What sample and bitrate should I set my sound card to...

32 but float contains more theoretical headroom without incurring distortion. 24 bit audio contains 144db of theoretical headroom. 32 has an insane amount higher and will allow channels that do not leave 32 but float to not incur digital distortion. This is the entire premise of modern DAWs.
Higher sample rates contains information up to the half the frequency of the sample rate, 44.1 has information up to 22.5Khz. Higher sample rates will influence different programs interact with the data. They will also change the way audio reacts when being pitched. They will also give different ways to interact with those hyper high frequencies. For example, a Maag EQ will have an air band for increasing volume with a 40Khz shelf. We can hear these effects because of various reasons.
Humans can not hear pitches above 20k but we can certainly perceive the effects of higher frequencies relationships with those we can hear.

You are an idiot and shouldn’t be advising anyone.

Smooth interpolation is not always what you seek in digital audio playback, so no, it's not objectively better

no because the conversion is the same
it will literally have the exact same result

>there is only one numerical method ever and it makes things smooth
Yikes...
What conversion, you literally have more information to work with, show me an algorithm that loses quality with more samples.

>We can hear these effects because of various reasons.
>Humans can not hear pitches above 20k but we can certainly perceive the effects of higher frequencies relationships with those we can hear.
this is the esoteric pseudoscience audiophiles believe in

An example of this is this:
Consider a recording of a snare at 44.1
It contains information for 0 to 22.5 kHz. We pitch this down one octave. Our fundamental has changed from 300hz to 150hz.
In fact all of the information has now been halfed.
Now consider hat in 48k sample rate we now have information from 0 to 24khz, that is 3.5k more of information.
When we pitch this down we will be able to perceive the information from 24k because it is now at 12k.
This is very useful if you are interested in sound design.

>this entire post

Attached: 1562325202159.jpg (638x1000, 61K)

Why do sound tech threads on Jow Forums always consist 95% of
>yikes
>ur dumb
>no u
>*wojakposting*

That is nearly the definition of a higher quality converter, to recreate from digital sample points, a smooth curve instead of the flat sharp curve that happens during clipping.
32 bit float is an internal method for avoiding this clipping behaviour.
Strong bait or retarded it doesn’t really matter does it

cope

Because Jow Forums is filled with faggots

read "Sampling fallacies and misconceptions"
people.xiph.org/~xiphmont/demo/neil-young.html

Also consider that most mics won’t record those u we higher frequencies, but a synth will generate them. We can abuse this to great effect for music and film. Ever heard a 96k frequency transposed to 12k? Try it yourself it’s quite bizarre.

I still don't see where I'm wrong. And if you can, give me an example with an open source codec.

You’re not wrong he is trolling.
That article is a good read though.

then watch this fucking video
youtube.com/watch?v=cIQ9IXSUzuM
he is wrong
This entire thread is a nest full of placebo effect damaged audiophiles

I’m not an audiophile I am a producer who uses freqs above hearing range for creative practises.
You can literally do this yourself.
Download a soft synth and create a tone in a high sample rate. Pitch it down. You have now perceived the effects of higher sample rates: it’s that simple.

Done, still not convinced.

You watched a 27 minute video in 5mins.
Dope skills bro

>I am a producer who uses freqs above hearing range for creative practises
you must be a shitty producer to throw your time away
>done
>video is 23 minutes long
ok

Are you fucking kidding yourselves I'm watching that shit right now? I asked a very simple question.

Sound design and experimenting isn’t throwing time away you smooth brained retard.
I was just joking don’t waste your time on this retards video.
What was your question? I’ll give you a straight answer right now.

>24 bit audio contains 144db of theoretical headroom. 32 has an insane amount higher
32bit float has 144db of dynamic range as well, but it can distribute it among a wider range.
32bit integer is different.

none of this is relevant to op, though.

Can you demonstrate how more samples = worse sound, how higher sample rates = worse sound, how higher bit rates = worse sound, and the mathematics behind all of it.

Well the truth is that they will probably just sound different although a human won’t be able to tell. Not better or worse unless you consider worse to be more distorted. You could test harmonic distortion to determine which sounds “better”but there are a lot of variables.
Sample rates above 44.1 will probably not be audibly different. Dynamic range will be, listen to transients.
More samples could sound worse depending on your converter. It absolutely is not a case of higher sounds better.
The only reason it will sound worse is hardware not converting it perfectly, which will ALWAYS happen. There is no product capable of truly recreating an analogue signal. The method of plotting points to convert will always produce a non sinusoidal conversion.

No I cannot display this mathematically that is some phd level stuff and I am a simple practitioner who is very interested in noise, loudness potential and psycho acoustics.

But it is never summed to cause digital distortion meaning internally one can work above digital 0 as long as it is never converted.
Ye it’s not relevant at all, OP should just use 24 bit 44.1.
In fact, if you were to use 32 bit you could cause yourself damage. I once generated a file in 32 bit float using PaulStretch that resulted in LOUD distortion, were talking 100’s of dB over 0.

Well thanks, video is interesting btw

Audio is very interesting stuff in general.
Music, the artistic application of audio is very complex. One can perceive the chord of a drum and the notes of a chord. We can perceive the timbre of a flute as opposed to the timbre of a sax. We can perceive crest of transients and we can perceive the delay between walls and a speaker.
This is all just nonsense and irrelevant but the point is, in the mix no one can hear your screen.

>don’t waste your time on this retards video
>don't waste your time learning something instead of believing anons placebo effects on a vientnamese basket weaving board
lmao at this board
you might learn something @10:00 after you stop being a high esotericist
youtube.com/watch?v=FG9jemV1T7I

Imagine being this much of an insufferable faggot

No one discussed anything about placebo effects. There are applicable uses for different sample rates and bit rates. That is not a discussion it’s a fact, I have already described ways you can do this yourself in any DAW with any synth.

Imagine having such a lacking understanding of a topic you engage with that you can’t even use your own words.
And no that doesn’t mean I don’t believe in citations. It means you are unable to explain your own thoughts.

>That is not a discussion it’s a fact, I have already described ways you can do this yourself in any DAW with any synth.
its called distortion you absolute mogoloid
there is no fucking reason for OP to use anything above 44khz/16bit and yet you subhuman audiophilic placeboeffect ridden spastics are still shitting this thread up with pseudoscientific babble

I already said OP should use 44.1 24 bit. There is an audible difference between 24 and 16 bit.
It’s not pseudoscience or babble you smooth brained retard.
It’s not distortion you spectacular retard it’s repitching inaudible frequencies to become audible.