Glass vs. Plastic optical cables

[Celeron_Phreak]

I was browsing Parts Express and I found some Glass Optical cables:

http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=180-954

I was wondering how much better performance (if any) one would get form these vs. plastic ones....?

[lowfat]

I doubt there will be any difference in sound. All optical cables should sound the same.

[Incesticide]

There won't be. Data transmitted though an optical cable will be digital, so as long as the cable is up to spec, there should be no difference. Optical versus coaxial with digital is another issue though, due to grounding and what-not, coax can introduce a hum. But I think the only advantage that glass has over plastic is range. But this is only an issue when you are discussing kilometers of cable.

[DaWiper]

You probably wont hear much difference. I myself got a "glass" cable, just to be sure :D

[Celeron_Phreak]

DaWiper, how fragile are they? When you coil it, how big is the diameter of each coil?

[RobxMcCarthy]

I wasn't aware that they made plastic optical cables. Most optical cables are indeed glass. The way that fiber optics works is that there's a high refractive index core and a lower refractive index "cladding". Light entering the fiber (multi mode or single mode) is constrained to the core. I haven't heard of this being done with plastic though, I'd assume the refractive indexes would be too low.

Either way I don't think you'll lose much data over a few feet. Optical fibers are meant to transmit data over hundreds of kilometers without loss. (unlike standard copper signal transmission).

And although you won't get a grould loop issue with an optical cable, depending on the clock of your source and your receiving photo-diode's timer you can introduce what's called "jitter" It happens when the clock frequencies don't agree with eachother and it makes the music sound harsh and well...jittery.

Also, glass fibers are very sturdy. They have a higher tensile strength than Coax and can bend to some rather tight radi. Although bending the fiber too tightly can result in loss of signal as the light can no longer remain constrained to the core around really tight bends. But then again, what are you planning on doing with this cable, mountain climbing?

[Celeron_Phreak]

lol, not quite McCarthy. :)

I decided to just order the regular Dayton 6' optical cable from parts express. If I have any spare cash in the future, I'll order a glass cable, just for comparison though.

[RobxMcCarthy]

Not quite what?

[Celeron_Phreak]

It won't be getting so much abuse that it's application could be compared to mountain climbing. :p

[JCLW]

All digital sound cables will sound identical. Glass, plastic, or coax. Either you get the 1s & 0s out the other end in a recognizable stream or you don't.

I usually use coax because it's cheaper - plus most people have a few 75ohm coax cables with RCA ends floating around from their video card purchases over the years.

[Emon]

I'd assume the refractive indexes would be too low.
Maybe too high? I was under the impression that plastics used in modern glasses have higher indices of refraction than glass, which is why the lenses are so much thinner.

[Celeron_Phreak]

I just recieved my Dayton 6' optical cable in the mail this morning. After comparing it to the one that came with the X-Mystique, I can personally say that I do hear a difference. Using the optical output now has that "warm" sound to it and the bass is punchy again. :)

[Randyman...]

The Human Brain hard at work! Pre-conceptions essentially ruin any chance of a fair A/B test in most "perception" type scenarios. That is what we get for being "Human".

FWIW - The Zeros and Ones in the SPDIF stream have NO CORRELATION to "Warm" or "Tighter Bass" - they have no correlation to audio AT ALL. You can not look at a SPDIF stream and say "that stream is a low frequency - I can tell" - the 1's and 0's are not "Analogous" to analog audio (that's why it is called digital, and not analog :p ). 1's and 0's do not make audio - the DAC and its reconstruction + filtering DO. The only issue that can be direclty applied to the resultant audio is jitter (still not a direct relation to "Analog" audio), and dropout. Lose the DATA, and you loose the associated Audio. Errors do affect sound, but I doubt swapping from a working 6' optical to a 6' glass fiber cable is reducing errors (unless the old cable was flawed). Just sayin'...

Humans are one-of-a-kind! What can we do? Just enjoy the multiple imagined realities we as humans create. I don't believe ANY more A/B tests unless they are true double blinds (don't see many of those in the Audiophile world, or those $1000 AC power cords would be history by now ;) ).

Good converters can make a world of difference, as this is where the "work" is done, but as long as the DAC is getting a fairly jitter-free clock, and an error-free data stream, digital interconnects will not affect "sound" (its DATA at this stage, not "Audio").

:cool:

[Emon]

Ahahaha yes, audiophile power cables. Biggest load of **** ever. Can't those people realize that the power goes through miles and miles of copper, sometimes aluminum cable, from the power plant to their house? And a 6' cable will make a difference...how? "But the bass is tighter, the soundstage is wider!" oh and one of my favorites, "the highs are less sibiliant!" god I hate audiophiles sometimes.

[bwanaaa]

The Human Brain hard at work! Pre-conceptions essentially ruin any chance of a fair A/B test in most "perception" type scenarios. That is what we get for being "Human".

FWIW - The Zeros and Ones in the SPDIF stream have NO CORRELATION to "Warm" or "Tighter Bass" - they have no correlation to audio AT ALL. You can not look at a SPDIF stream and say "that stream is a low frequency - I can tell" - the 1's and 0's are not "Analogous" to analog audio (that's why it is called digital, and not analog :p ). 1's and 0's do not make audio - the DAC and its reconstruction + filtering DO. The only issue that can be direclty applied to the resultant audio is jitter (still not a direct relation to "Analog" audio), and dropout. Lose the DATA, and you loose the associated Audio. Errors do affect sound, but I doubt swapping from a working 6' optical to a 6' glass fiber cable is reducing errors (unless the old cable was flawed). Just sayin'...

Humans are one-of-a-kind! What can we do? Just enjoy the multiple imagined realities we as humans create. I don't believe ANY more A/B tests unless they are true double blinds (don't see many of those in the Audiophile world, or those $1000 AC power cords would be history by now ;) ).

Good converters can make a world of difference, as this is where the "work" is done, but as long as the DAC is getting a fairly jitter-free clock, and an error-free data stream, digital interconnects will not affect "sound" (its DATA at this stage, not "Audio").

:cool:


I thought there was something called 'jitter' that affected the timing of the pulses. If the 1s and 0s do not come at the right times, a 1 might be used instead of a 0 and vice versa.

[Randyman...]

Nope. Jitter is indeed related to timing imperfections, but jitter will not be high enough to cause errors in the reciever (if you have enough jitter to actually cause errors, there are serious problems with something in your system - probably a bad cable or a bad optical transmitter/reciever). Jitter WILL, however, cause timing inaccuracies in the audio, which can smear the soundstage.

The MOST IMPORTANT place to have low jitter is in the studio, at the initial A/D conversion during the recording session (and any analog to digital transfers - like from 2" 24 Track reel-to-reel into 24 Channels of ADC to get into a DAW for editing and mixing). If this initial A/D "capture" was created with a jittery clock, no amount of "jitter correction" can un-do the damage. The damage is now part of the digital snapshots (the samples were not "sliced" equally - some "slices" may be "Thicker" than others, and the minute timing BETWEEN samples drifts).

Interestingly, jitter also does NOT AFFECT digital-to-digital transfers (like from an ADC to your Soundcard's SPDIF input). As long as the samples are sent and recieved one-by-one w/o dropouts, a digital-to-digital transfer is immune to jitter (just like transferring data from a HD to the Southbridge). Digital-to-Digital transfers and DSP procesing can ADD jitter, but it does not have any effects to the "audio" (still a data stream) at this time. Jitter only comes into play during CONVERSION to a real-time analog audio signal. If the jitter is causing timing imperfections, the resultant audio will also be plagued with these timing errors ("smeared" if you will).

Another point of interest, a GOOD DAC can actually resolve most jitter (assuming an initially "clean" ADC capture), as the DA will have a small sample buffer, and then it will usually "Re-Clock" the bitstream to a PLL type loop (Phase-Locked-Loop), and any jitter can be eliminated BEFORE the digital stream is converted to a "Realtime Analog" signal. This is NOT true at the initial AD conversion. A bad AD capture is not un-doable.

Something like that :cool:

[Celeron_Phreak]

So then, what effect does a precision machined plug on an optical yield?

[Randyman...]

So then, what effect does a precision machined plug on an optical yield? A joke, or are you serious? :shrug: Gold plating will certainly help an optical connection, so why not a metal jack, too ;) (Joking).

Considering the Toslink "socket" will still be molded plastic on the unit itself (I haven't seen metal toslink jacks on any unit - even AD converters costing $4000 use plastic Toslink jacks, but metal jacks might exist), and a slight "misfit" will only slightly lower the brightness/intensity of the optical signal (it shouldn't change the digital data, just make the optical reciever see a slightly weaker light - but still well within tolerance), I'd have to say no difference what so ever... The Optical portion of the Cable is still what "Makes the connection", as the optical cable still protrudes from the plug.

They do look cool, and they might also improve the ruggedness of the cable, but nothing to improve the light transfer (Jitter/Errors). I just bought 4 used 3' Optical/ADAT cables for $10 on eBay - with the metal ends. They work fine - just as fine as the thin plastic terminated ones I already had for my RME Digiface and RME ADI-8 ADAT AD/DA converter on PC#2 in my sig (My dedicated DAW that runs Nuendo).

:cool:

[Celeron_Phreak]

I'm not talking about a metal machined connector, I'm talking about a plastic one. Look at the optical cables on partsexpress.com and see for yourself.

[Randyman...]

Are you referring to the "Polished Tip"? Or the "gold plated plug tip" (looking at your inital link)? Polishing can help increase signal strength (I guess - but this goes back to not actually "Changing" the data - just making the "light" brighter which might help with crappy equipment with poor transmitters/recievers), but the "gold plated plug tip" is a load of hogwash IMO.

Please also notice that Parts Express page also says: "The cable contains 65 glass fibers to ensure the best frequency response..." . Since when does the digital CABLE dictate the frequency response of a digital transfer? It CAN NOT!!! It either passes the Digital Data stream effectively w/ few to no errors, or it would create TONS of errors at the reciever end if the cable could not handle standard SPDIF bandwidth. The Frequency Response is dictated souly by the sample rate, and the specific filtering in the AD/DA converters. A digitial interconnect is NOT transferring analog AUDIO!!! It is transferring 1's and 0's with NO CORRELATION to analog audio!!!

If you HD could not reliably send 1's and 0's to your Southbridge, you would have a corrupt OS before you knew it. Bit-Accurate transfers are easily attained inside a PC at bandwidths that dwarf ANY kind of Digital Audio transmissions - so why wouldn't a descent Toslink cable be good enough to transfer a measely 44.1K - 96K bandwidth audio @ 16-24bits? What about Fibre Channel? No data corruption there, and those bandwidths are through the roof!!! We are talking about DATA, not analog audio when referencing Digital Audio and transmissions of this digital audio. It is only "Analog" AT the ADC's input and DAC's output.

:cool:

[DaWiper]

What happens with the sound if one single bit gets screwed up? Lets say the audio signal is a perfect sine.
Does spdif has any kind of crc checking/error correction?

[Emon]

I don't think so, but it shouldn't need it. UDP is a connectionless protocol that has no kind of error correction and it works great for file transfers and internet/LAN gaming, among other things.

[Randyman...]

IIRC, the SPDIF protocol uses a form of error correction called "Cross Interleave Reed-Solomon Code" which has 25% redundancy.

Bottom line - if it sounds better to you - that's fine. Just be informed that the human brain is one VERY odd organ, and what may appear to be "A definite change" might be nothing at all (Literally, all in your head :) )... This is why "Double Blind Tests" will sometimes yeild very different results from what your pre-conceptions would subconciously suggest.

:cool:

[DaWiper]

Lol, In some way I think I'm almost willing to say that if it sounds better in your head that is all that matters. In the end that's where the sound is suppose to go. :D

Isn't "Cross Interleave Reed-Solomon Code" only used internally in compact disc players?

[Incesticide]

If glass cables with precision machined connectors make you happy, all the more power to you. But that money could be spent elsewhere, with better results, making you even happier ;)

[Avatar28]

For what it's worth, the toslink standard is actually expecting the normal plastic cables, not glass. So I really don't see how it could make a difference.

Now, that said, I HAVE heard a difference in quality between TOSLINK and a coax S/PDIF connection. Shortly after I got my current receiver, I had both hooked up to two different input selectors. When playing music one time I switched between them and I heard a slight but definitely noticable difference between the two. The coax actually had a slightly clearer sound to it. Where with optical I could hear a slight bit of distortion, when it switched to coax the distortion was less noticable. I couldn't honestly say the frequency response was any better or anything like that, just clarity (I would be hard pressed at this point to say if the sound stage was any better or not either).

I don't know why there was a difference. Certainly I had expected NOT to find one at all, which is why it so surprised me. If I had to guess, I'd say that either the receiver or the DVD player was introducing a slight bit of distortion in the conversion from electrical to optical or vice versa. Of the two I'm more inclined to believe the fault lay with the DVD player.

That was with a PCM audio stream. With encoded formats like dolby digital and dts it shouldn't be possible to have ANY difference between the two.

Edit: DaWiper, you are correct that that error correction is only used internally on the CD player not s/pdif. I have done a bit more research and found a couple of links that go into technical detail on the signals.
http://members.chello.nl/~m.heijligers/DAChtml/digcom/digcom.html

Also if you read the minidisk FAQ (http://www.minidisc.org/faq_sec_4.html) it also indicates there is no real error correction with S/PDIF, only parity. One quote that seem to relate to the discussion at hand:

SPDIF: (Sony/Philips Digital InterFace): This is the interconnect that is most often used on consumer DAT machines. The connectors are standard RCA phono connectors. This type of connector may also be lableled "IEC Type II" or simply "Digital I/O". Standard analog phono cables can usually be used for the digital data, however some cables that are designed for analog may not be able to carry the high rates needed for the digital data, especially over long distances. Many high-end audio stores carry special digital phono cables that solve this problem. [The pro-audio FAQ says not to use audio cables, but that video cables will work].

It also linked to this page (http://www.cl.cam.ac.uk/Research/SRG/HAN/docs/sp-dif.txt) which seems to contain some nice information on the s/pdif interface and was actually somewhat easier to read than the first link.