DIGITAL OUT: Optical Cable vs. Coaxial Cable

The main reason Toslink is crap has nothing (or VERY LITTLE) to do with the cable quality. It has to do with the cheap transceivers (the device you plug the cable into at each end) used in Toslink systems. It's convenient and sometimes necessary when it's important to electrically isolate equipment from each other, but I agree with rcrerar that a good 75ohm will outperform any Toslink, no matter what it's made of.

Glass fiberoptic cable (AT&T - ST) is another story altogether, but it's found only on the most expensive consumer digital equipment (like Wadia, for instance.)

i would have agreed with nsgarch two weeks ago but when i changed optical cables it was night and day. the only explanation could be the cable itself. there are allot of discussions on this forum and others about whether quality of either digital cables matters and i never knew what to believe. now i have proof and am quite frankly surprised at the outcome.

Richard, I agree there are differences in Toslink cable construction; but I wonder if you compared the MIT Toslink cable with an equally "better" metal one?

My earlier point was that Toslink performance can be pushed only so far until it runs into the limiting factor of the transceiver units. A friend of mine who used to be with Wadia explained that the most expensive Toslink transceivers cost the manufacturers maybe $5 - $15 while the cheapest AT&T transceivers start around $100!

Of course I suppose there can be variations in 75ohm interface quality too. For instance, my friend Larry thinks coax is best, but ONLY if it employs BNC (bayonet) connectors (as opposed to RCA connectors.)

Of course I suppose there can be variations in 75ohm interface quality too. For instance, my friend Larry thinks coax is best, but ONLY if it employs BNC (bayonet) connectors (as opposed to RCA connectors.)

That would be highly system dependent, as it would be dependent on the jitter reduction capabilities of the dac (if any), the risetime and falltime of the output signal from the transport or other digital source, the degree of impedance mismatch between cable, connectors, and interface circuits at both ends, and the length of the cable (which affects the timing of reflections and re-reflections caused by impedance mismatch; 1.5 meters is generally considered optimal, shorter than that non-optimal).

See this paper: http://www.positive-feedback.com/Issue14/spdif.htm

And this and other threads here at Audiogon: http://forum.audiogon.com/cgi-bin/fr.pl?ddgtl&1237611683

Given these kinds of system dependencies, and others which most likely exist in the case of optical connections, I think it stands to reason that experiences and opinions will differ on the op's question.

Regards,
-- Al

When I referred to "variations in the 75ohm interface" I meant variations in the ability of different types of connectors to maintain the integrity of the RF signal across the interface; and not the quality of construction. Sorry for the misunderstanding.