The Day AudioNut Beat The Shannon Limit & Made Infinite BW Drums

Discussion in 'How to Win Friends and Influence People' started by AudioNut, Dec 13, 2018.

  1. AudioNut

    AudioNut Acquaintance

    Joined:
    Oct 2, 2018
    Likes Received:
    55
    Trophy Points:
    18
    Location:
    Ottawa, ON
    Apologies for being a little lazy with my earlier comment regarding percussion and bandlimiting. That was just intended as a simplification to avoid muddying the issue with input considerations. It *is* true that percussion in the real world is effectively not bandlimited, subject only to physical constraints relating to air movement. However, obviously all practical real-world microphones are bandlimited. Good, modern microphones do tend to have wider bandwidth than Redbook sampling permits, which is why any issue exists.

    I don't have enough time today to respond to all of the other comments. I will add, just for fun (because I think it's kind of cool, not because it's relevant to any of this) that if anyone graduated with an MSEE prior to 2007, it will probably blow your mind to learn that faster-than-Nyquist signalling is actually one of the hottest research topics in signals now. Faster-than-Nyquist (i.e., higher than Shannon capacity) used to be one of those things that everyone with proper training believed was impossible in practice (apart from one neglected 1975 Bell Labs paper that explored the issue), but now it's a viable contender for the 5G+ telecom standards. The math is really nuts though.
     
  2. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    So percusions are effecitvely not bandlimited in real world... Can you show me which one in these charts?
    1. http://2.bp.blogspot.com/_hwnCNuE0G...zxRHlIc/s1600/Interactive-Frequency-Chart.png
    2. http://www.fourmasterscaraudio.co.uk/caraudio/wp-content/uploads/2015/10/Frequency-chart.png
    3. https://aeveringtonblog.files.wordp...cy-spectrum-instrument-ranges-and-eq-tips.jpg
    4. http://www.zytrax.com/tech/audio/audio.html#frequencies
    In 4. above, the 6th harmonic of a Cymbal is claimed to reach 24 kHz. This is a far cry from "effectively not bandlimited".

    Furthermore, why should one give two shits about ultrasonics? What is the issue here?

    EDIT (toned down since I may be misreading @AudioNut's intentions - not sure if a troll):

    I know beamforming very well (including the "really nuts" math). It does not beat the Shannon Limit. It increases S/N and therefore capacity. This thread does not deal with the Shannon Limit.
     
    Last edited: Dec 13, 2018
  3. AudioNut

    AudioNut Acquaintance

    Joined:
    Oct 2, 2018
    Likes Received:
    55
    Trophy Points:
    18
    Location:
    Ottawa, ON
    I don't mean to be rude, but I made clear in my original post that remark was just for fun and not relevant to any of this (I mean my post literally says: "because I think it's kind of cool, not because it's relevant to any of this").

    As for your point about the Shannon Limit and Faster-than-Nyquist, there's no other way of putting this, but you're being flat out ignorant. Faster-than-Nyquist signalling does increase the effective Shannon capacity of a channel. I can point you to dozens of recent papers, but here's a good start:
    I honestly don't know why you created a new thread to flame me and call me a "troll". It's pointless to contribute to online forums when people behave like this. Enjoy your little echo chamber.
     
  4. Thad E Ginathom

    Thad E Ginathom Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    14,130
    Trophy Points:
    113
    Location:
    India
    Just my ignorant opinion... No, I don't know exactly what Ultrabike does for a job, and I certainly don't know his qualifications. But I have a strong impression that he does do this sort of thing for a living. I don't know if you've bothered to get to know the guy and pick up on that? Not that that is any guarantee of him always being right, but, ignorant? I doubt it.

    Of course, I don't know you about you either. But Ultrabike's been around since forever, whereas you haven't.

    No, I am absolutely not suggesting CV/Certificate waving here! Just history makes it more likely that I am going to listen (in sofar as I have a clue what you guys are on about) to Ultrabike. And I'm probably not alone here.

    Whatever...
     
  5. Xen

    Xen Friend

    Pyrate
    Joined:
    Sep 30, 2015
    Likes Received:
    207
    Trophy Points:
    43
    Location:
    Houston, TX
    Shannon Limits main equation is C=B log [sub]2[/sub] (1+(S/N)).
    C = Shannon Limit of information (bits/s) per channel
    B = Amount of spectrum used (more spectrum, more information can be crammed into a channel)
    S = Signal (I think Faster-than-Nyqist increases this part)
    N = Noise (I bet Faster-than-Nyqist increases this part, too)

    So...Semantics? Shannon Limit (1948 by ... Shannon) is a limit, so only reachable under optimal conditions that will never be present in the real world (infinite spectrum or infinite signal or zero noise). Increasing "effective Shannon Limit" just means we are getting closer to the Shannon Limit.

    As an analogy, Carnot theorized the perfect heat engine in 1824, and we have YET to design something that can exceed the perfect Carnot cycle. We are getting closer to maximum efficiency, but will never reach it as the most efficient heat engine (a car's engine is a prime example) requires an infinitely hot heat source or a heat sink at absolute zero. 2nd Law of Thermodynamics is a really big female dog...

    A easy to read article about Shannon Limits and Submarine Fiber Optics that carry data (including the internet) across the ocean.
    https://www.ciena.com/insights/articles/Shannons-Limit-or-Opportunity.html
     
  6. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    To be explicit, the practice of starting on one topic (sampling signals at higher a frequency greater than Nyquist causes aliasing - no duh, that's what LPFs are for), and continually forking new topics such as FTN, FTL travel, or time travel research, will result in a permaban. This "thread cancer" will not be tolerated going forward.
     

Share This Page