|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
98.248.201.11
In Reply to: RE: Auraliti & Berkeley Audio Alpha-1 DAC posted by Tony Lauck on October 12, 2009 at 09:06:22
The initial version won't support external clock. I'm looking at word clock and super clock support but those are not a good solution in practice. The core issue is controlling the clock to match the sample rate. The source knows the sample rate. If the clock originates at the DAC then the DAC somehow needs to know what sample rate to switch to. There is no standardized way to do this that I am aware of. The people I know who have tried this find its an interesting demo but don't use it for normal listening.
Follow Ups:
I agree with you. This external clock to dac business isn't flexible and is just as affected by impedance matching and cabling issues. Yes, the clock cable can make a big audible difference.Nobody seems to want to do it but it is much better having a good external clock with relocker. This way, you can actually feed a clock to the sound card, clean up the aes clock, discard the aes clock and generate your own, and compare the sonic results directly. (more or less)
In my experience, it is very important to have low jitter aes/spdif signal (into dac) which is symmetrical and has no impedance mismatch kinks (reflections)or high frequency oscillations. When fed to a good dac, this gives the best result regardless of clocking scheme. (provided that an excellent cable is used also).
It is not good enough to have a PC sound card aes output (most if not all are terrible irrespective of claims.)
Edits: 10/13/09
This is a classic trade off between quality and ease of use. :-(
I observe the following:
1. A user who bothers to set the DAC manually to the correct speed should get the best possible sound, i.e. bit perfect to the DAC and no transport jitter, only DAC master clock jitter.
2. A user who doesn't bother to set the DAC manually should still get very good sound, regardless of setting. This can be done by running the DAC at whatever speed it is set to and having the playback software make the necessary SRC. (The software will have all of the necessary information available to it.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
May be, but this is a theoretical statement not necessarilly reflected in practice.
Add
A user who sets his/her fs manually also risks having crashes and bangs from the LS.
What exactly is going to crash and bang?
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
It depends on the dac/clock combination. Some dacs don't like abrupt fs changes. Also dangerous in direct coupled playback systems (for the LS)
Are you saying that some DACs output a glitch when the SR changes, even if the digital input is silent at this point? Which DACs? What is the amplitude of this glitch (e.g. compared to a 0 dBFS digital signal)? Is it greater? If so, then such a DAC is dangerous, IMO.
I can see damage to speakers if the analog gain is too high. I wouldn't think this would happen if the analog gain has been adjusted so that 0 dBFS is within the limits of the amplifier and speaker. But perhaps not with real high power stuff. I can see blowing drivers if there is sustained DC or high power ultrasonic signals, both of which can persist for some time with no audible sign, but for a short transient it seems to me that something is seriously wrong with a system if a brief glitch can do permanent damage. I know there are people who like to run this way, drive cars at 180 MPH on the highway, etc., and I've blown my share of LS drivers over the years, melted KT88 and EL34 plates, etc...
Apart from the more youthful enthusiasts, I should think most of us value a system that operates consistently and dependably more than one that puts out the best possible sound on a few occasions in between repairs. I prefer my audio system to be more like a Mercedes than a Ferrari, or a Lotus Elan (the latter causing me to have to take the bus to work for an entire winter in my more youthful days after I blew the engine on a track).
But yes, a warning is definitely in order. (Worse would be if headphones are used, as damage to hearing is more serious than to equipment.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
As I said, some dacs are prone.
No, I don't go round measuring amplitudes of thuds; I usually mute my system before making changes. This doesn't mean that I don't (nor others) occasiobnally do something untowards.
I don't go around setting 0dB levels either; this kind of thing I used to do years ago. Now, I make sure L&R are matched; phase is correct; digital level not over etc etc.
And some systems merrily play the content at the wrong speed when driven from an external clock. Different DAC's handle the sample rate changes differently. Most of the pro market devices I have been near do it gracefully. Some consumer stuff generates ticks and pops until it realizes that the sample rate changed.
One intrepid high end dealer I know fried several sets of really expensive tweeters trying the external clock stuff.
I really don't want to prescribe or dictate the best way to do any of this. However I also only want to support stuff I know will work on the platform. As we figure out how to support alternative ways to use the Auraliti platform we will be happy to share as much as possible. We recognize that this is a rapidly evolving environment.
And dCS and Lynx won't work together, each quoting specs at the user. I think it's mostly Lynx's problem.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: