[tot]

Quote:

Originally Posted by OverlordXenu /img/forum/go_quote.gif How does a cable change the way 10101010101 sounds? Does the $100 cable add more zeros? Does the $20 cable take away ones? Because I don't see how anything outside altering the bit stream is going to make digital cables affect the sound.

You should read a bit how S/PDIF works, and this has been covered many times.

But in short, it is not just bits in digital signal. The DAC recovers the clock using leading edges of the bits. However, the edges are not exactly correct and have timing errors. The errors are caused by the senders clock and output circuitry, the cable and the DAC's input circuitry.

 

[stevenkelby]

Tot, as I understand it there is no such thing as a digital signal through a S/PDIF cable, they are analogue signals treated as digital. The cable is actually carrying a wave, there is a range of values equaling a 1 and a range for 0. If the wave strays outside that range, buts get lost. Timing is an also an issue as you say, which is corrected within a certain range, if it strays outside that range, it's lost. Does that sound correct?

 

[tot]

No, it is a square wave which changes sides (+max or -max) for every bit. If the bit to send is one, the second half of the wave goes to zero.

 

[stevenkelby]

Thanks. Is the value of the square wave always of the same size for a 1 or a 0, or does it vary within a range?

What is the wave made of anyway? Voltage? Frequency?

 

[tot]

The wave is voltage in the cable. I don't recall what the limit is, but basically you have three possible values: +max, 0, -max.

The frequency of the signal in the cable for CD is 2 * 32 * 44100 * 2 =~5.6Mhz.

 

[slwiser]

I want digital cables that are just adequate to feed my Lavry DA10 a good signal that it can resolve and control. I think any more is not necessary. But I am open to be educated.

I surprisingly found that better cables do make a difference in certain locations. Specially cables that carry the ground signal can be different from each other. There are secondary effects of signal design that at the resolutions of a lot of equipment around here can be noted but just not understood very well. This may also be true on the digital side but I have not come to even know of those effects yet.

 

[jaduffy007]

Miles....try to use BNC connectors on coaxial cables if possible as they are true 75ohm connectors. RCA are not.

Just for the record...digital ICs *can* make a very significant difference in sound.

DH Labs is a great, affordable place to begin *your experiments* to see if you hear a difference.

Good luck!

 

[ruZZ.il]

There's also a problem that arises from a wave transferred through medium changes (ie, impurities in a fiber optic cable, resistance mismatch in current carried signals, the borders between different types of glass/plastic, etc..) so like when a string with a wave traveling down it that hits a point where the string is tied to a different medium the wave will be partially reflected and partially transfered (depending on the differences in medium) as is with any digital signal through a cable, or light or whatever. In a cable, this reflected wave would hit the other end of the cable, and create another reflected/transferred wave, and so forth.. from any given wave, multiple reflections sum up within the cable always being partially transfered to the receiving end, that is to determine whether a signal is either a 1 or a zero (for instance above half a volt or below half volt).. if this reflected signal sums up with enough other reflected signals it could potentially cause a fake positive (ie, above half a volt). There are other phenomena, like dispersion. This is an effect on how the individual sinusoidal signals that are put together to create a square wave travel in the first place. the higher frequencies travel faster, flattening a wave. this could put the timing off on long lengths of cable, or even low quality cables, depending on the frequencies used. These are only some of the digital distortions a signal may be submitted to before reaching a receiver, and the quality of the receiver determines how well it copes with these things. How well it can differentiates between an intended or unintended step and the time it determines is between them. One must remember that light behaves in a wave like manner too, so these effects occur even in optical devices.. bottom line is that this is an analog world, with mostly sinusoidal signals, and digital science is made up from an analog nature, and just because one doesn't know of something nor understand it, that it doesn't mean it isn't so. I also understand that my understanding IS limited, and often inaccurate and that I often simplify things to the verge of being incorrect, so this is all with a bucket of salt. but something to think about

made you read haha...

Wave Reflection at an Impedance Discontinuity

 

[mofonyx]

I use Supra X-ZAC

 

[riffer]

Blue Jeans Cables using the Belden 1694A cable with Canare RCA or BNC connectors.

 

[Omom]

Quote:

Originally Posted by riffer /img/forum/go_quote.gif Blue Jeans Cables using the Belden 1694A cable with Canare RCA or BNC connectors.

2x...

 

[OverlordXenu]

So, is that $100 cable going to work any better than that $20 BJC?

 

[dvw]

Quote:

Originally Posted by tot /img/forum/go_quote.gif But in short, it is not just bits in digital signal. The DAC recovers the clock using leading edges of the bits. However, the edges are not exactly correct and have timing errors. The errors are caused by the senders clock and output circuitry, the cable and the DAC's input circuitry.

This is partially correct but not inaccurate. Bits are still bits unless they changed. DAC do recover clocks but today's DAC also retimes.

The bit are actually represented by two symbols. The advantage of biphase coding is clock recovery is easier and simpler. The tradeoff is higher bandwidth.

For more detail and precise description.

http://www.epanorama.net/documents/audio/spdif.html

 

[tot]

Quote:

Originally Posted by dvw /img/forum/go_quote.gif Bits are still bits unless they changed. DAC do recover clocks but today's DAC also retimes.

I am not trying to argue that jitter should matter, a proper DAC should reclock and be immune to incoming jitter. Historically this was not true and I am not convinced that all today's DAC are "proper" ones.

Bits should go through just fine with a proper cable on short runs, that should not be a problem.

 

[zer061zer0]

Not sure if i can seek some help here, I have a laptop that uses one of the heaphone out, a 3.5mmsocket as the spdif out. I tried using a coax cable with a adaptor to a mono3.5mm plug to my LITE-AH DAC it dun work. tried a normal 3.5-3.5mm cable with a 3.5mm-to-RCA jack adaptor, it don't work again.

Can I know if this is correct and if there is any settings to take note of? I am using a HP tx1000 series laptop, running on vista, i change to the digital out setting but my headamp doesn't give any output at all.

Did i use the wrong cable? or was my connection/settings wrong?