Home Computer Audio Asylum

Music servers and other computer based digital audio technologies.

RE: wasn't referring to the connection

One can separate the physical layer from the protocol. In the case of I2S this is a matter of drivers and receivers. I wouldn't say that the engineering investment in developing a longer run I2S would be much of a problem. The problem is lack of standardization and hence a limited marketplace over which to amortize fixed costs. The process of standardization would be politically difficult and hence costly. IMO standardization is probably unjustified by the potential benefit given the limited size of the audiophile marketplace.

I2S is a better protocol from an audiophile perspective than SPDIF because it does not multiplex clock and data over the same wires. This means that bandwidth limitations on the cable (or required at the driver to meet government EMC standards) do not add signal dependent jitter. However, with long cable runs bandwidth will be less than a direct point to point connection on a circuit board and this means that the rise time of a clock signal will be greater. With the slower rise time comes a lower slope on the voltage/time curve and that implies that noise on the signal (and ground if the signal isn't differential) translates into more jitter in the received clock signal. Therefore there will still be a potential source of jitter associated with the cable, but the situation will be potentially better with I2S than with SPDIF because one of the two possible types of jitter has been eliminated.

If I2S is run with the clock at the DAC then jitter effects associated with the cable and its drivers and receivers can be eliminated but at the expense of complexity. When the clocking comes at the source the relative phase of the clock and data at the receiver are not significantly changed as a function of cable length but this is not the case when the clock is sent back to the source. There will be a fixed delay between the phase of the clock and the data and this delay will vary with the length of the cable and this must be taken into account in the timing budget. (This type of system can be very simple if the cable length is fixed. This is the solution Seymour Cray used for the I/O system on the CDC 6600 computer sold in the 1960's.) There is an additional problem with running I2S clocked at the DAC. Unless an audiophile wants to play audio files recorded at only one sample rate he will have to suffer the distortion of a sample rate conversion or the inconvenience of manually switching the sample rate at the DAC unless additional signalling is provided that allows the source to tell the DAC what rate it should run at. One of the benefits of using an async USB interface is that the USB protocol allows the computer to send this control information, as well as other things such as volume control. (Some USB DACs may have an analog volume control built in.)

There is no theoretical reason why running I2S over a 10 meter cable clocked at the source couldn't be as good as with a local clock at the DAC, but it would require that the noise induced jitter provided by the clock transmission on the cable be kept well below the phase noise of the oscillator itself. Which approach will give better performance in any particular case will depend on the quality of design and execution. (This is the generic: YMMV.)



Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar


This post is made possible by the generous support of people like you and our sponsors:
  Analog Engineering Associates  


Follow Ups Full Thread
Follow Ups
  • RE: wasn't referring to the connection - Tony Lauck 07:57:07 04/20/12 (2)
    • info - Joe Murphy Jr 13:16:41 04/20/12 (1)

FAQ

Post a Message!

Forgot Password?
Moniker (Username):
Password (Optional):
  Remember my Moniker & Password  (What's this?)    Eat Me
E-Mail (Optional):
Subject:
Message:   (Posts are subject to Content Rules)
Optional Link URL:
Optional Link Title:
Optional Image URL:
Upload Image:
E-mail Replies:  Automagically notify you when someone responds.