Recording at 48000Hz??

  • Thread starter Thread starter Croww
  • Start date Start date
Well, first off, by definition, digital isn't 100% accurate.

Your converters need to see 2 points to convert audio without aliasing, but that is different from the sort of sonic accuracy that we are talking about here.

At 48k, if you were to record a 24k sine wave, you would get two points for every second of audio. It would record 2 amplitudes of that wave each second. Simple math will tell you that if you record that same wave at 96k, you'd get 4 amplitudes/second. This is much more accurate, and would be an audible difference if we could hear this high. (if we had dog's ears, this would be completely obvious.

Now, granted, we can't hear 24k. But when you take the idea of this into account, it would make sense that you would get more detail with higher samplerates, but it would only be much more obvious in the higher end of frequencies.

Read up on Nyquist theory for more info about this.

Im not convinced that two dots inside a bandlimited signal is any less accurate than 100
 
Yes, we only hear to 20k (more like 16k for most of us, even with healthy ears). But when we move up to 96k, you give your converters that many more samples to capture shorter wavelength (higher frequency) information. So you'll get a much more accurate high end. 44.1 and 48 sound fine (great even, if you can mix!) but there is a lot more "approximation" when it comes to high frequencies. .

It's now official, you have no idea what you are talking about or how digital audio works. Pat your self on the back, you have made the internet a little dumber today.
 
It's now official, you have no idea what you are talking about or how digital audio works. Pat your self on the back, you have made the internet a little dumber today.

Well, rather then just insulting me, why not shed some light on the subject? I'm willing to listen to what you have to say!
 
Well, rather then just insulting me, why not shed some light on the subject? I'm willing to listen to what you have to say!

Insults are more fun and take less time.:p

Perceived differences aside, because I don't claim to hear anything or take either side of the issue--

The idea that extra data points taken with higher sample rates increases the resolution of the reproduced wave is false. As long as the sampling frequency is more than twice the frequency to be sampled, the wave can be perfectly reproduced with two data points. No more are needed, and more data points don't give you a more accurate reproduction.

There is a lot going on in the hardware that reproduces that 20kHz. The samples are put through an iterative (repeating) process that will perfectly reproduce a frequency up to the Nyquist limit.

A 20kHz signal is reproduced the same whether it is sampled at 48k or 96k. It's amplitude doesn't matter, whether it's a sine, triangle, or square wave doesn't matter. 96k just gives you the ability to sample higher frequencies.

I won't claim to truly understand it, but this .pdf paper lays it out pretty well.

http://www.google.com/url?sa=t&ct=res&cd=1&url=http%3A%2F%2Fwww.lavryengineering.com%2Fdocuments%2FSampling_Theory.pdf&ei=hF6vR-_ZOJyGgwT4-s0b&usg=AFQjCNFpOgtvkqkJxFP_CsY6aKQK5_0IvQ&sig2=aE7D1zzdPIFnoIkEoahMwA


The bottom line is, if you hear a difference between something tracked at 96k vs. 48k, it's not because 96k is somehow more accurate.
 
Hmmmm. I read that whole article. The math is WAY over my head!

Aside from the math, which I'm sure could be convincing if I understood it, both camps make convincing arguments.

What I said in my posts was what I was taught in school, and made sense to me, but now I'm on the fence!!

Good thing my mbox only does up to 48k anyways!
 
I generally record at 48K 24 bit, you might be putting it on a dvd one day!
 
Hmmmm. I read that whole article. The math is WAY over my head!

Aside from the math, which I'm sure could be convincing if I understood it, both camps make convincing arguments.

What I said in my posts was what I was taught in school, and made sense to me, but now I'm on the fence!!

Good thing my mbox only does up to 48k anyways!

The argument on sample rates has been argued sooooo many times, why not just read the millions of threads on this subject.

TimOBrien summarized it well.
 
The main reason for having higher sample rates is to allow for better quality anti-alias filters (and indeed reconstruction filters).

This allows the filter to either be shallower or have a higher turnover frequency meaning that we can maintain a linear phase relationship across the audible range. As has been touched on already, a sample rate of 44.1kHz doesn't leave much room for a filter, maybe a couple of kHz.
 
That is soooooo not true that I don't even want to start...

I shouldn't say that -- Most "less seasoned" recordists I know record at high sample rates (mostly because of marketing hype). HOWEVER, 70-80% (depending on which polling data you read) of full-time audio professionals track at the target frequency.

Even the designers of some of the greatest digital conversion on the planet will plainly tell you that if you can hear the difference between 44.1/48 and 88.2/96kHz in their converters, they're broken. Heck, I track classical sessions at 44.1kHz. Always in 24-bit of course - That's a different story.

That all said - Yes, resampling to lower rates from higher rates (especially "goofy" conversion like 48 to 44.1kHz) can damage the audio. Some SRC's are very decent, some are very bad. Depends on what you're using on what.

I always want to have the possibility that my recordings could be used for vinyl or a different medium than CD. What do you do if you recorded everything at cd quality, and along comes a really wicked format that catches on, or if you get something licensed to film (which is usually done at least at 24bit 48kHz)... Maybe in some of the ultra high end converters, you can get just as good of a result at 44.1 but, a lot of the ones I've used in my own personal studio (around the $800 - $1200 range) have had a bit of a difference when dealing with some tracks beyond 48kHz.. mostly in plugin reverbs, and some soft synths. Guitar and shit like that, doesn't make so much of a difference, tho.. I find it's more prevalent when it's being generated via software at different sample rates. I can actually hear a bit of a difference on some material, which is why I do most of my stuff that has heavy synth use at 24bit 96kHz. Rock stuff, usually I do at 24bit 48kHz tho.
 
Well, rather then just insulting me, why not shed some light on the subject? I'm willing to listen to what you have to say!

THis is like the eternal pissing match subject of this forum. Why doesn't everyone realize...some people can hear the difference, some people can't. mathematics doesn't prove anything when you're dealing with something like this. Our ears don't hear in equations.

If you can't hear the difference, then don't bother, as it wouldn't improve your mixes with what you'd gain anyway.. if you can, use it when it's needed.

Why is this such a bitter issue with so many people? It's just silly.

I do a lot of stuff at 24bit 96kHz, but I think I might reduce some to 48k, as even though there's an audible difference between those to me, I'm starting to not give so much of a shit with my allready distorted loud obnoxious mixes lol. Just do what works...about the only time it starts to get stupid is when you go up to 192kHz, I can't hear any difference at all with anything between 96k or 192k.. the difference between 88.2k and 96k is tiny even.
 
Thats kind of like saying "Prayer works, except when it doesnt"

Its either testable or its not. No real test has shown people can hear much above 20khz if even that.

If there IS a difference, its gotta be something to do with the antialias system, and why Dan Lavry's idea of a 60khz target makes sense IMHO
 
Its either testable or its not. No real test has shown people can hear much above 20khz if even that.

Well, that's sort of a different issue, isn't it? It's general fact that 20khz is the top of our hearing, but its closer to 16.5 for most men, and 17.5 for most women. 44.1 reproduces 22.050 and down right? So if people can hear a difference between 44.1 and 88.2, wouldn't that mean that it would be because of higher detail, rather than larger bandwidth?? :confused:

Also, yeah, the 60khz thing in that article totally made sense to me, but wouldn't that also lead one to assume that 96khz would therefore also sound better than 44.1/48? Even if it isn't as efficient as far as computer performance?

As far as "testing" anything like this... I used to run sound at a club and we had an 01v96. we usually ran it at 48k just due to the fact that you lose 2 fx if you go 96, but we tried a/b-ing 48/96 one night and all three of us working the gig could hear a difference. And I've had way too many long nights at loud clubs so I'm sure my hearing is nowhere near 20k.

Anyways, tell me how wrong I am now!!! :D Haha, not trying to start a fight or a bitter debate about anything. If I was wrong about the math, that's fine, but like someone else said, the ears don't hear equations!
 
THis is like the eternal pissing match subject of this forum. Why doesn't everyone realize...some people can hear the difference, some people can't. mathematics doesn't prove anything when you're dealing with something like this. Our ears don't hear in equations.

My point is that if you hear a difference, it's not because 96k makes more accurate samples or captures more accurate data than 48k. There is something else that accounts for the difference you hear.
 
Well, that's sort of a different issue, isn't it? It's general fact that 20khz is the top of our hearing, but its closer to 16.5 for most men, and 17.5 for most women. 44.1 reproduces 22.050 and down right? So if people can hear a difference between 44.1 and 88.2, wouldn't that mean that it would be because of higher detail, rather than larger bandwidth?? :confused:

Those numbers are wrong. The highest reported frequency heard by humans is an astounding 120 kHz.

I'm not saying that all humans can hear above 16.5 kHz, and you don't hear ultrasonic sounds in the same way as you hear ordinary sounds, but to say that humans can't hear it at all is just plain factually incorrect, and there are plenty of studies that have proven so repeatedly.

:D
 
sure... but who cares really.. it doesn't matter in reality why, it just does make a difference. Nobody is going to buy a record because of the sampling frequency used on it.

Just use whatever works...who bloody cares... if someone wants to record everything in 8 bit and it sounds great, then good, do it.
 
44.1 reproduces 22.050 and down right?

No- for two reasons-

First, the sample rate must be MORE THAN double the highest frequency to be sampled.

Second, the input filters cut off sharply starting at 20k. By the time you hit 22k, the signal is down about 100db. So while yes, a 44.1kHz sample rate could approach 22k, stuff that high never makes it in.



So if people can hear a difference between 44.1 and 88.2, wouldn't that mean that it would be because of higher detail, rather than larger bandwidth?? :confused:

Not necessarily. There are lots of reasons you could hear a difference.
 
sure... but who cares really.. it doesn't matter in reality why, it just does make a difference. Nobody is going to buy a record because of the sampling frequency used on it.

Just use whatever works...who bloody cares... if someone wants to record everything in 8 bit and it sounds great, then good, do it.

I love this quote!...

Could it be possible be some people (just a few at least) get a little dissapointed when they spend their hard-earned on a device that 'captures' 57874khz, but are told that 44.1 or 48k is really all that is required for high quality recording ?....Sure, I believe that the math matters, but maybe the quality of the music and musicianship is worth more than those farking numbers.??
oh...I'm very happy since i went back to 44.1 big black dogs @ 24 pope shaped yogurt throwers....but hey, that's just me....
 
Back
Top