VideoHelp Forum
+ Reply to Thread
Results 1 to 18 of 18
Thread
  1. Member SingSing's Avatar
    Join Date: Apr 2001
    Location: U.S.A.
    Search Comp PM
    I used up the only coax spdif connection on my surround amp.

    I was ready to buy a $2 $2 Toslink optical cable off ebay.

    But I read optical Toslink have bad start, due to using LED and low band width optical sensor and cable.

    So, did optical Toslink and cable improved enough to be as good and hiss free as coax connection ?

    Is there a field test result for toslink cable performance by brand or make ?
    Quote Quote  
  2. Member
    Join Date: Oct 2004
    Location: Freedonia
    Search Comp PM
    The problems with TOSlink are that its limited in size and bandwidth. It can't carry high def lossless audio like DTS-MA and it can't be longer than about 10 meters. Quality was never a problem with it.

    Maybe you just need a switch like this for more S/PDIF inputs
    http://www.monoprice.com/Product?seq=1&format=2&p_id=3027&CAWELAID=1329448176&CAGPSPN=...FQyNaQodViEAdQ

    I have no experience with that switch or Monoprice. That's just a suggestion.
    Quote Quote  
  3. Wow, I never realized that the coax connection was an over all better set up than the TOSlink ones. I could never really hear a difference so I used them interchangeably. Thanks for this!

    --dES
    "You can observe a lot by watching." - Yogi Bera
    http://www.areturningadultstudent.com
    Quote Quote  
  4. Member Cornucopia's Avatar
    Join Date: Oct 2001
    Location: Deep in the Heart of Texas
    Search Comp PM
    There is NO, repeat: NO difference between Optical vs Coaxial forms, except length & interchange solidity (which favors coax) and freedom from EMI/RFI (which favors optical). BOTH use the SPDIF protocol so both have the same restrictions it has.
    Quality is IDENTICAL.

    Scott
    "When will the rhetorical questions end?!" - George Carlin
    Quote Quote  
  5. Hmmm....thanks Scott!!!!

    --dES
    "You can observe a lot by watching." - Yogi Bera
    http://www.areturningadultstudent.com
    Quote Quote  
  6. Member SingSing's Avatar
    Join Date: Apr 2001
    Location: U.S.A.
    Search Comp PM
    Hi, Scott,

    It is not easy to working out a decent optical link in physical layer. Optical fiber are designed for laser and not LED, especially when cheap lens can cause collimation loss, cheap multi-mode fiber will loss power and bandwidth. Consumer also don't clean the tip of the fiber. These problems can loss up to 80% of light.

    A simple test is to shine the red light from the light port as well as the end of the fiber on a piece of white paper, and look at their intensity, and check the shape of the dot. You don't have to go down to the optical lab to tell the different.

    Power and bandwidth loss cause Jitter, and cause clock recovery problem, plus timing and data error and impact the quality of the outcome.

    When Toshiba first developed Toslink, it did not work well. It get acceptable after people graduated to using better optical components and fiber.
    Since they are no test data available, then One can run into problems when throwing a bunch of ebay stuff together.

    For coax cable in Audio gears always sit in enclosed metal box,es so emi/rfi hardly can causes issues. Also, humming cannot caused by a digital clocked interface.
    Quote Quote  
  7. Member
    Join Date: Mar 2011
    Location: Nova Scotia, Canada
    Search Comp PM
    Jitter is the THD of modern technobabble. There isn't anything there that won't show up in other distortion measurements. Scott is right. As usual.
    Quote Quote  
  8. Member SingSing's Avatar
    Join Date: Apr 2001
    Location: U.S.A.
    Search Comp PM
    Originally Posted by Hoser Rob View Post
    Jitter is the THD of modern technobabble. There isn't anything there that won't show up in other distortion measurements. Scott is right. As usual.
    The Jitter in optical signaling is about its transmit and recovered clock. It is not about not about audio, which is in a Different domain.
    Quote Quote  
  9. Jitter is real and can be easily measured and usually for TOSLink is higher than on even cheap coax, usually jitter on TOS is approx 5x bigger than on coax.

    Hum is not a problem as coax use galvanic insulation (thus any hum problems is more related to poor grounding - ground loops or simple device malfunction).

    Today jitter is less problem than in past as most of device use buffers and re-clocking buffer content.

    In past some companies instead TOSLink selected standards from network market - such as glass fiber optic, SC/ST pigtails etc.
    Quote Quote  
  10. Member Cornucopia's Avatar
    Join Date: Oct 2001
    Location: Deep in the Heart of Texas
    Search Comp PM
    Transmission jitter is inconsequential and irrelevant while within the bounds of standard digital reclocking & error correction (which it almost always is). Jitter is really only a quality factor when sampling (A->D), resampling, or playing out (D->A). Don't bring up red herrings.

    Toslink is perfectly fine for short distance, tame environment stuff.

    You want HD lossess, super-high samplerate, or multi-multi-channel? Shouldn't use either one, rather use HDMI or multi-cable AES or SDI.

    Scott
    "When will the rhetorical questions end?!" - George Carlin
    Quote Quote  
  11. Member SingSing's Avatar
    Join Date: Apr 2001
    Location: U.S.A.
    Search Comp PM
    Originally Posted by pandy View Post
    Jitter is real and can be easily measured and usually for TOSLink is higher than on even cheap coax, usually jitter on TOS is approx 5x bigger than on coax.
    Do note that your measurement equipment has their built in Jitter, caused by their front ends. Take me a while to figure out why I cannot measure TIE Jitter under 7 ps

    Originally Posted by pandy View Post
    In past some companies instead TOSLink selected standards from network market - such as glass fiber optic, SC/ST pigtails etc.
    You are saying Toslink is not using glass fiber ? what does Toslink use ? plastics ?
    Is SPDIF a modulated signal or a base-band signal on coax ?

    We should always have glass fiber optic, SC/ST pigtails etc. in our drawers, they come in handy.

    Originally Posted by Cornucopia View Post
    Transmission jitter is inconsequential and irrelevant while within the bounds of standard digital reclocking & error correction.
    Scott
    I don't think Spdif has error correction.
    I also don't think spdif transport can use reclocking, that require the receiver to train its local clock, and that add undesirable delay in audio.
    Sight, I like simple stuff like spdif, but sometime they have limitation.
    Last edited by SingSing; 23rd Jul 2014 at 11:51.
    Quote Quote  
  12. Member Cornucopia's Avatar
    Join Date: Oct 2001
    Location: Deep in the Heart of Texas
    Search Comp PM
    TOSLink uses either glass or high-quality ("optical grade") plastics in their fiber. Optical fiber uses laser light (aka COHERENT light) as its modus operandi. LED would not work at all, as it isn't coherent light, so it wouldn't make it very far down the fiber before reflections/diffractions/diffusion would overwhelm the signal with noise. You are mistaken.

    SPDIF has built-in error detection (correction would be built into the receiving device/algorithm). You are mistaken.

    The amount of delay incurred from reclocking is also inconsequential: the buffer would cover a range of a few samples (~84microseconds=4 samples @48kHz), whereas the delay in synchronized audio would only become noticeable at around 200-500 milliseconds. And many common devices (cell phones, PC drivers) routinely have a delay of 2-50 msec without any notice at all. Reclocking is done by the receiving device, not the transport itself (sometimes using internal crystal clock, sometimes - in pro venues - using master clock). You are mistaken.

    Scott
    "When will the rhetorical questions end?!" - George Carlin
    Quote Quote  
  13. TOS Link use plastic fiber optic (quite thick when compared to usual for networks glass fibers). Light source are red LED. Nothing fancy, yes, jitter is introduced at 2 conversion and transmission itself - this is why it's so high when compared to coax, also bandwidth for TOS is usually limited way bellow 20MHz (rarely more than 5 - 8Mbps speed are supported). Nowadays jitter is less problematic than in past as modern audio architecture usually assume to have buffer (FIFO) with own clock that reclock data and do jitter filtering.
    Quote Quote  
  14. Member Cornucopia's Avatar
    Join Date: Oct 2001
    Location: Deep in the Heart of Texas
    Search Comp PM
    (Partial) RETRACTION: Evidently, they DO use LED (incoherent) light for their transmission. My mistake. I will have to investigate how that can hold up through such distances...
    However, they DO come in both HQ plastic AND glass. I own a set (of glass). It IS thinner than the plastic, but is MUCH more fragile.
    Also, yes, jitter is higher than coax. Except, it doesn't really make a difference as long as it still falls within the boundaries of the reclocking buffer.

    Regardless, the limitations of TOSlink in particular, or the SPDIF protocol in general, are often still more than satisfactory for consumer electronics needs. If you were happy with a coax connection before, and your environment is tame & stable and your distances short, you will also be happy with the TOSlink. No better, no worse.

    Scott
    "When will the rhetorical questions end?!" - George Carlin
    Quote Quote  
  15. Member
    Join Date: May 2005
    Location: Netherlands
    Search Comp PM
    Is the new 4G mobile phone network of concern maybe? A - here in Holland - recommended brand coax cable has this new cable with extra shielding to avoid interference which would be caused by 4G. This cable is for indoor home use of cable-tv systems. I don't know what to think about this, but if it were to be true, I can imagine that 75 ohm coaxial interlinks can be affected too. I never experienced any trouble with coax so far. Maybe optical signal travel will suffer much less from radiowaves interference?

    As far as differences between coax and TOSlink, for the interlinks of both kind I've always used, I couldn't tell them apart.
    Quote Quote  
  16. Originally Posted by Cornucopia View Post
    However, they DO come in both HQ plastic AND glass. I own a set (of glass). It IS thinner than the plastic, but is MUCH more fragile.
    Strange as transmitters and receivers are designed to deal with thick plastic fiber (around 0.6 - 0.8mm) - thus glass fiber must use some magnifying solution to adapt fiber diameter to the diameter expected by transmitter/receiver.

    Originally Posted by Ennio View Post
    I can imagine that 75 ohm coaxial interlinks can be affected too. I never experienced any trouble with coax so far. Maybe optical signal travel will suffer much less from radiowaves interference?

    As far as differences between coax and TOSlink, for the interlinks of both kind I've always used, I couldn't tell them apart.
    Definitely fiber-optic is not sensitive to external EM field this is why so frequently fiber is used in heavy industry and for example power electric as it is able to deal with data/signal transmission.
    But LTE case seem to be not so big issue as LTE use frequency around 780 - 860MHz and to improve coax immunity usually ferrite bead's at both ends are sufficient.
    Quote Quote  
  17. remove please
    Last edited by pandy; 25th Jul 2014 at 11:05. Reason: post combined...
    Quote Quote  
  18. Member SingSing's Avatar
    Join Date: Apr 2001
    Location: U.S.A.
    Search Comp PM
    Originally Posted by Cornucopia View Post
    (Partial) RETRACTION: Evidently, they DO use LED (incoherent) light for their transmission. My mistake. Scott
    Two facts : 1. Toslink does use LEDs. 2. Even HeNe Laser is not totally coherence.

    Originally Posted by Cornucopia View Post
    I will have to investigate how that can hold up through such distances... Scott
    Fiber optic losses are mostly at their entry and exit points. Also, SPDIF is digital. SO, as long as the level at the receiver is higher than the detector threshold, the signal will get demodulated.
    Quote Quote