Video Asylum

Posted by blue_z (A) on April 11, 2011 at 13:43:05

In Reply to: RE: SATA II cable upgrade in HD/BD players posted by Bom on April 09, 2011 at 06:06:05

Hi there

> Are you sure that there is no difrference for digital video signal?

If you carefully reread my post, I qualified my response by writing "packetized digital data". That is because not all digital interfaces are equally reliable.

At least three layers of a digital interface need to be evaluated.

1. The physical layer.
This is the cable (and connectors) that you can actually see and touch. This cable could be coaxial with just a single conductor and shield (as used by S/PDIF digital audio), or a multiconductor (as used by SATA) for several signals.

2. The electrical layer.
This is where it starts to get technical and less comprehensible since you cannot see these concepts. S/PDIF digital audio uses single-ended signaling (same as unbalanced analog audio using RCA connectors). SATA is a full duplex (two directions) communications link and uses differential signaling to reject noise (same a balanced analog audio using XLR connectors).
Note that digital demodulation techniques are tolerant of noise and signal degradation far better than analog demodulation. Analog signals are degraded by noise and attenuation when the waveform is changed because the information is carried by the waveform itself. But a digital signal can be tolerant of changes to the waveform since demodulation (translating "symbols" to bits) relies on ranges of amplitude or phase (rather than precise values or conditions). It's ironic that "digital" whatever is perceived/marketed as precise and analog is inaccurate, but it is actually the other way 'round when it comes to demodulation.

3. The data layer.
This layer organizes the data, and can provide identification and data integrity checks. S/PDIF digital audio uses a rather simple protocol, and streams the data. SATA uses SCSI command and response packets to transfer data, obtain device status and perform device operations. Data transfers can be validated, and if data corruption is detected then the data packet can be retransmitted or the entire operation can be retried if necessary. The data protocol adds reliability to the underlying transport layers (in the same manner that TCP is reliable but IP and Ethernet are not).


The SATA interface is far more sophisticated than the S/PDIF digital audio interface. You're comparing a *reliable* data connection used as a computer peripheral (where data corruption cannot be tolerated) to a data connection that does not have serious consequences if data corruption goes undetected.

> If there is no difference for digital video signal like you said, why do different HDMI cables have different picture?

I don't know much about HDMI other than it carries an uncompressed digital video signal and that is has some bi-directional capability (mostly for copy protection schemes rather than data integrity). But I'm quite sure that there is no agreement that "different HDMI cables have different picture" in the technical community.

I responded to your original question about SATA cables. I do not want to get into a S/PDIF and jitter or HDMI cable discussion.

Regards