Optical vs Coax SBT>Rega DAC

[Golly640]

I went through the same dilemma with my Cyrus sytem that has a CD transport to a preamp Dac. After a week running the cables in there were no similarities. A good quality digital coax outperforms optical cables in terms of detail and clarity. The only benefit with an optical is if you need to have longer than 1 to 2 meters where you might just lose a bit of detail on the coax.

 

[Mustbedef]

Hi

I've just found this thread and thought I'd mention my experiences on the subject of the 'Rega Dac coax or Toslink?' debate. I've owned my Dac for about 6 months now and have been experiencing intermittent random clicking noises (about once or twice in a listening session) while the music was playing which has been quite annoying. I wondered whether it might be a fault with the unit or something like a ground loop problem between the player and dac, so I bought a Van Den Hul Optocoupler to see what happend. It appears to have cured the clicks, but the sound is also smoother and less grainy than with the coax. It sounds faster and more natural with at least as much detail as the coax. The coax is an 'XLO' digital cable which my Rega dealer sold me and could be the cause of the problems, but I didn't want to buy another coax and find I was in the same situation. My player is a Rega Jupiter 2000 so there shouldn't be a compatability issue there.

I do think my system is quite revealing and small differences are really noticable.

 

[john dolan]

USB out sounds better than coax or optical into my M-dac using the EDO.

 

[spockfish]

The amount of FUD spread about this is just unbelievable. We're talking about the digital domain, and there you only have 0's and 1's. So either the bit is transferred correctly or not, and in general the transport protocol takes care of some error correction.

So: either the data is being sent over or not, the latter presenting itself as 'drop out's. All the stories about "better detailed, smoother" etc are pure bogus and I dare anyone to do a blind test.

The *only* difference that exists is that optical cables tend to suffer from jitter more. And although in general the idea is that people can not hear 'jitter' the latest experiments reveal otherwise above a certain level (measured in time).

However, in your specific case you have a (very good) Rega DAC, which has a circuit that (tries to) eliminate jitter to the bare minimum, resulting in virtually no difference between COAX or optical.

 

[andyjm]

The *only* difference that exists is that optical cables tend to suffer from jitter more. And although in general the idea is that people can not hear 'jitter' the latest experiments reveal otherwise above a certain level (measured in time).

Hmmn. Not sure what you mean by 'latest experiments'.

Jitter has been known about since the earliest digital systems were designed. At extreme levels it causes bit errors, for digital audio much lower levels of jitter (while not causing bit errors) are still audible.

Work was conducted by Bell Labs in the early 60's to establish the audibility threshold of jitter when they introduced the T1 PCM system in the US. This was for 'telephone grade' audio, so not directly applicable to HiFi.

More recently, the BBC research department published a paper in 1974 about the audibility threshold of jitter - they identified that it dependend on the type of jitter, not just its magnitude. One conclusion was that jitter below 35nS RMS was inaudible to 95% of the population - this I guess is an upper bound of jitter acceptability.

http://downloads.bbc.co.uk/rd/pubs/reports/1974-11.pdf

Various tests (available on the web if you google) have established that jitter correlated to the programme material is more objectionable and detectable at lower levels than random jitter. S/PDIF has a tendency to introduce correlated jitter.

Levels of jitter below 250pS RMS seem to be undetectable in all the studies I have read. Most modern sources are able to achieve levels far better than this. Tough to find published specs, but to put this in context, the Squeezebox Transporter (which made a virtue of its low jitter levels) was able to achieve jitter below 15pS RMS.

 

[spockfish]

I meant exactly the same. With 'latest' I was referring to a recent study (I can't find the link back in my history :-(), which was basically comparable to the study from the BBC.

I was more referring to the statements made like 'smoother' and stuff like that which makes no sense what soever.

But as we have comparable systems: have you already had a look at the BlueSound streamer yet? http://www.bluesound.com/products/node

 

[Mustbedef]

Spockfish, jump down off your high horse and give people a bit of respect. I stand firmly by what I said about what I heard and in case you didn't read my post properly, I did say that I thought that there may be a problem either with my coax or my dac unit. In which case a reduction in jitter is perfectly feasible. When I said 'smoother' I meant less grain to the sound and this is still the case.

I have a simple experiment for you to try on your own system. Insert a 20k Alps pot in to a pair of interconnects and fit them between your source and your 'direct input' (if you have one) on your Rega Brio amp. Care is required when controlling the volume and I cannot be held responsible if you blow your gear up. As good as the Brio is, I guarantee that once you've heard the results of protecting the source signal from the murky preamp circuit you will no longer be happy with your amp and you'll want to upgrade. The extra detail and bigger soundstage will jump out at you. I used to own a Rega Cursa/Maia setup and this experiment started me off on the unattainable holy grail for perfect sound. But until your system is as capable as mine in terms of resolution, I'll pass on your advice and opinions.