[Chu]

Quote:

Originally Posted by LawnGnome /img/forum/go_quote.gif This is incorrect. If it doesn't receive a signal during the time period, it is evaluated as a 0 or off. So a signal with severe timing issues would be more like this: 101010 ideal signal 1 0 1 0 1 0 Your idea of jitter 10001000100 More proper representation. The periods where signal aren't received are a an 0. Because the receiver is either getting a light pulse = 1 = ON, or no light pulse = 0 = OFF. There is either 0 or 1, no grey, no in between, no blanks.

This is not quite true, for pretty much the exact reason you list

Most digital encoding schemes are set up so that the voltage (or in this case, stream of light) is forced to swing every once in a while no matter what data is being sent.

There is a lot of literature out there on the best ways to do this, but they took the easy way out for digital audio over toslink. 01=1, 10=0. 00 or 11 is an error. It might be the other way around

Back to the op. This means, at a bare minimum, 50% of the bits you are sending over that cable are dedicated to finding flaws in the signal. There are more bits being sent that are dedicated to evaluating the link (and thus the cable) then there are to actually play back your music. As long as your cable is good enough s.t. the link is good, your cable is fine.

 

[d-cee]

don't hurt me! i'm just borrowing someone else's illustration

i don't claim to be able to hear jitter (as i've never been able to remove/add it in my system)

but i won't rule it out completely either, stranger things affect SQ in this hobby of ours =)

 

[dvw]

Quote:

Originally Posted by Chu /img/forum/go_quote.gif This is not quite true, for pretty much the exact reason you list Most digital encoding schemes are set up so that the voltage (or in this case, stream of light) is forced to swing every once in a while no matter what data is being sent. There is a lot of literature out there on the best ways to do this, but they took the easy way out for digital audio over toslink. 01=1, 10=0. 00 or 11 is an error. It might be the other way around Back to the op. This means, at a bare minimum, 50% of the bits you are sending over that cable are dedicated to finding flaws in the signal. There are more bits being sent that are dedicated to evaluating the link (and thus the cable) then there are to actually play back your music. As long as your cable is good enough s.t. the link is good, your cable is fine.

The coding scheme is called Manchester bi-phase. The idea is actually to have a transition every bit. 00 and 11 have no transition, therefore they will never happen. This is used to make clock recovery easier. This is also used in 10 Mbit interface. So there is always two symbols in every bit and is not bandwidth efficient but it is adequate for audio application.

For error correction they used an interleaving method. I don't remember from the top of my head. But any way this system is pretty robust from an error perspective.

 

[unkle11]

hi thanks for all the replies, i assume the above arguments about toslink applies to digital coaxial also?

 

[Chu]

Quote:

Originally Posted by unkle11 /img/forum/go_quote.gif hi thanks for all the replies, i assume the above arguments about toslink applies to digital coaxial also?

Some arguments do, some don't. Digital Coax is subject to all the same problems that normal cables are when it comes to interference, transmission loss, etc.

However, they are still transmitting a digital signal, so the all-or-nothing arguments still apply.

The protocols used are incredibly robust. To give you an idea of the quality of cable needed to ensure a good link, when TI were designing their chipsets for encoding/decoding digital audio, in their labs they were able to recover a bit-perfect signal from a six foot run of coat hangers.

 

[unkle11]

that.. boggles.. my mind.. i have no idea what to do now.. especially since the only coat hangers i have are plastic... should i just head out to radioshack? lol

 

[rb67]

Haha it'd be a much more pain in the butt to try to hook up some stiff wire without terminations!

Follow the facts. Just get a well made cable that won't break with quality connections that will last. If a claim sounds like rubbish, it most likely is.

 

[bdshort]

Just for reference, I find that the metal coathangers from Wal-Mart give me a much wider soundstage and tighter bass than hangers from Target or K-Mart.

 

[rb67]

Hm... I've always found the ones from target to be more accurate.

 

[peelax]

I agree, the ones from target give much more transparency to the music, with a deeper bass. Possibly because the mid range is slightly rolled off? Although I heard someone wrapped ESR paper around the Wal-Mart and the sound stage was greatly improved!

 

[Chri5peed]

I don't really believe, but what the hey!

My Optical cable is my cheapest interconnect. I'd rather not use an adaptor, is there anywhere that does a decent Toslink to Mini lead, under $50.

Its just that I wouldn't want it sticking out or is it like 1/8 to 1/4 adaptors, i.e. not very obtrusive?

 

[Dept_of_Alchemy]

Quote:

Originally Posted by peelax /img/forum/go_quote.gif I agree, the ones from target give much more transparency to the music, with a deeper bass. Possibly because the mid range is slightly rolled off? Although I heard someone wrapped ESR paper around the Wal-Mart and the sound stage was greatly improved!

I have heard that too, it must be true. Although it's been reported that the k-mart ones scale much better with brilliant pebbles than the other brands.

 

[evilking]

I'm glad people are starting to see sense.

 

[Chri5peed]

Glass does let bigger '1's through.

 

[peelax]

It is important to make sure your optical is plugged in the right way round too, contrary to popular belief glass does have a direction, preferring to let photons flow one way more than another. As manufacturers generally don't label the correct direction of photon flow it is important to conduct listening tests to determine it. If you have the cable plugged in the wrong way it will sound more fatiguing with more digital harshness, if it is right you get a more analogue sound with a blacker background. This is due to the way in which micro fractures in the glass reflect photons causing photonic standing waves in the cable. These micro fractures are directional, as they tend to form at the same sort of angle. The resulting standing waves interfere with the digital signal making the rising and falling of the ones and zeros more difficult to detect, in effect adding jitter and timing issues to the signal.

If you cannot detect which direction you should be plugging in your optical cable I can test it for you with my equipment for a nominal fee. I will also cryo treat it to align the silicon molecules properly, something that (amazingly) most cable manufacturers do not do.