at knowingly or unknowingly. People can freely choose their own sample rate, without being shamed into not being professional by yourself and safe in the knowledge that the choice of 44.1kHz is very much likely to be inconsequential in the real world.

But 96 kHz is the professional industry standard sample rate for recording and working with source audio for the reasons we've discussed. Even if it's only slightly better, most pros choose slightly better over slightly worse if given a choice.

Generally speaking, as music makers we'll take every advantage we can get, no matter how small, regardless of our level.

[audiojunkie]

My point was, that their (Steely Dan’s) albums are held up as recording masterpieces, and yet they didn’t have the ability to use 24/96. They didn’t even have digital editing for the most part. It was largely analog mixing (manually) and tape splicing. Plenty can be done without the higher sample rate. For most people, 24/44.1 won’t be the weakest part of their production, it will be the recording quality (microphones), gain staging, giving each instrument its proper space in the mix, the choice of instruments and sounds, the accuracy of the instrument playing, the song writing, the production choices, etc, etc. I could go on and on. My point is that while yes, higher bit rate and sample rate is useful, there are many other things that may well be more important in the end product. The emphasis and endless debate over 44.1 or 48 is not nearly as important in the greater scheme of things, when looking at the production as a whole. It’s like looking beyond the mark, or not seeing the forest for the trees. It has to be kept within a proper perspective.

In essence, recording at 24/96 is a nicety, but alone is not the essential requirement that determines a quality production.

[vurt]

this week, ive been recording, analog synths, at 12 bit, 22500hz

admittedly, they're for a sampler.

[jethrobull]

this week, ive been recording, analog synths, at 12 bit, 22500hz

admittedly, they're for a sampler.

Just not good enough. You're obviously not a Pro

[PatchAdamz]

Lets Make Some Music.png

[osiris]

16 pages and no clear winner. I came just to say I do 48 but something I just thought of: some VSTs have the ability to change sample rates, Reaktor comes to mind and I haven't but should check they are set as the same sample rate as the DAW, because i don't know if that switches auto or what.

[El°HYM]

They can be set to the Host SR or independent up to 192kHz. Some Ensembles automatically double up the internal SR e.g. from 48kHz to 96kHz.

[jamcat]

My point was, that their (Steely Dan’s) albums are held up as recording masterpieces, and yet they didn’t have the ability to use 24/96. They didn’t even have digital editing for the most part. It was largely analog mixing (manually) and tape splicing. Plenty can be done without the higher sample rate.

And my point was that analogue recording doesn't have the same challenges as end-to-end digital recording. Analogue has its own problems for sure, but unnatural frequencies due to multi-process error accumulation isn't one of them. Analogue's main problem is limited dynamics due to low headroom and a high noise floor, which is what Steely Dan obsessed over. If they had been recording in the digital domain, they would have been obsessing over sample rates and aliasing filters instead.

[El°HYM]

I have a feeling this will not end very good.