Could not this easily be tested by a spectrum analyzer or something ?

___________________


http://soundcloud.com/subtara

quote:

Originally posted by ******** Nah man something is in those bits. It translates to something. Also the higher frequnecy the processor runs at the more bandwidth exists in the noise level. Like was in the link.. those cycles are bieng used for a filter or form, so it does directly relate. Also the encoding system would be different which may create different gaps. I disagree though... sounds beyond the audible human range can still be rendered audible .. and perceivable. I understand what you are saying and why, but you are wrong. Its probably in the field of "I can't really tell I don't give a damn" but for correctness it does matter and does exist.. but probably wont be relevant for most people. I've read studies on ultrasonics etc.. UHF, ULF etc.. and it does exist and is "perceivable" in many circumstances. It just isn't cognitional most of the time. Unless you are really intune. I could go into effects of atomic vibration on the human personal magnetic field etc.. but it may not lend much credence to providing information you are willing to accept. Frankly I could go into neural response based on soundwaves in how UHF and ULF and other beyond the audible range effect neural response, but I'm not going to at this point.

Understand, I'm not taking on perceptible audio quality of a high sample rate and high resolution audio, which I'm all for (I want a Korg MR-1000 for mastering purposes - that's some truly astonishing sample-rates, right there). I actually agree that there is a perceptible difference in quality, the higher the spectrum one is able to obtain.

quote:

Originally posted by ******** what you arn't clueing into is the nyquist limit thing 2x the maximum frequnecy.. http://en.wikipedia.org/wiki/Nyquist_frequency ex 96 allows up to double 48khz .. higher than 44 only allows 22khz. which goes into foldback dual channel and all that stuff... but IT DOES MATTER. ps music ins't ONLY for humans.. its for everyone..

This is more apt to the poster I was pointing to, however I can see where you're not understanding everything that I am understanding so I'll be abundantly succinct with this post. I pretty much agree with everything you're saying.








Even my kitties get a little hyper when I've got techno playing

quote:

Originally posted by EddieZilker 96kHz doesn't pertain to the audible spectrum of sound. It's referencing the number of cycles a processor takes per second. The higher the resolution of the initial recording, the better it's able to withstand processing and conversion down to CD quality without producing aliasing artifacts. It has nothing to do with canine or human hearing.

No. 96kHz refers to the frequency of samples. This means you can reproduce a perfect sine wave of 48kHz, according to the Nyquist sampling theorem (Which you should really read up on).
You are correct that it can reduce aliasing by moving it up to the inaudible range (Not every type of processing produces aliasing by the way). But this is already done internally in almost any decent synth/effect using oversampling. Working with ultra high sample rates is just redundant and can cause several problems. Not to mention it eats up a ton of harddrive space and CPU.

quote:

Originally posted by Eldritch No. 96kHz refers to the frequency of samples. This means you can reproduce a perfect sine wave of 48kHz, according to the Nyquist sampling theorem (Which you should really read up on). You are correct that it can reduce aliasing by moving it up to the inaudible range (Not every type of processing produces aliasing by the way). But this is already done internally in almost any decent synth/effect using oversampling. Working with ultra high sample rates is just redundant and can cause several problems. Not to mention it eats up a ton of harddrive space and CPU.

again redundancy is a matter of perspective depending on the type of music, the type of studio ...

Bob Katz, one of the most respected mastering engineers devotes a substantial amount of pages explaining the benifits of 96khz. Despite the upsampling that might occur, you have to keep in mind that each time you downsample, you will add noise to the signal path. Whether you can hear it, well some people seem to think so and those people are paid alot of money.

It is really similar to my bit thread. The real question is whether it will benifit the given situation.

Last edited by RichieV on Apr-26-2009 at 01:56

quote:

Originally posted by Eldritch No. 96kHz refers to the frequency of samples. This means you can reproduce a perfect sine wave of 48kHz, according to the Nyquist sampling theorem (Which you should really read up on). You are correct that it can reduce aliasing by moving it up to the inaudible range (Not every type of processing produces aliasing by the way). But this is already done internally in almost any decent synth/effect using oversampling. Working with ultra high sample rates is just redundant and can cause several problems. Not to mention it eats up a ton of harddrive space and CPU.

Why would I even entertain a theory invented by some loony, right-wing crackpot?

quote:

Originally posted by RichieV again redundancy is a matter of perspective depending on the type of music, the type of studio ... Bob Katz, one of the most respected mastering engineers devotes a substantial amount of pages explaining the benifits of 96khz. Despite the upsampling that might occur, you have to keep in mind that each time you downsample, you will add noise to the signal path. Whether you can hear it, well some people seem to think so and those people are paid alot of money. It is really similar to my bit thread. The real question is whether it will benifit the given situation.

You're contradicting yourself. Why downsample at all when you can work in 44.1kHz to begin with?

quote:

Originally posted by Eldritch You're contradicting yourself. Why downsample at all when you can work in 44.1kHz to begin with?

Because working in a higher sample rate yields better results when you, inevitably, have to down-size.

quote:

Originally posted by Eldritch You're contradicting yourself. Why downsample at all when you can work in 44.1kHz to begin with?

because there is only 1 instance of downsampling.Compare this to the multiple up and downsampling of each synth and effect , well you can imagine is much more than 1. That is pretty much the jist of why there are many benifits.

quote:

Originally posted by EddieZilker Because working in a higher sample rate yields better results when you, inevitably, have to down-size.

Yet you have provided no facts to back up your claim.

quote:

Originally posted by RichieV because there is only 1 instance of downsampling.Compare this to the multiple up and downsampling of each synth and effect , well you can imagine is much more than 1. That is pretty much the jist of why there are many benifits.

Are you just making things up as you go along?
Internal oversampling is done in multiples. Not resampling to a fixed rate like in the case of 44.1kHz -> 96kHz or vice versa.
So a 96kHz signal would still be oversampled.

RichieV Eddi and ELdritch are all wrong.

The best sample rate is 1.21 giggawatts.

___________________
i'm the alchemist without the cyst without a doubt out of the mist
dig out the grout expose the pest to take it out without the rest

they will attest my patients restorations from the best

i'm from the west i'm not an acmeist only the alchemist

quote:

Originally posted by Eldritch Yet you have provided no facts to back up your claim. Are you just making things up as you go along? Internal oversampling is done in multiples. Not resampling to a fixed rate like in the case of 44.1kHz -> 96kHz or vice versa. So a 96kHz signal would still be oversampled.

What happens when that synth which is clocked at 44 upsamples to 96 and eventually downsampled to a bus at 44 khz ?
What happens when that particular effect(upsampled then downsampled) is on a 44 bus ?

The result is alot of downsampling and alot of noise.
How many synths and effects do you use in your chain?


And i've provided you with a book by one of the most renowned mastering engineers who seems to think you are wrong. What exactly have you provided other than childish accusations.

read pages 52 - 53 in Katz book. Or 253 is you really just don't getit. Although i would feel really bad about this concerning i purchased it , you can probably find a link pretty easily. I did to find the exact page numbers since i'm not at home.

But honestly, what is with the attitude. I'm almost certain there are elements we all don't quite present correctly. Quit being such a little B.I think something everyone will agree with is that only professional mastering/engineers will benifit and regardless if i'm wrong or your wrong, i think we can both agree on that.

Last edited by RichieV on Apr-26-2009 at 13:24