The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
if higher sample rate doesnt matter then why .... Effects Pedals, Units & Accessories
Old 20th September 2011
  #511
Moderator
 
narcoman's Avatar
 

Quote:
Originally Posted by UnderTow View Post
All periodic waves can be de-constructed into sine waves. Noise is aperiodic, isn't cyclic and thus can not be de-constructed into sine waves AFAIK.

Fourier said something to the effect that all music is sine waves (that might be what you are referring to) and indeed all musically recognisable tones can be seen as series of summed sine waves which means that almost all music is sine waves but it is not 100%!

That said, any series of sample points still can only have one possible de-constructed wave form.

Alistair
True, noise in itself isn't periodic but for it to traverse through any physical medium as sound it must necessarily be waves. Sound white noise is merely a summation of all frequencies; in that sense sound as white noise is sine waves - true white noise is all of a spectrum at equal power, in other words EVERY sine wave.

White noise isn't periodic any more than the speaking the word "numbnuts". But you could break both down into constituent parts, it's just that true white noise would take a long time

I'm not convinced that white noise in sound actually ever occurs. Perhaps radio static through a speaker? But even that is filtered when turned into sound.... not sure....
Old 20th September 2011
  #512
Lives for gear
 
DaveUK's Avatar
Carry on then .I for one have downloaded enough PDF's from this thread to keep me reading for a long time .....,saying nowt till then !
Oh and thanks for filling in the gaps (or joining the dots heh heh)
Old 20th September 2011
  #513
Gear Nut
 

Quote:
Originally Posted by MandyC View Post
3.no I dont think the points have to be anything...that's the problem...you can draw whatever you like in between those points...
No - because if you did - no two digital to analogue converters would sounds *remotely* the same. They all do it like this:


where T = 1/fs is the sampling interval, fs is the sampling rate, and sinc(x) is the normalized sinc function, i.e. sin(x)/x

(lifted from:
Whittaker–Shannon interpolation formula - Wikipedia, the free encyclopedia )

Hope this helps!
Old 20th September 2011
  #514
Quote:
Originally Posted by MandyC View Post
sure you can construct a sine wave from those point, but as I said, I can also construct a condo, a bicycle and a ladder...for it to be relevant to this argument we need at least 44,100 dots on that page...wouldn't you say at that point that any guess work between those samples could be fairly accurate

3.no I dont think the points have to be anything...that's the problem...you can draw whatever you like in between those points...


let me ask you this...is this semantics again?...are you just saying that because sound in its' purest form (a sine wave) exists on a curve always...that any point between two symetrical points (samples lets say) at the top of the curve is greater in amplitude than those same two points?
The fact that an AD converts sound into 1s and 0s, and a DA recreates that sound with any sort of accuracy surely convinces you that it works, right?

Quote:
Originally Posted by MandyC View Post
jeez I just realized I've been calling him Ocean this whole time...like an idiot...no wonder you guys wont listen to me
Don't know about anyone else, but I was starting to wonder...I thought there was some great theorist we didn't know about!

Quote:
Originally Posted by MandyC View Post
yes I get your points and they are valid...let me throw this out to you....what if all of a sudden we realized that gravity wasn't the earth pulling us towards it, but displaced space pushing us against it and that every mass doesn't actually have a gravitational pull but rather displaces the fabric of space which then pushes towards that mass...I know this is not on topic, but it's plausible and at the very least possible...so again...considering a possibility is not such a bad thing...let's face it, if the idea was completely ridiculous this thread would be over.
I must also confess no real knowledge of this past A-level...but I'll bow to narco's space-time knowledge. The only reason this thread is not over is because there's various posters (yourself included, though you're very pleasant with it) arguing against the facts!
Old 21st September 2011
  #515
Gear Head
 

Quote:
Originally Posted by UnderTow View Post
All periodic waves can be de-constructed into sine waves. Noise is aperiodic, isn't cyclic and thus can not be de-constructed into sine waves AFAIK.
It doesn't matter if it's aperiodic, it's still sine waves. Now it depends on how you are looking at the wave - the so-called 'window' - you use to observe it as to how you may determine its form. However, at it's fundamental level - and the state in which it applies in physical reality as a propagating wave - it is all sine. (at least in accordance with our current understanding of the limits of our physical knowledge of the universe).

This is why a rainbow happens. Have you ever seen a rainbow whilst taking a shower? The random droplets of water dither your noisy white light in to the various colours of the rainbow through refraction. It's also why your 3D glasses work at the cinema. Since waves are actually three-dimentional spirals, you can polarize them even when the source light may be construed as "noise".
Old 21st September 2011
  #516
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by narcoman
True, noise in itself isn't periodic but for it to traverse through any physical medium as sound it must necessarily be waves.
Quote:
Originally Posted by child of Gaia View Post
It doesn't matter if it's aperiodic, it's still sine waves. Now it depends on how you are looking at the wave - the so-called 'window' - you use to observe it as to how you may determine its form. However, at it's fundamental level - and the state in which it applies in physical reality as a propagating wave - it is all sine. (at least in accordance with our current understanding of the limits of our physical knowledge of the universe).
Thanks for the correction guys. For some reason I was thinking it would be referred to differently but the more I think about it, the more it makes sense.

Quote:
This is why a rainbow happens. Have you ever seen a rainbow whilst taking a shower?
Sure. Or watering the garden. Or after knocking that water hydrant over with the car. ;-) (That last bit didn't really happen ).

Quote:
The random droplets of water dither your noisy white light in to the various colours of the rainbow through refraction. It's also why your 3D glasses work at the cinema. Since waves are actually three-dimentional spirals, you can polarize them even when the source light may be construed as "noise".
Interesting. I never thought of light in that way. (And this is exactly why it is worth continuing these kind of threads. Wait long enough and you are bound to get a different perspective on something).

Alistair
Old 21st September 2011
  #517
Lives for gear
 
Greg Curtis's Avatar
 

Quote:
Originally Posted by unity
... and this can very easily be demonstrated wrong the next time you stand on the earth and don't float away.
I'll test this the next time I'm on Earth.
Old 21st September 2011
  #518
Quote:
Originally Posted by child of Gaia View Post
[...] Since waves are actually three-dimentional spirals, you can polarize them even when the source light may be construed as "noise".
Just when I was feeling confident with intersample peaks...
Old 21st September 2011
  #519
Lives for gear
 

Quote:
Originally Posted by Alexey Lukin View Post
I think that I've come up with a formulation that we can all agree upon (including OAG):

The concept "intersample peaks" barely means that analog waveform may peak between time instances where it has been (or will be) sampled. It is not necessarily a problem, but may become a problem when such peaks are unaccounted for and clipped, which would be called "intersample clipping".
close enough
i like the previous formulation better
but note if the d/a is desgined right there is no clipping

i wonder if some of the alleged clipping is just
the distortion from truncating the sinc like recreation waves

i would feel much better to totally omit intersample in the defintions. there is nothing between the samples in the digital domain. there are no samples in the analog domain.
hence no intersamples ever exist anywhre except to confuse people with meaningless jargon.
Old 21st September 2011
  #520
Lives for gear
 

Quote:
Originally Posted by DaveDaveDave View Post
No - because if you did - no two digital to analogue converters would sounds *remotely* the same. They all do it like this:


where T = 1/fs is the sampling interval, fs is the sampling rate, and sinc(x) is the normalized sinc function, i.e. sin(x)/x

(lifted from:
Whittaker–Shannon interpolation formula - Wikipedia, the free encyclopedia )

Hope this helps!
great math
no hardware can do that
Old 21st September 2011
  #521
Lives for gear
 

Quote:
Originally Posted by UnderTow View Post
You, Narcoman and Theblue1 are giving him waaaay too much credit! With comments like "infinite sample rate is identical to analog !!" it is clear he really does not understand Digital Audio.

Just look at this post: https://www.gearslutz.com/board/6866622-post27.html

I state very clearly "If you have your converters calibrated so that 0 dB FS is equivalent to say +4 dBu, this signal in this picture will cause a peak of over +10 dBu in the analogue domain!"

and he responds: "digital Samples do not equate to analog signals that way!! thta is the way you drew them. you cannot mix domains!!!"

He just does not get it at all.

Also let's not forget his very first post in this thread. "Some have derided my observation that higher sample rates do improve quality." It doesn't increase quality. It increases bandwidth.

He has also claimed he can make a linear compressor that could compress things as much as you want with zero distortion. Compression and linear in the same sentence is already an oxymoron. Compressing as much as you want with zero distortion is plain stupid.( https://www.gearslutz.com/board/6883786-post11.html )

oldanalogueguy does not understand the most basic principles of analogue audio let alone digital audio.

Alistair
higher sample rates do increase bandwidth
but we throw the extra away
the result is more accuarcy in the audio band

you can try to mix the domains
but it is illogical nonsense

do your nonsense again
draw an infinite number of samples on the audio like you all do so often
they will be identical
a nice smooth audio curve
no peaks no problems
perhaps you would prefer isomorphic rather than "identical"
Old 21st September 2011
  #522
Lives for gear
 

Quote:
Originally Posted by psycho_monkey View Post
Assuming your "ocean" is OAG, he may not have belittled anyone in this thread directly, but he certainly has in others (in fact, he mocked my degree in another one, albeit quite lamely) and in persisting in this train of confusion, he's kind of belittling the work of everyone.

Narcoman likewise has not made ANY snarky remarks to my knowledge - if you knew the guy personally, you'd know that he has the credentials to back up his comments, and whilst you may have been doing this stuff as long as he has, your comments show you don't understand the science behind it anything like as well as he does. Which is absolutely fine, no one's saying you have to to make good recordings, but don't pretend like you do.
....and no-one is denying that! well, not "non-functioning" but poorly designed (and often cheap) DA conversion. That's not the point. The whole argument has stemmed from OAG's stubborn assertion that "intersample peaks do not exist", and it's taken 17 pages to ascertain that he's not arguing with the way things work, merely the terminology - basically thinks the world should realign itself to his way of wording things. However, his poor way of communicating this (along with the provocative stance and way of posing questions, which no-one else on this thread has done!) makes 17 pages of confusion, several knowledgable people explaining things in simple words, whereas if he'd actually stated what he meant in plain English at the start, that would have all been avoided! Really, a 2 page thread at most has been turned into a 17 page argument because of one person's lack of ability to communicate what they mean, along with a stubborn streak.

are you referring to you spl degree?
sorry if you felt demeaned


lots of folks have denied the d/a has anything to do with it
including peaks and clipping
Old 21st September 2011
  #523
Lives for gear
 

as to gravity and quantum stuff
you need a 248 dimensional model to unite them
one guy almost has it done

each dimension equates to a particle in his approach

i am wondering if gravity is not really the warping of time
instead of gravity appearing as to slow time down

worse is that we can never know if there are not more universes on our ten dimensional membrane or more universes along the eleventh dimension of quantum theory . what happens when we go to 248 is going to take a while to figure out !!
Old 21st September 2011
  #524
Lives for gear
 
dcollins's Avatar
Quote:
Originally Posted by oldeanalogueguy View Post
higher sample rates do increase bandwidth
but we throw the extra away
the result is more accuarcy in the audio band

you can try to mix the domains
but it is illogical nonsense

do your nonsense again
draw an infinite number of samples on the audio like you all do so often
they will be identical
a nice smooth audio curve
no peaks no problems
perhaps you would prefer isomorphic rather than "identical"

This thread is far from its peak. I predict great things to come.

Entropy increasing.


DC
Old 21st September 2011
  #525
Gnu
Here for the gear
 

Quote:
Originally Posted by oldeanalogueguy View Post
i would feel much better to totally omit intersample in the defintions. there is nothing between the samples in the digital domain. there are no samples in the analog domain.
hence no intersamples ever exist anywhre except to confuse people with meaningless jargon.
Wow, this thread is still here? That's awesome.

Yes, you're arguing for the sake of argument; the last straw for me was watching you lamely attempt to pick a fight over dBFS. For the last time, no one tried to evoke "intersamples", and I've honestly never heard anyone anywhere attempt to use the word "intersample" as a noun. It's an adjective. The prefix is "inter-", from the Latin meaning "between", as in between the freaking samples. You have quite loudly reminded us that there is information between the samples, and continue to insist that no one understands that the source of this information exists solely in the analog domain. I don't think anyone cares anymore whether or not you believe how reliable the mechanics are behind the reproduction of this "lost" information by A/D conversion, but I do because you obviously need to find Jesus, who will be represented by a Jawa for the purposes of this post.
Old 21st September 2011
  #526
Moderator
 
narcoman's Avatar
 

Quote:
Originally Posted by oldeanalogueguy View Post
great math
no hardware can do that
as long as you window it you can get stupidly close. You know that.

The rest of it : it's clear it's the term you don't like. Well - let's call the same "things" flutabogs.
Old 21st September 2011
  #527
Gear Nut
 

This thread reminds me of this lyric:
"It was just like being on a fast ride at the fun fair. The sort you want to get off because its scary. And then as soon as you're off you want get straight back on again ..."
- Walk Away Rene, Billy Bragg
Old 21st September 2011
  #528
day 457, page 18 the thread trudges on past the body's of the dead forum contributors. when the end comes, it will be cockroaches and this thread, fighting over food.
Old 21st September 2011
  #529
Moderator
 
narcoman's Avatar
 

Quote:
Originally Posted by swim View Post
In the interest of resolution [of conflict, and otherwise], let's also stipulate that flutabogs exist in both the analog and digital world.
ha!
Old 21st September 2011
  #530
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by oldeanalogueguy View Post
higher sample rates do increase bandwidth
but we throw the extra away
the result is more accuarcy in the audio band
Binglybinglybing! And we are back at square zero.

Alistair
Old 21st September 2011
  #531
Moderator
 
narcoman's Avatar
 

Quote:
Originally Posted by UnderTow View Post
Binglybinglybing! And we are back at square zero.

Alistair
hurrrah!!! I love it!
Old 21st September 2011
  #532
Quote:
Originally Posted by oldeanalogueguy View Post
close enough
i like the previous formulation better
but note if the d/a is desgined right there is no clipping

i wonder if some of the alleged clipping is just
the distortion from truncating the sinc like recreation waves

i would feel much better to totally omit intersample in the defintions. there is nothing between the samples in the digital domain. there are no samples in the analog domain.
hence no intersamples ever exist anywhre except to confuse people with meaningless jargon.
As noted, from a communication/signal theory point of view, signal exists, even in the digital domain. It doesn't exist in the form of an analog signal in current, of course. It exists as the sample values and the math formulae used for DA reconstruction.

Signal doesn't disappear at AD and magically reappear at DA. It exists all along the signal chain -- even if the signal is transduced into other forms.

And that signal, so defined, does contain those peaks.


PS... I'm with swim on the term peaks. I actually have often used the term intersample overs a number of times in the past, since it seems more pertinent to the potential problem being described.
Old 21st September 2011
  #533
Lives for gear
 
DaveUK's Avatar
Groundhog thread
Old 21st September 2011
  #534
Gear Maniac
 

Quote:
Originally Posted by unitymusic View Post
Actually this is already a theory, but it's not necessarily only true at the expense of the "pull" theory being wrong. Lack of something can't push, only pull, like a vacuum. If space were displaced around an object with gravity, the only way for an outside object to be pushed towards that object would be the object with the gravitational "pull" pulling space and time towards it. If it were a pushing phenomenon, the gravity theory would have to be reversed so that the larger object had a lack of gravity comparatively, and this can very easily be demonstrated wrong the next time you stand on the earth and don't float away.

the larger object would displace more space...similar to say a pool...therefore creating more force towards the mass doing the displacing (crumpling like an accordion) space...similar to your body displacing water in a pool...not exactly the same...but anyway..."crumple theory" aside


I see that narc is an immovable object on this one in...

the beauty of our imagination is that you can also consider alternate possibilities, discounting them where necessary and embracing them where need be...
Old 21st September 2011
  #535
Quote:
Originally Posted by oldeanalogueguy View Post
i would feel much better to totally omit intersample in the defintions. there is nothing between the samples in the digital domain. there are no samples in the analog domain.
how about "implied signal peak between recorded samples"? cos that's not like cumbersome or anything...."intersample peak" is easily visualised once you understand how digital audio works...it's fine as a term.

What would YOU propose it's called? Because there's a phenomenon that needs a descriptor..regardless of whether it's a problem with a well designed DAC, the fact remains that the highest recorded sample may not represent the highest voltage when the waveform is recreated.
Old 21st September 2011
  #536
Quote:
Originally Posted by MandyC View Post
the larger object would displace more space...similar to say a pool...therefore creating more force towards the mass doing the displacing (crumpling like an accordion) space...similar to your body displacing water in a pool...not exactly the same...but anyway..."crumple theory" aside


I see that narc is an immovable object on this one in...

the beauty of our imagination is that you can also consider alternate possibilities, discounting them where necessary and embracing them where need be...
Thing is, none of this is relevant to how digital audio works. We've built AD/DA converters that do a good job, regardless if there's room for improvement. Much of quantum mechanics is still in the theoretical stage. Maybe the reason narcoman is immoveable is because he's got a better understanding of it than you?!
Old 21st September 2011
  #537
Lives for gear
Quote:
Originally Posted by psycho_monkey View Post
What would YOU propose it's called?
Don't go there, man. GS needs not additional cans'o'worms.
Old 21st September 2011
  #538
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by MandyC View Post
the beauty of our imagination is that you can also consider alternate possibilities, discounting them where necessary



Alistair
Old 21st September 2011
  #539
Moderator
 
narcoman's Avatar
 

Quote:
Originally Posted by MandyC View Post
the larger object would displace more space...similar to say a pool...therefore creating more force towards the mass doing the displacing (crumpling like an accordion) space...similar to your body displacing water in a pool...not exactly the same...but anyway..."crumple theory" aside


I see that narc is an immovable object on this one in...

the beauty of our imagination is that you can also consider alternate possibilities, discounting them where necessary and embracing them where need be...
There is never an alternate possibility against a theorem. Rallying against fact does not make one a free thinker.... Using such theorem to investigate new possibilities is what you should be doing, but you're so far you have locked yourself off from trying to understand how it all works in favour of believing a novice plausible story. Unfortunately your choosing which side to believe from a standpoint of ignorance of how it all works.

Read the papers, read Shannon, understand the theorems and THEN come back and tell us what you think.

That all sounds insulting - it isn't meant to.

As for being an "immovable static thinker" I'm afraid you couldn't be more wrong. The great brilliance of our imagination is coupling it with understanding from prior work and developing it further. Your own comment of "consider alternates, discount where necessary" is EXACTLY what one does to establish new work. Sampling theorem isn't new work - it's established.... I'm giving you guidance as to which cul de sac paths to discard. You are belittling experience and expertise.


Re- your gravitational theory - abstract thought is great, but you need proof to make it worthwhile. Secondly - current thinking isn't a million miles away from this BUT I am no expert in astro-physics. I am, somewhat, an acknowledged and published expert when it comes to digital theorem. Tough eh ? heh


Or to put it bluntly - maybe the reason I'm immovable on this is cuz I've got a fuc.king PhD in the area.
Old 21st September 2011
  #540
Quote:
Originally Posted by DaveUK View Post
Groundhog thread
I've never heard that term before.

Perfect.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump